| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 27 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 46:21:41 |
| Calls: | 632 |
| Calls today: | 3 |
| Files: | 1,187 |
| D/L today: |
24 files (29,813K bytes) |
| Messages: | 176,482 |
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Cursitor Doom <cd@notformail.com> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
My Farnell power supply also does that. There are two switches, one for
the mains and one for the HT output. You just keep the output switched
off until the voltage has had time to settle down.
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to >>their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew
up transistor circuits.
You could hang a big MOV or TVS across your load circuit.
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
John Larkin--- Synchronet 3.21a-Linux NewsLink 1.2
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>say, they both overshoot after warming up and then settle back down to >>>their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
On Wed, 8 Oct 2025 22:06:38 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
My Farnell power supply also does that. There are two switches, one for >the mains and one for the HT output. You just keep the output switched
off until the voltage has had time to settle down.
Might be the same supply if it's the 350V one by the sound of it;
that's a Farnell as well. It has 6 ranges, course and fine controls,
mA & V moving coil meters and a bunch of fixed low voltage outputs for heaters and stuff. Same as yours?
I'm not mad keen on your solution, though. I tend to be rather absent
minded!
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>say, they both overshoot after warming up and then settle back down to >>>their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any,
but then the output tends to take quite a while to get up to it's stable >level.
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's >>>>an initial voltage surge way beyond their indicated output levels. One >>>>of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>say, they both overshoot after warming up and then settle back down to >>>>their expected output levels. The 350V one approaches 400V shortly >>>>after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>but I'd like to mod these so they max-out at their indicated voltages >>>>and no more. Something preferably simple, like a crowbar, perhaps, but >>>>a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>this question (well, I suppose it *could* be if someone chooses to >>>>>make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>an initial voltage surge way beyond their indicated output levels. One >>>>>of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>say, they both overshoot after warming up and then settle back down to >>>>>their expected output levels. The 350V one approaches 400V shortly >>>>>after switch-on and the heftier one surges to over 600V before >>>>>settling back. Now, that may have been acceptable enough 50 years ago, >>>>>but I'd like to mod these so they max-out at their indicated voltages >>>>>and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>I'm not about to do that with supplies that are in all other respects, >>>very good indeed and will probably still be around for many more years >>>than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all.
I might adopt that.
Cursitor Doom <cd@notformail.com> wrote:
On Wed, 8 Oct 2025 22:06:38 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
My Farnell power supply also does that. There are two switches, one for
the mains and one for the HT output. You just keep the output switched
off until the voltage has had time to settle down.
Might be the same supply if it's the 350V one by the sound of it;
that's a Farnell as well. It has 6 ranges, course and fine controls,
mA & V moving coil meters and a bunch of fixed low voltage outputs for
heaters and stuff. Same as yours?
Yes, that's it. The series stabiliser is two EL34 valves in parallel
and it has a very clever circuit for keeping them within their safe
operating area.
I'm not mad keen on your solution, though. I tend to be rather absent
minded!
Think of the HT switch as "Default Off" and only switch it on when you
are ready to do a test - then switch it straight off again. Absent
minded and working with high voltages can be a dangerous combination.
Take care!
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a
disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any,
but then the output tends to take quite a while to get up to it's stable
level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>> say, they both overshoot after warming up and then settle back down to >>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew >>>>> up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either, >>>> as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>> I'm not about to do that with supplies that are in all other respects, >>>> very good indeed and will probably still be around for many more years >>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all.
I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
On Thu, 9 Oct 2025 10:15:22 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
On Wed, 8 Oct 2025 22:06:38 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to >> >> this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's >> >> an initial voltage surge way beyond their indicated output levels. One >> >> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >> >> say, they both overshoot after warming up and then settle back down to >> >> their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >> >> but I'd like to mod these so they max-out at their indicated voltages >> >> and no more. Something preferably simple, like a crowbar, perhaps, but >> >> a bit more civilized. Any suggestions?
My Farnell power supply also does that. There are two switches, one for >> >the mains and one for the HT output. You just keep the output switched >> >off until the voltage has had time to settle down.
Might be the same supply if it's the 350V one by the sound of it;
that's a Farnell as well. It has 6 ranges, course and fine controls,
mA & V moving coil meters and a bunch of fixed low voltage outputs for
heaters and stuff. Same as yours?
Yes, that's it. The series stabiliser is two EL34 valves in parallel
and it has a very clever circuit for keeping them within their safe >operating area.
I'm not mad keen on your solution, though. I tend to be rather absent
minded!
Think of the HT switch as "Default Off" and only switch it on when you
are ready to do a test - then switch it straight off again. Absent
minded and working with high voltages can be a dangerous combination.
Take care!
Yeah, well, I've been goofing around with EHT stuff for more decades
now than I care to admit to so I must have *some* level of
self-preservation in my method!
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>>> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>> up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either, >>>>> as the low votage PSU is the one I'm using for that section of the
test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>>> I'm not about to do that with supplies that are in all other respects, >>>>> very good indeed and will probably still be around for many more years >>>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all.
I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and
getting an arc between the contacts can wreck them.
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>> this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One >>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>> say, they both overshoot after warming up and then settle back down to >>>> their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>> but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but >>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a
disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>> but then the output tends to take quite a while to get up to it's stable >>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
Cursitor Doom <cd@notformail.com> wrote:
[...]
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old
vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
I have uploaded the circuit diagram of that power supply to: ><http://www.poppyrecords.co.uk/other/Farnell350vPSU.gif>
and a sketch of the operating principle to: ><http://www.poppyrecords.co.uk/other/HTPSUprinciple.gif>John Larkin
Hope this helps.
On Fri, 10 Oct 2025 22:02:58 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
[...]
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old
vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
I have uploaded the circuit diagram of that power supply to: >><http://www.poppyrecords.co.uk/other/Farnell350vPSU.gif>
Gosh, that's horrible.
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>>>> wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>> switch-on, too, but that's not as concerning to me currently.
We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>> up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either, >>>>>> as the low votage PSU is the one I'm using for that section of the >>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>>>> I'm not about to do that with supplies that are in all other respects, >>>>>> very good indeed and will probably still be around for many more years >>>>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all.
I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and
getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>> this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's >>>>> an initial voltage surge way beyond their indicated output levels. One >>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>> say, they both overshoot after warming up and then settle back down to >>>>> their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a
disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>>> but then the output tends to take quite a while to get up to it's stable >>>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
On Fri, 10 Oct 2025 22:02:58 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
[...]
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old
vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
I have uploaded the circuit diagram of that power supply to: ><http://www.poppyrecords.co.uk/other/Farnell350vPSU.gif>
Gosh, that's horrible.
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>> wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>>>> an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages >>>>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either, >>>>>>> as the low votage PSU is the one I'm using for that section of the >>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>>>>> I'm not about to do that with supplies that are in all other respects, >>>>>>> very good indeed and will probably still be around for many more years >>>>>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It
could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and
getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts
move together they will eventually get to a spacing which a high voltage
can spark over. You get a glow discharge for a microsecond or so until
the surfaces start to heat up under ion bombardment and start distorting >into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may
not be all that great.
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>> say, they both overshoot after warming up and then settle back down to >>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>> disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>>>> but then the output tends to take quite a while to get up to it's stable >>>>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time.
Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>>> wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either, >>>>>>>> as the low votage PSU is the one I'm using for that section of the >>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>>>>>> I'm not about to do that with supplies that are in all other respects, >>>>>>>> very good indeed and will probably still be around for many more years >>>>>>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and
getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts
move together they will eventually get to a spacing which a high voltage
can spark over. You get a glow discharge for a microsecond or so until
the surfaces start to heat up under ion bombardment and start distorting
into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may
not be all that great.
We are all grateful that you don't actually design electronics.
On Sat, 11 Oct 2025 16:35:22 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>>> disturbance - you can configure them to be critically damped, where >>>>>> there only a very small overshoot, or over-damped where there isn't any, >>>>>> but then the output tends to take quite a while to get up to it's stable >>>>>> level.
The issue is probably nasty states caused by tubes heating up in the >>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time.
Reading books is really hard too.
Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
What in the world about this thread invoked Trump? We were talking
about tube power supply transients.
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>>> wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either, >>>>>>>> as the low votage PSU is the one I'm using for that section of the >>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and >>>>>>>> I'm not about to do that with supplies that are in all other respects, >>>>>>>> very good indeed and will probably still be around for many more years >>>>>>>> than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and
getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts >>move together they will eventually get to a spacing which a high voltage >>can spark over. You get a glow discharge for a microsecond or so until
the surfaces start to heat up under ion bombardment and start distorting >>into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may
not be all that great.
We are all grateful that you don't actually design electronics.
John Larkin--- Synchronet 3.21a-Linux NewsLink 1.2
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>> this question (well, I suppose it *could* be if someone chooses to
make it so). Anyway, they both have a design flaw inasmuch as there's >>>>> an initial voltage surge way beyond their indicated output levels. One >>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>> say, they both overshoot after warming up and then settle back down to >>>>> their expected output levels. The 350V one approaches 400V shortly
after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a
disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>>> but then the output tends to take quite a while to get up to it's stable >>>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>> say, they both overshoot after warming up and then settle back down to >>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>> disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>>>> but then the output tends to take quite a while to get up to it's stable >>>>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time. Your recommendation merely tells us merely that you like it - and you like Donald Trump.
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
tells me that it is yet another biography of Richard Feynman, and I've already read a couple of his autobiographies (and heard him lecture in person in Cambridge, as part of a large audience) . Feynman had lots of skills - safecracker was one of them - but I don't recall anything about practical electronics,
john larkin <jl@glen--canyon.com> wrote:
On Fri, 10 Oct 2025 22:02:58 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
[...]
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old
vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might
work with my Solartron HV PSU too, since that uses a similar system.
I have uploaded the circuit diagram of that power supply to:
<http://www.poppyrecords.co.uk/other/Farnell350vPSU.gif>
Gosh, that's horrible.
What have you found wrong with it?
On Sat, 11 Oct 2025 07:48:53 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the >>>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts
move together they will eventually get to a spacing which a high voltage >>> can spark over. You get a glow discharge for a microsecond or so until
the surfaces start to heat up under ion bombardment and start distorting >>> into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may
not be all that great.
We are all grateful that you don't actually design electronics.
John, you can be such a *bitch* at times! :-D
Bill Sloman <bill.sloman@ieee.org> wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>>> disturbance - you can configure them to be critically damped, where >>>>>> there only a very small overshoot, or over-damped where there isn't any, >>>>>> but then the output tends to take quite a while to get up to it's stable >>>>>> level.
The issue is probably nasty states caused by tubes heating up in the >>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time. Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
tells me that it is yet another biography of Richard Feynman, and I've
already read a couple of his autobiographies (and heard him lecture in
person in Cambridge, as part of a large audience) . Feynman had lots of
skills - safecracker was one of them - but I don't recall anything about
practical electronics,
In different book ("You are joking mister Feynman") Feynman wrote that
as a kid he earned money repairing radio receivers. One receiver was
making horrible noise after turn on, but worked OK after it was warm.
This was two tube design with both tubes of the same type. Feynman conjectured that one of tubes was heating faster and was responsible
for the noise. He swapped the tubes and this solved the problem.
Liz Tuddenham <liz@poppyrecords.invalid.invalid> wrote:
john larkin <jl@glen--canyon.com> wrote:
On Fri, 10 Oct 2025 22:02:58 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
Cursitor Doom <cd@notformail.com> wrote:
[...]
Let me get this right: if I use your method and just switch the HT
switch on/off for testing, I won't see any surge? I have *so* much old >> >> vintage test gear I simply cannot be thoroughly familiar with every
single piece of it; no one could. Anyway, you're saying just toggle
the HT on and off once the main deal has warmed up? If so, that might >> >> work with my Solartron HV PSU too, since that uses a similar system.
I have uploaded the circuit diagram of that power supply to:
<http://www.poppyrecords.co.uk/other/Farnell350vPSU.gif>
Gosh, that's horrible.
What have you found wrong with it?
Voltage divider setting output voltage is connected in wrong way.
On the picture trimpot T1 is connected to 0V line, but it must be
connected to slightly higher voltage.
Minor glitch is that left to milliamp meter line is marked with
-580V, on the right with -350V. I do not think that millamp
meter is dropping 230V.
On Fri, 10 Oct 2025 07:43:33 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>> say, they both overshoot after warming up and then settle back down to >>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>> disturbance - you can configure them to be critically damped, where
there only a very small overshoot, or over-damped where there isn't any, >>>>> but then the output tends to take quite a while to get up to it's stable >>>>> level.
The issue is probably nasty states caused by tubes heating up in the
wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
He doesn't need to as he's already one. In his own mind, anyway. ;-)
On 12/10/2025 1:48 am, john larkin wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the >>>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts
move together they will eventually get to a spacing which a high voltage >>> can spark over. You get a glow discharge for a microsecond or so until
the surfaces start to heat up under ion bombardment and start distorting >>> into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may
not be all that great.
We are all grateful that you don't actually design electronics.
Electronic circuit design is an art, and you aren't a master of it.
https://en.wikipedia.org/wiki/William_McGonagall#
does come to mind.
The point I made about relay contacts strikes me a sensible. Mercury
wetted relays would be less vulnerable to that kind of damage - the
mercury surface can't microweld - but it is a point that a competent
circuit designer would keep in mind. You seem to fall into the barely >competent category.
On Sat, 11 Oct 2025 07:48:53 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org> >>wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com> >>>>>>>>> wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew >>>>>>>>>> up transistor circuits.
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the >>>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit.
Not sure they'd be any good for longer duration surges as these are >>>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable. >>>>>>>>>>
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't >>>move anything like as fast as electrons in free space. As the contacts >>>move together they will eventually get to a spacing which a high voltage >>>can spark over. You get a glow discharge for a microsecond or so until >>>the surfaces start to heat up under ion bombardment and start distorting >>>into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may >>>not be all that great.
We are all grateful that you don't actually design electronics.
John, you can be such a *bitch* at times! :-D
On 12/10/2025 1:53 am, john larkin wrote:
On Sat, 11 Oct 2025 16:35:22 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>> switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>>>> disturbance - you can configure them to be critically damped, where >>>>>>> there only a very small overshoot, or over-damped where there isn't any,
but then the output tends to take quite a while to get up to it's stable
level.
The issue is probably nasty states caused by tubes heating up in the >>>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time.
Reading books is really hard too.
Not if you reading rate is unusually fast, and mine is.
Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
What in the world about this thread invoked Trump? We were talking
about tube power supply transients.
It's a measure of your susceptibility to confidence tricksters.
I note that you have snipped the link I posted to the actual book
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
which revealed what it was actually about, and my comment that a
biography of Richard Feynman didn't seem to be all that relevant to thread.
Bill Sloman <bill.sloman@ieee.org> wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>> after switch-on and the heftier one surges to over 600V before
settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at
switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>>> disturbance - you can configure them to be critically damped, where >>>>>> there only a very small overshoot, or over-damped where there isn't any, >>>>>> but then the output tends to take quite a while to get up to it's stable >>>>>> level.
The issue is probably nasty states caused by tubes heating up in the >>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time. Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
tells me that it is yet another biography of Richard Feynman, and I've
already read a couple of his autobiographies (and heard him lecture in
person in Cambridge, as part of a large audience) . Feynman had lots of
skills - safecracker was one of them - but I don't recall anything about
practical electronics,
In different book ("You are joking mister Feynman") Feynman wrote that
as a kid he earend money repairing radio receivers. One receiver was
making horrible noise after turn on, but worked OK after it was warm.
This was two tube design with both tubes of the same type. Feynman >conjectured that one of tubes was heating faster and was responsible
for the noise. He swapped the tubes and this solved the problem.
Does anybody here use mercury or mercury-wetted relays? They are big, expensive, toxic, and unreliable.
Digikey has one mercury tilt sensor for $203 plus $35 shipping.
john larkin <jl@glen--canyon.com> wrote:
[...]
Does anybody here use mercury or mercury-wetted relays? They are big,
expensive, toxic, and unreliable.
Digikey has one mercury tilt sensor for $203 plus $35 shipping.
I haven't used a mercury relay for many years but I did have to repair a >mercury tilt switch last year.
It was part of the contol system for the water supply in a house that
used pumped water from a well. The switch was mounted on a wooden plank
that spanned across the top of the water tank in the attic, operated by
a see-saw arrangement with a large cork float hanging by a chain on one
side and a balance weight hanging on the other side. It controlled the >fairly substantial motor of a pump that drew water from the well some >distance away and delivered it to the tank.
It had stopped working after a builder went up in the attic several
months previously, so the owner had returned the system to manual
control. When I investigated, I found the entire assembly in the bottom
of the tank, under water and still connected to the 240v mains supply.
It appears the builder had carelessly knocked it in but (sensibly)
decided not to try to retieve it from the water.
It was a horrible mess and I had to machine up new brass terminals, make >flexible braids and replace the over-centre springs. I fixed the plank
so it couldn't fall in again. The switch has worked now for over a
year and I expect it will outlive me and it's owner (...and the
builder!)
On Sun, 12 Oct 2025 15:12:20 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 12/10/2025 1:48 am, john larkin wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>>
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the >>>>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit. >>>>>>>>>>Not sure they'd be any good for longer duration surges as these are >>>>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts >>>> move together they will eventually get to a spacing which a high voltage >>>> can spark over. You get a glow discharge for a microsecond or so until >>>> the surfaces start to heat up under ion bombardment and start distorting >>>> into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may >>>> not be all that great.
We are all grateful that you don't actually design electronics.
Electronic circuit design is an art, and you aren't a master of it.
I did get a few pages in The Art of Electronics AOE3, and I'm named 22
times in the X chapters, and I'm in the dedications, and in the index.
I admit that I have no patents related to stirred water baths. I do
have a few customers.
https://en.wikipedia.org/wiki/William_McGonagall#
does come to mind.
The point I made about relay contacts strikes me a sensible. Mercury
wetted relays would be less vulnerable to that kind of damage - the
mercury surface can't microweld - but it is a point that a competent
circuit designer would keep in mind. You seem to fall into the barely
competent category.
Most ordinary relays will make bad "dry" contacts, namely do bad
low-level switching, without enough voltage and current to spark the
contacts a bit and blast through oxide.
Does anybody here use mercury or mercury-wetted relays? They are big, expensive, toxic, and unreliable.
Digikey has one mercury tilt sensor for $203 plus $35 shipping.
On Sun, 12 Oct 2025 11:18:17 +0100, Cursitor Doom <cd@notformail.com>
wrote:
On Sat, 11 Oct 2025 07:48:53 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom <cd@notformail.com> >>>>>>> wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--canyon.com> >>>>>>>> wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom <cd@notformail.com> >>>>>>>>> wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom <cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. It blew
I've a couple of vintage HT power supplies which date from around 50
years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's
an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages
and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>>
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any favors either,
as the low votage PSU is the one I'm using for that section of the >>>>>>>>>> test rig.
You could hang a big MOV or TVS across your load circuit. >>>>>>>>>>Not sure they'd be any good for longer duration surges as these are >>>>>>>>>> (~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make space and
I'm not about to do that with supplies that are in all other respects,
very good indeed and will probably still be around for many more years
than the new replacement would be!
The timer idea put forward by you and Edward is the frontrunner so >>>>>>>>>> far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea at all. >>>>>>>> I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't
move anything like as fast as electrons in free space. As the contacts >>>> move together they will eventually get to a spacing which a high voltage >>>> can spark over. You get a glow discharge for a microsecond or so until >>>> the surfaces start to heat up under ion bombardment and start distorting >>>> into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may >>>> not be all that great.
We are all grateful that you don't actually design electronics.
John, you can be such a *bitch* at times! :-D
Sure.
I'm friendly and cheerful and helpful to decent people, but don't much
like creeps.
Sorry.
On Sun, 12 Oct 2025 15:21:58 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 12/10/2025 1:53 am, john larkin wrote:
On Sat, 11 Oct 2025 16:35:22 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,Almost every second order control loop gives a ringing response to a >>>>>>>> disturbance - you can configure them to be critically damped, where >>>>>>>> there only a very small overshoot, or over-damped where there isn't any,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to
this question (well, I suppose it *could* be if someone chooses to >>>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>>>> an initial voltage surge way beyond their indicated output levels. One
of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I
say, they both overshoot after warming up and then settle back down to
their expected output levels. The 350V one approaches 400V shortly >>>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago,
but I'd like to mod these so they max-out at their indicated voltages >>>>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>
but then the output tends to take quite a while to get up to it's stable
level.
The issue is probably nasty states caused by tubes heating up in the >>>>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time.
Reading books is really hard too.
Not if you reading rate is unusually fast, and mine is.
I read a book in, typically, two days, or one if I'm on vacation. 400
WPM.
I just read The Secret Garden in a day and a half. It's a children's
book, very well written, profound in places.
The result is a huge accumulation of books. If it's not something
worth rereading many times, I donate them to little free libraries. We
have maybe 5 of them along my walk from home to work.
https://www.mercurynews.com/wp-content/uploads/2022/11/MAG-LITTLEFREE-12.jpg?w=1568
Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
What in the world about this thread invoked Trump? We were talking
about tube power supply transients.
It's a measure of your susceptibility to confidence tricksters.
I note that you have snipped the link I posted to the actual book
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
which revealed what it was actually about, and my comment that a
biography of Richard Feynman didn't seem to be all that relevant to thread.
There is a part about how, as a kid, he fixed a radio by swapping two
tubes. He figured that an oscillation could be caused by different
filament warmup rates.
On Sun, 12 Oct 2025 13:41:17 -0000 (UTC), antispam@fricas.org (Waldek Hebisch) wrote:
Bill Sloman <bill.sloman@ieee.org> wrote:
On 11/10/2025 1:43 am, john larkin wrote:
On Fri, 10 Oct 2025 16:56:50 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 10/10/2025 1:58 am, john larkin wrote:
On Thu, 9 Oct 2025 18:19:49 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>>> wrote:
On 9/10/2025 3:57 am, Cursitor Doom wrote:
Gentlemen,
I've a couple of vintage HT power supplies which date from around 50 >>>>>>>> years ago; valve jobs. Anyway, the valve bit is not really relevant to >>>>>>>> this question (well, I suppose it *could* be if someone chooses to >>>>>>>> make it so). Anyway, they both have a design flaw inasmuch as there's >>>>>>>> an initial voltage surge way beyond their indicated output levels. One >>>>>>>> of these PSUs goes up to a max of 350VDC and the other to 400VDC. As I >>>>>>>> say, they both overshoot after warming up and then settle back down to >>>>>>>> their expected output levels. The 350V one approaches 400V shortly >>>>>>>> after switch-on and the heftier one surges to over 600V before >>>>>>>> settling back. Now, that may have been acceptable enough 50 years ago, >>>>>>>> but I'd like to mod these so they max-out at their indicated voltages >>>>>>>> and no more. Something preferably simple, like a crowbar, perhaps, but >>>>>>>> a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>> switch-on, too, but that's not as concerning to me currently.
Almost every second order control loop gives a ringing response to a >>>>>>> disturbance - you can configure them to be critically damped, where >>>>>>> there only a very small overshoot, or over-damped where there isn't any,
but then the output tends to take quite a while to get up to it's stable
level.
The issue is probably nasty states caused by tubes heating up in the >>>>>> wrong order.
See the book Genius by James Gleick.
Or so John Larkin wants to tell us. Citing a book called "Genius"
doesn't make you one.
OK, don't read it. I can see why you wouldn't want to.
Even downloading books costs money and time. Your recommendation merely
tells us merely that you like it - and you like Donald Trump.
https://en.wikipedia.org/wiki/Genius:_The_Life_and_Science_of_Richard_Feynman
tells me that it is yet another biography of Richard Feynman, and I've
already read a couple of his autobiographies (and heard him lecture in
person in Cambridge, as part of a large audience) . Feynman had lots of
skills - safecracker was one of them - but I don't recall anything about >>> practical electronics,
In different book ("You are joking mister Feynman") Feynman wrote that
as a kid he earend money repairing radio receivers. One receiver was
making horrible noise after turn on, but worked OK after it was warm.
This was two tube design with both tubes of the same type. Feynman
conjectured that one of tubes was heating faster and was responsible
for the noise. He swapped the tubes and this solved the problem.
Yes, that was my Feynman reference. I thought it was in "Genius" but I
guess it's in Surely You Are Joking.
He was a good circuit designer as well as a serious physicist. That's
an unusual combination.
On 13/10/2025 3:31 am, john larkin wrote:
On Sun, 12 Oct 2025 15:12:20 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 12/10/2025 1:48 am, john larkin wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org> >>>> wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman
<bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen--
canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen-- >>>>>>>>>>> canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
Gentlemen,We had a Heathkit tube bench supply that did that at turnon. >>>>>>>>>>>> It blew
I've a couple of vintage HT power supplies which date from >>>>>>>>>>>>> around 50
years ago; valve jobs. Anyway, the valve bit is not really >>>>>>>>>>>>> relevant to
this question (well, I suppose it *could* be if someone >>>>>>>>>>>>> chooses to
make it so). Anyway, they both have a design flaw inasmuch >>>>>>>>>>>>> as there's
an initial voltage surge way beyond their indicated output >>>>>>>>>>>>> levels. One
of these PSUs goes up to a max of 350VDC and the other to >>>>>>>>>>>>> 400VDC. As I
say, they both overshoot after warming up and then settle >>>>>>>>>>>>> back down to
their expected output levels. The 350V one approaches 400V >>>>>>>>>>>>> shortly
after switch-on and the heftier one surges to over 600V before >>>>>>>>>>>>> settling back. Now, that may have been acceptable enough 50 >>>>>>>>>>>>> years ago,
but I'd like to mod these so they max-out at their
indicated voltages
and no more. Something preferably simple, like a crowbar, >>>>>>>>>>>>> perhaps, but
a bit more civilized. Any suggestions?
Your pal,
CD.
PS: I've another low voltage PSU that gives a current surge at >>>>>>>>>>>>> switch-on, too, but that's not as concerning to me currently. >>>>>>>>>>>>
up transistor circuits.
It's not doing the heaters in the tubes I'm testing any >>>>>>>>>>> favors either,
as the low votage PSU is the one I'm using for that section >>>>>>>>>>> of the
test rig.
You could hang a big MOV or TVS across your load circuit. >>>>>>>>>>>Not sure they'd be any good for longer duration surges as >>>>>>>>>>> these are
(~4 seconds)
Or put a time delay relay inside the supply. Or a pushbutton >>>>>>>>>>>> enable.
Or buy a new power supply!
No, if I did that I'd have to junk the existing ones to make >>>>>>>>>>> space and
I'm not about to do that with supplies that are in all other >>>>>>>>>>> respects,
very good indeed and will probably still be around for many >>>>>>>>>>> more years
than the new replacement would be!
The timer idea put forward by you and Edward is the
frontrunner so
far.
John Larkin
Highland Tech Glen Canyon Design Center
Lunatic Fringe Electronics
How about a pushbutton that muat be pushed once after powerup? It >>>>>>>>>> could even light up to remind you to push it.
Yeah, I can see where you're coming from here. Not a bad idea >>>>>>>>> at all.
I might adopt that.
Actually, all you need is a dpdt relay and a pushbutton.
Too easy, no fun.
Finding relays rated for high voltages takes a bit of an effort, and >>>>>>> getting an arc between the contacts can wreck them.
It only has to close on high voltage, not open.
That's easy.
It might look that way until you think about it. Relay contacts don't >>>>> move anything like as fast as electrons in free space. As the contacts >>>>> move together they will eventually get to a spacing which a high
voltage
can spark over. You get a glow discharge for a microsecond or so until >>>>> the surfaces start to heat up under ion bombardment and start
distorting
into the spikes which can deliver enough current to sustain an arc.
You can get microwelding when the relay finally closes. Relay life may >>>>> not be all that great.
We are all grateful that you don't actually design electronics.
Electronic circuit design is an art, and you aren't a master of it.
I did get a few pages in The Art of Electronics AOE3, and I'm named 22
times in the X chapters, and I'm in the dedications, and in the index.
I admit that I have no patents related to stirred water baths. I do
have a few customers.
https://en.wikipedia.org/wiki/William_McGonagall#
does come to mind.
The point I made about relay contacts strikes me a sensible. Mercury
wetted relays would be less vulnerable to that kind of damage - the
mercury surface can't microweld - but it is a point that a competent
circuit designer would keep in mind. You seem to fall into the barely
competent category.
Most ordinary relays will make bad "dry" contacts, namely do bad
low-level switching, without enough voltage and current to spark the
contacts a bit and blast through oxide.
Does anybody here use mercury or mercury-wetted relays? They are big,
expensive, toxic, and unreliable.
Mercury-wetted reed switches aren't big, and they weren't particularly expensive when I was using them. Regular dry reed switches are good for about 10 million contact closures, and mercury-wetted reeds are good for
100 million, if you operate them within their specifications - anything
but unreliable.
Digikey has one mercury tilt sensor for $203 plus $35 shipping.
That's not what we were talking about.
Pickering used to be a reliable source of mercury wetted reed relays but
now they only cater to the legacy market. Health and safety is now
nervous about mercury, even the tiny amount encapsulated in a glass reed capsule.
The mercury can only get out if the capsule is smashed, so they aren't toxic.
On Sun, 12 Oct 2025 19:29:41 +0100, liz@poppyrecords.invalid.invalid
(Liz Tuddenham) wrote:
john larkin <jl@glen--canyon.com> wrote:
[...]
Does anybody here use mercury or mercury-wetted relays? They are big,
expensive, toxic, and unreliable.
Digikey has one mercury tilt sensor for $203 plus $35 shipping.
I haven't used a mercury relay for many years but I did have to repair a >mercury tilt switch last year.
It was part of the contol system for the water supply in a house that
used pumped water from a well. The switch was mounted on a wooden plank >that spanned across the top of the water tank in the attic, operated by
a see-saw arrangement with a large cork float hanging by a chain on one >side and a balance weight hanging on the other side. It controlled the >fairly substantial motor of a pump that drew water from the well some >distance away and delivered it to the tank.
It had stopped working after a builder went up in the attic several
months previously, so the owner had returned the system to manual
control. When I investigated, I found the entire assembly in the bottom
of the tank, under water and still connected to the 240v mains supply.
It appears the builder had carelessly knocked it in but (sensibly)
decided not to try to retieve it from the water.
It was a horrible mess and I had to machine up new brass terminals, make >flexible braids and replace the over-centre springs. I fixed the plank
so it couldn't fall in again. The switch has worked now for over a
year and I expect it will outlive me and it's owner (...and the
builder!)
I bought a ton of old electronic surplus from some deceased ham's
estate and discovered a bunch of old mercury tilt switches among the 'treasures' after they'd been delivered. I know when they would have
been made, but clearly it was a *long* time ago. There must have been
half an ounce of mercury metal in each switch easily! In bygone times, everything was over-engineered it seems. Fabulous quality as well.
On 10/12/25 21:07, Bill Sloman wrote:
On 13/10/2025 3:31 am, john larkin wrote:
On Sun, 12 Oct 2025 15:12:20 +1100, Bill Sloman <bill.sloman@ieee.org>
wrote:
On 12/10/2025 1:48 am, john larkin wrote:
On Sat, 11 Oct 2025 16:11:12 +1100, Bill Sloman <bill.sloman@ieee.org> >>>>> wrote:
On 11/10/2025 1:42 am, john larkin wrote:
On Fri, 10 Oct 2025 17:04:09 +1100, Bill Sloman
<bill.sloman@ieee.org>
wrote:
On 10/10/2025 9:59 am, john larkin wrote:
On Thu, 09 Oct 2025 23:54:04 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
On Thu, 09 Oct 2025 07:53:59 -0700, john larkin <jl@glen-- >>>>>>>>>> canyon.com>
wrote:
On Wed, 08 Oct 2025 23:27:27 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
On Wed, 08 Oct 2025 12:24:09 -0700, john larkin <jl@glen-- >>>>>>>>>>>> canyon.com>
wrote:
On Wed, 08 Oct 2025 17:57:31 +0100, Cursitor Doom
<cd@notformail.com>
wrote:
Pickering used to be a reliable source of mercury wetted reed relaysThat may be perfectly true, but lawyers can be very toxic...bad for your health & wealth!
but now they only cater to the legacy market. Health and safety is now
nervous about mercury, even the tiny amount encapsulated in a glass
reed capsule.
The mercury can only get out if the capsule is smashed, so they aren't
toxic.
