| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 42 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 01:51:16 |
| Calls: | 220 |
| Calls today: | 1 |
| Files: | 824 |
| Messages: | 121,543 |
| Posted today: | 6 |
Found a teardown video and saw a custom ribbon cable utilized to connect
the keyboard to the mobo - so that's a complication. After a bit of searching, they utilized one of the four usb ports in the hub for the keyboard logic.
Has anyone managed a similar project with the experience to share?
This past summer was a good time to lower the power bill and overall temperature of the den by transitioning to rpi's for my computing
needs. The main pc was replaced by a pi400. Lovely, silent, low
power. Lovely.
Well, mostly so. I really don't like the keyboard and question why they
went with it. The full sized keyboard is something I miss, you know,
with the full row of function keys, dedicated number pad, full sized
arrows, the pageup.down,insert,delete cluster. The keys themselves suck
- where they often fail to register key presses and I have to fix
spelling quite often. I had hoped that the keys would wear in and solve
the problem, but it really just seems to be bad design. Hopes are it's
simply bad luck and a bad keyboard.
This past summer was a good time to lower the power bill and overall temperature of the den by transitioning to rpi's for my computing
needs. The main pc was replaced by a pi400. Lovely, silent, low
power. Lovely.
Well, mostly so. I really don't like the keyboard and question why they
went with it. The full sized keyboard is something I miss, you know,
with the full row of function keys, dedicated number pad, full sized
arrows, the pageup.down,insert,delete cluster. The keys themselves suck
- where they often fail to register key presses and I have to fix
spelling quite often. I had hoped that the keys would wear in and solve
the problem, but it really just seems to be bad design. Hopes are it's
simply bad luck and a bad keyboard.
The problem is easy to solve by plugging in a usb keyboard, but it
defeats the appeal and purpose of the all-in-one design.
How hard would it be to rig the mobo into a full size keyboard shell?
Design a new lower shell of a chosen keyboard to accomodate the mobo and
port holes for 3d printing. A search online didn't yield evidence that
it has been done yet. But if this thing with it's massive heat sync can
fit in this small keyboard, it shouldn't have issues fitting in full
sized keybord.
Found a teardown video and saw a custom ribbon cable utilized to connect
the keyboard to the mobo - so that's a complication. After a bit of searching, they utilized one of the four usb ports in the hub for the keyboard logic.
Daniel <me@sc1f1dan.com> wrote:
This past summer was a good time to lower the power bill and overall
temperature of the den by transitioning to rpi's for my computing
needs. The main pc was replaced by a pi400. Lovely, silent, low
power. Lovely.
Well, mostly so. I really don't like the keyboard and question why they
went with it. The full sized keyboard is something I miss, you know,
with the full row of function keys, dedicated number pad, full sized
arrows, the pageup.down,insert,delete cluster. The keys themselves suck
- where they often fail to register key presses and I have to fix
spelling quite often. I had hoped that the keys would wear in and solve
the problem, but it really just seems to be bad design. Hopes are it's
simply bad luck and a bad keyboard.
The problem is easy to solve by plugging in a usb keyboard, but it
defeats the appeal and purpose of the all-in-one design.
How hard would it be to rig the mobo into a full size keyboard shell?
Design a new lower shell of a chosen keyboard to accomodate the mobo and
port holes for 3d printing. A search online didn't yield evidence that
it has been done yet. But if this thing with it's massive heat sync can
fit in this small keyboard, it shouldn't have issues fitting in full
sized keybord.
https://forum.diyperks.com/user-projects/raspberry-pi-400-x-mechanical-keyboard/
https://www.printables.com/model/318803-raspberry-pi-400-mechanical-keyboard-upgrade
https://hackaday.io/project/175844-the-mechanical-pi-400 https://www.tomshardware.com/news/orthopi-raspberry-pi-400-ortholinear-mod
are a few examples I found by searching 'rpi 400 mechanical keyboard'.
There are no doubt others.
Found a teardown video and saw a custom ribbon cable utilized to connect
the keyboard to the mobo - so that's a complication. After a bit of
searching, they utilized one of the four usb ports in the hub for the
keyboard logic.
The keyboard has a Holtek HT45R0072 (one time programmed) microcontroller to convert the matrix to USB. Your options are either to repurpose this microcontroller, at which point you need to match the keyboard matrix of the Pi400 (hard):
https://www.40percent.club/2020/12/orthopi.html
or just ignore it and use a USB port. Another alternative is to cut the traces to the Holtek (or unsolder it) and use its USB lines: https://www.40percent.club/2020/11/raspberry-pi-400-keyboard-controller.html
This past summer was a good time to lower the power bill and overall temperature of the den by transitioning to rpi's for my computing
needs. The main pc was replaced by a pi400. Lovely, silent, low
power. Lovely.
Well, mostly so. I really don't like the keyboard and question why they
went with it. The full sized keyboard is something I miss, you know,
with the full row of function keys, dedicated number pad, full sized
arrows, the pageup.down,insert,delete cluster. The keys themselves suck
- where they often fail to register key presses and I have to fix
spelling quite often. I had hoped that the keys would wear in and solve
the problem, but it really just seems to be bad design. Hopes are it's
simply bad luck and a bad keyboard.
The problem is easy to solve by plugging in a usb keyboard, but it
defeats the appeal and purpose of the all-in-one design.
How hard would it be to rig the mobo into a full size keyboard shell?
Design a new lower shell of a chosen keyboard to accomodate the mobo and
port holes for 3d printing. A search online didn't yield evidence that
it has been done yet. But if this thing with it's massive heat sync can
fit in this small keyboard, it shouldn't have issues fitting in full
sized keybord.
Found a teardown video and saw a custom ribbon cable utilized to connect
the keyboard to the mobo - so that's a complication. After a bit of searching, they utilized one of the four usb ports in the hub for the keyboard logic.
Has anyone managed a similar project with the experience to share?
Daniel
Well, I'm still on the quest for my modded Pi400. I'm currently on one
of the keyboards that seemed promising in terms of shell depth
(Logitech K120 with silicon membrane). The illusion, I realize, is in
the full-sized keys and their travel with the membrane below. I'm sort
of addicted to this thing, as it's plugged into my pi400.
On 2024-12-02, Daniel <me@sc1f1dan.com> wrote:
Well, I'm still on the quest for my modded Pi400. I'm currently on one
of the keyboards that seemed promising in terms of shell depth
(Logitech K120 with silicon membrane). The illusion, I realize, is in
the full-sized keys and their travel with the membrane below. I'm sort
of addicted to this thing, as it's plugged into my pi400.
I am typing this on an original IBM Model M. Since I have a disassembled spare sitting next to me, I just checked: a normal Banana Pi Board (same
size as an old RPi) fits fine into the space under the keys at the back - so if you de-solder the GPIO header and solder in an angled connector, this might work ;-)
You would need to convert the keyboard to USB, eg. using https://www.schwingen.org/modelm-usb/
This past summer was a good time to lower the power bill and overall temperature of the den by transitioning to rpi's for my computing
needs. The main pc was replaced by a pi400. Lovely, silent, low
power. Lovely.
Well, mostly so. I really don't like the keyboard and question why they
went with it. The full sized keyboard is something I miss, you know,
with the full row of function keys, dedicated number pad, full sized
arrows, the pageup.down,insert,delete cluster. The keys themselves suck
- where they often fail to register key presses and I have to fix
spelling quite often. I had hoped that the keys would wear in and solve
the problem, but it really just seems to be bad design. Hopes are it's
simply bad luck and a bad keyboard.
The problem is easy to solve by plugging in a usb keyboard, but it
defeats the appeal and purpose of the all-in-one design.
How hard would it be to rig the mobo into a full size keyboard shell?
Design a new lower shell of a chosen keyboard to accomodate the mobo and
port holes for 3d printing. A search online didn't yield evidence that
it has been done yet. But if this thing with it's massive heat sync can
fit in this small keyboard, it shouldn't have issues fitting in full
sized keybord.
Found a teardown video and saw a custom ribbon cable utilized to connect
the keyboard to the mobo - so that's a complication. After a bit of searching, they utilized one of the four usb ports in the hub for the keyboard logic.
Has anyone managed a similar project with the experience to share?
Daniel
On 12/10/24 17:48, Daniel wrote:
Wellp, glad I didn't very far on my homebrew pi400 project. Now that
they quietly released the Pi500, I'll be focusing on that now.
I still have to watch the youtube reviews of it.
I read Jeff Geerling's recent write-up on it (https://www.jeffgeerling.com/blog/2024/pi-500-much-faster-lacks-m2)
as a "mixed review", and would be wary of investing even the $120 for
one (the kit, or $90 for the standalone). YMMV
Daniel wrote:
Now that they quietly released the Pi500, I'll be focusing on thatnow.
I read Jeff Geerling's recent write-up on it (https:// www.jeffgeerling.com/blog/2024/pi-500-much-faster-lacks-m2) as a "mixed review", and would be wary of investing even the $120 for one (the kit,
or $90 for the standalone).
Jonathan B. Horen wrote:
Daniel wrote:
Now that they quietly released the Pi500, I'll be focusing on thatnow.
I read Jeff Geerling's recent write-up on it (https://
www.jeffgeerling.com/blog/2024/pi-500-much-faster-lacks-m2) as a
"mixed review", and would be wary of investing even the $120 for one
(the kit, or $90 for the standalone).
Presumably, before too long there will be a Pi5x0 where the M.2 socket
and associated components are fitted.
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced
in so no power.
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced >>> in so no power.
There were some nearby component missing too, maybe the Rev1 board has
bugs, and they need a Rev2 (or whatever) but his soldering job looked
relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no engineer.
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced
in so no power.
There were some nearby component missing too, maybe the Rev1 board has
bugs, and they need a Rev2 (or whatever) but his soldering job looked relatively clumsy (bridged pins etc).
On 11/12/2024 16:46, Daniel wrote:
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced >>> in so no power.
There were some nearby component missing too, maybe the Rev1 board has
bugs, and they need a Rev2 (or whatever) but his soldering job looked
relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no engineer.
Why would you put support chips on if you were not going to install an
M.2 socket anyway?
I think he called it right in that this is a bare bones model, and more upmarket versions will emerge in due course.
The Natural Philosopher <tnp@invalid.invalid> wrote:
On 11/12/2024 16:46, Daniel wrote:
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced
in so no power.
There were some nearby component missing too, maybe the Rev1 board has >> bugs, and they need a Rev2 (or whatever) but his soldering job looked
relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no engineer.
Why would you put support chips on if you were not going to install an
M.2 socket anyway?
I think he called it right in that this is a bare bones model, and more upmarket versions will emerge in due course.
Looking at: https://www.jeffgeerling.com/sites/default/files/images/pi-500-m2-socket-install.jpg
https://www.jeffgeerling.com/sites/default/files/images/pi-500-pcb-bottom.jpeg
I think there's a MOSFET and some decoupling capacitors missing (to the left of the middle larger hole), so the power isn't connected. There may be a pair of missing capacitors in series with the PCIe lane, but I can't immediately see the footprints for those.
...I have to fix spelling quite often.
Re: Re: Homebrew pi400^^^^^^^
By: Daniel to Daniel on Mon Dec 02 2024 17:03:16
...I have to fix spelling quite often.
Go on any social media site. Nobody would notice, unfortunately.
Glad to see someboy still bothers to do it, though.
Theo <theom+news@chiark.greenend.org.uk> wrote:
The Natural Philosopher <tnp@invalid.invalid> wrote:
On 11/12/2024 16:46, Daniel wrote:
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket
hoping it was properly designed. He found the power lines weren't traced
in so no power.
There were some nearby component missing too, maybe the Rev1 board has >> > >> bugs, and they need a Rev2 (or whatever) but his soldering job looked >> > >> relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no >> > > engineer.
Why would you put support chips on if you were not going to install an
M.2 socket anyway?
I think he called it right in that this is a bare bones model, and more
upmarket versions will emerge in due course.
Looking at:
https://www.jeffgeerling.com/sites/default/files/images/pi-500-m2-socket-install.jpg
https://www.jeffgeerling.com/sites/default/files/images/pi-500-pcb-bottom.jpeg
I think there's a MOSFET and some decoupling capacitors missing (to the left >> of the middle larger hole), so the power isn't connected. There may be a
pair of missing capacitors in series with the PCIe lane, but I can't
immediately see the footprints for those.
Ah, I see them now. To the right of the spot on the silver chip are four tiny spots for passives, oriented at compass bearing 135 degrees. I can't tell from the photo or video whether they're populated or not, but those
will be the series capacitors.
Theo
Theo <theom+news@chiark.greenend.org.uk> writes:Wait for the Pi 501...with it all one the board already
Theo <theom+news@chiark.greenend.org.uk> wrote:
The Natural Philosopher <tnp@invalid.invalid> wrote:
On 11/12/2024 16:46, Daniel wrote:
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket >>>>>>> hoping it was properly designed. He found the power lines weren't traced
in so no power.
There were some nearby component missing too, maybe the Rev1 board has >>>>>> bugs, and they need a Rev2 (or whatever) but his soldering job looked >>>>>> relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no >>>>> engineer.
Why would you put support chips on if you were not going to install an >>>> M.2 socket anyway?
I think he called it right in that this is a bare bones model, and more >>>> upmarket versions will emerge in due course.
Looking at:
https://www.jeffgeerling.com/sites/default/files/images/pi-500-m2-socket-install.jpg
https://www.jeffgeerling.com/sites/default/files/images/pi-500-pcb-bottom.jpeg
I think there's a MOSFET and some decoupling capacitors missing (to the left
of the middle larger hole), so the power isn't connected. There may be a >>> pair of missing capacitors in series with the PCIe lane, but I can't
immediately see the footprints for those.
Ah, I see them now. To the right of the spot on the silver chip are four
tiny spots for passives, oriented at compass bearing 135 degrees. I can't >> tell from the photo or video whether they're populated or not, but those
will be the series capacitors.
Theo
I'm going to wait for someone to hack a working solution then copy
it.
On 13/12/2024 23:55, Daniel wrote:
Theo <theom+news@chiark.greenend.org.uk> writes:Wait for the Pi 501...with it all one the board already
Theo <theom+news@chiark.greenend.org.uk> wrote:I'm going to wait for someone to hack a working solution then copy
The Natural Philosopher <tnp@invalid.invalid> wrote:
On 11/12/2024 16:46, Daniel wrote:
Andy Burns <usenet@andyburns.uk> writes:
Daniel wrote:
Jeff did a teardown on his second channel and soldered a m.2 socket >>>>>>>> hoping it was properly designed. He found the power lines weren't traced
in so no power.
There were some nearby component missing too, maybe the Rev1 board has >>>>>>> bugs, and they need a Rev2 (or whatever) but his soldering job looked >>>>>>> relatively clumsy (bridged pins etc).
Yeah, it was pretty bad, at least he was honest about it.
I am sure things were missing to keep the circuitry disabled and I'm no >>>>>> engineer.
Why would you put support chips on if you were not going to install an >>>>> M.2 socket anyway?
I think he called it right in that this is a bare bones model, and more >>>>> upmarket versions will emerge in due course.
Looking at:
https://www.jeffgeerling.com/sites/default/files/images/pi-500-m2-socket-install.jpg
https://www.jeffgeerling.com/sites/default/files/images/pi-500-pcb-bottom.jpeg
I think there's a MOSFET and some decoupling capacitors missing (to the left
of the middle larger hole), so the power isn't connected. There may be a >>>> pair of missing capacitors in series with the PCIe lane, but I can't
immediately see the footprints for those.
Ah, I see them now. To the right of the spot on the silver chip are four >>> tiny spots for passives, oriented at compass bearing 135 degrees. I can't >>> tell from the photo or video whether they're populated or not, but those >>> will be the series capacitors.
Theo
it.
The Natural Philosopher <tnp@invalid.invalid> writes:
On 13/12/2024 23:55, Daniel wrote:
I'm going to wait for someone to hack a working solution then copyWait for the Pi 501...with it all one the board already
it.
I've heard rumor of it but is this confirmed?
On 15/12/2024 21:11, Daniel wrote:
The Natural Philosopher <tnp@invalid.invalid> writes:
On 13/12/2024 23:55, Daniel wrote:
I'm going to wait for someone to hack a working solution then copyWait for the Pi 501...with it all one the board already
it.
I've heard rumor of it but is this confirmed?
I was actually semi-joking.
But the demand for a fast Pi with nvme in a neat package is there all
right.
