• Why are BFR181 transistors so fragile?

    From Jean-Pierre Coulon@coulon@cacas.pam.obs-nice.fr to sci.electronics.design on Thu Feb 26 18:59:40 2026
    From Newsgroup: sci.electronics.design

    I am trying to replace my BFR193 transistors with BFR181 but they no
    longer do their transistor job once I solder them. I am at Ic = 2 mA, Vce
    = 3 V.

    Am I the only one is this case ?
    --
    Jean-Pierre Coulon
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Thu Feb 26 10:41:53 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 18:59:40 +0100, Jean-Pierre Coulon <coulon@cacas.pam.obs-nice.fr> wrote:

    I am trying to replace my BFR193 transistors with BFR181 but they no
    longer do their transistor job once I solder them. I am at Ic = 2 mA, Vce
    = 3 V.

    Am I the only one is this case ?

    Yikes. 20 mA abs max! It must be a really tiny chip.

    Please post your schematic.

    The c-b junctions of those tiny RF parts make amazing diodes.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Jeroen Belleman@jeroen@nospam.please to sci.electronics.design on Thu Feb 26 19:55:42 2026
    From Newsgroup: sci.electronics.design

    On 2/26/26 18:59, Jean-Pierre Coulon wrote:
    I am trying to replace my BFR193 transistors with BFR181 but they no
    longer do their transistor job once I solder them. I am at Ic = 2 mA,
    Vce = 3 V.

    Am I the only one is this case ?


    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Jeroen Belleman
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Thu Feb 26 11:02:25 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 19:55:42 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 18:59, Jean-Pierre Coulon wrote:
    I am trying to replace my BFR193 transistors with BFR181 but they no
    longer do their transistor job once I solder them. I am at Ic = 2 mA,
    Vce = 3 V.

    Am I the only one is this case ?


    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Jeroen Belleman

    Good point. It might oscillate itself to death if enough power supply
    is available.




    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Jean-Pierre Coulon@coulon@cacas.pam.obs-nice.fr to sci.electronics.design on Thu Feb 26 20:50:50 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Thu Feb 26 20:05:54 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Jeroen Belleman@jeroen@nospam.please to sci.electronics.design on Thu Feb 26 21:54:14 2026
    From Newsgroup: sci.electronics.design

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Thu Feb 26 13:06:21 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    And where's the output?

    There doesn't seem to be enough energy available to blow up
    transistors.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Thu Feb 26 21:20:48 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the
    infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why
    use these in the first place?
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Thu Feb 26 16:44:48 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz
    for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk >>
    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why
    use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Fri Feb 27 05:10:18 2026
    From Newsgroup: sci.electronics.design

    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk >>>
    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why
    use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Fri Feb 27 11:19:44 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman >>><jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>>>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk >>>>
    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>infineon models and not that simplistic mytotor model). Adding >>>resistance or bead to Q2 base will stop it, but you're correct - why
    use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    Version 4.1
    SHEET 1 6000 680
    WIRE 608 -416 368 -416
    WIRE 864 -416 608 -416
    WIRE 864 -368 864 -416
    WIRE 368 -336 368 -416
    WIRE 608 -256 608 -416
    WIRE 864 -256 864 -288
    WIRE 272 -224 256 -224
    WIRE 368 -224 368 -256
    WIRE 368 -224 336 -224
    WIRE 368 -208 368 -224
    WIRE 544 -208 368 -208
    WIRE 368 -160 368 -208
    WIRE 608 -128 608 -160
    WIRE 720 -128 608 -128
    WIRE 832 -128 720 -128
    WIRE -16 -112 -80 -112
    WIRE 128 -112 64 -112
    WIRE 208 -112 128 -112
    WIRE 256 -112 256 -224
    WIRE 256 -112 208 -112
    WIRE 304 -112 256 -112
    WIRE 608 -96 608 -128
    WIRE 720 -96 720 -128
    WIRE 128 -64 128 -112
    WIRE -80 -48 -80 -112
    WIRE 368 -32 368 -64
    WIRE 608 0 608 -16
    WIRE 720 0 720 -32
    WIRE 720 0 608 0
    WIRE 208 16 208 -112
    WIRE 432 16 208 16
    WIRE 608 16 608 0
    WIRE 608 16 512 16
    WIRE 128 32 128 0
    WIRE 608 48 608 16
    WIRE 608 176 608 128
    WIRE 608 304 608 256
    FLAG 368 -32 0
    FLAG 864 -256 0
    FLAG 608 304 0
    FLAG 128 32 0
    FLAG -80 -48 0
    SYMBOL npn 304 -160 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR181
    SYMBOL res 352 -352 R0
    SYMATTR InstName R9
    SYMATTR Value 5.6k
    SYMBOL cap 336 -240 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 0.1p
    SYMBOL res 80 -128 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R1
    SYMATTR Value 10k
    SYMBOL voltage 864 -384 R0
    SYMATTR InstName V2
    SYMATTR Value 5
    SYMBOL npn 544 -256 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR181
    SYMBOL res 592 -112 R0
    SYMATTR InstName R2
    SYMATTR Value 1.8k
    SYMBOL res 592 32 R0
    SYMATTR InstName R11
    SYMATTR Value 10k
    SYMBOL voltage 608 160 R0
    SYMATTR InstName V3
    SYMATTR Value -5
    SYMBOL cap 704 -96 R0
    SYMATTR InstName C2
    SYMATTR Value 100n
    SYMBOL res 528 0 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R3
    SYMATTR Value 4k
    SYMBOL cap 112 -64 R0
    SYMATTR InstName C3
    SYMATTR Value 22p
    TEXT -32 384 Left 2 !.MODEL BFR181 NPN(IS=1.0519E-18 ISE=1.2603E-14 ISC=1.1195E-17 XTI=3 BF=96.461\n+ NR=0.87757 IKF=0.12146
    IKR=0.24951 XTB=1.5 VAF=22.403 VAR=5.1127 VJE=0.73155\n+ VJC=1.1633 RE=2.1372 RC=2.2171 RB=9.9037 RBM=6.6315 IRB=0.00069278
    CJE=1.8168f\n+ CJC=3.1969E-13 XCJC=0.082903 FC=0.99768 NF=0.90617 BR=16.504 NE=1.7631 NC=1.6528\n+ MJE=0.43619 MJC=0.30013
    TF=1.7028E-11 TR=2.7449n ITF=0.0010549 VTF=0.12571\n+ XTF=0.33814 EG=1.11 VCEO=20 ICRATING=20m MFG=SIEMENS)
    TEXT -80 232 Left 2 !.tran 1 startup
    TEXT -80 264 Left 2 !.options plotwinsize=0


    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Fri Feb 27 09:04:17 2026
    From Newsgroup: sci.electronics.design

    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman >>>><jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>>>>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>infineon models and not that simplistic mytotor model). Adding >>>>resistance or bead to Q2 base will stop it, but you're correct - why >>>>use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    The usual Spice models don't include package parasitics. Lead and
    wirebond inductances encourage oscillation.

    Many RF parts have their wirebonds specifically tuned to optimize
    operation in one RF band. LDMOS for example.

    Ancient TO-5 parts like 2N2219 had LONG wirebonds so a simple emitter
    follower would often oscillate. The fix is a 100 ohm series base
    resistor to kill the Q.



    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Fri Feb 27 18:08:46 2026
    From Newsgroup: sci.electronics.design

    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman >>>><jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better >>>>>>>> chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist:
    * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p

    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>infineon models and not that simplistic mytotor model). Adding >>>>resistance or bead to Q2 base will stop it, but you're correct - why >>>>use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    Version 4.1
    SHEET 1 6000 680
    WIRE 608 -416 368 -416
    WIRE 864 -416 608 -416
    WIRE 864 -368 864 -416
    WIRE 368 -336 368 -416
    WIRE 608 -256 608 -416
    WIRE 864 -256 864 -288
    WIRE 272 -224 256 -224
    WIRE 368 -224 368 -256
    WIRE 368 -224 336 -224
    WIRE 368 -208 368 -224
    WIRE 544 -208 368 -208
    WIRE 368 -160 368 -208
    WIRE 608 -128 608 -160
    WIRE 720 -128 608 -128
    WIRE 832 -128 720 -128
    WIRE -16 -112 -80 -112
    WIRE 128 -112 64 -112
    WIRE 208 -112 128 -112
    WIRE 256 -112 256 -224
    WIRE 256 -112 208 -112
    WIRE 304 -112 256 -112
    WIRE 608 -96 608 -128
    WIRE 720 -96 720 -128
    WIRE 128 -64 128 -112
    WIRE -80 -48 -80 -112
    WIRE 368 -32 368 -64
    WIRE 608 0 608 -16
    WIRE 720 0 720 -32
    WIRE 720 0 608 0
    WIRE 208 16 208 -112
    WIRE 432 16 208 16
    WIRE 608 16 608 0
    WIRE 608 16 512 16
    WIRE 128 32 128 0
    WIRE 608 48 608 16
    WIRE 608 176 608 128
    WIRE 608 304 608 256
    FLAG 368 -32 0
    FLAG 864 -256 0
    FLAG 608 304 0
    FLAG 128 32 0
    FLAG -80 -48 0
    SYMBOL npn 304 -160 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR181
    SYMBOL res 352 -352 R0
    SYMATTR InstName R9
    SYMATTR Value 5.6k
    SYMBOL cap 336 -240 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 0.1p
    SYMBOL res 80 -128 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R1
    SYMATTR Value 10k
    SYMBOL voltage 864 -384 R0
    SYMATTR InstName V2
    SYMATTR Value 5
    SYMBOL npn 544 -256 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR181
    SYMBOL res 592 -112 R0
    SYMATTR InstName R2
    SYMATTR Value 1.8k
    SYMBOL res 592 32 R0
    SYMATTR InstName R11
    SYMATTR Value 10k
    SYMBOL voltage 608 160 R0
    SYMATTR InstName V3
    SYMATTR Value -5
    SYMBOL cap 704 -96 R0
    SYMATTR InstName C2
    SYMATTR Value 100n
    SYMBOL res 528 0 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R3
    SYMATTR Value 4k
    SYMBOL cap 112 -64 R0
    SYMATTR InstName C3
    SYMATTR Value 22p
    TEXT -32 384 Left 2 !.MODEL BFR181 NPN(IS=1.0519E-18 ISE=1.2603E-14 ISC=1.1195E-17 XTI=3 BF=96.461\n+ NR=0.87757 IKF=0.12146
    IKR=0.24951 XTB=1.5 VAF=22.403 VAR=5.1127 VJE=0.73155\n+ VJC=1.1633 RE=2.1372 RC=2.2171 RB=9.9037 RBM=6.6315 IRB=0.00069278
    CJE=1.8168f\n+ CJC=3.1969E-13 XCJC=0.082903 FC=0.99768 NF=0.90617 BR=16.504 NE=1.7631 NC=1.6528\n+ MJE=0.43619 MJC=0.30013
    TF=1.7028E-11 TR=2.7449n ITF=0.0010549 VTF=0.12571\n+ XTF=0.33814 EG=1.11 VCEO=20 ICRATING=20m MFG=SIEMENS)
    TEXT -80 232 Left 2 !.tran 1 startup
    TEXT -80 264 Left 2 !.options plotwinsize=0



    You should really use the measured S-params and use MO/ADS etc.
    (simulated together wit the layout). But there are spice models in the
    files Infineon files bfr193-spar6ghz-noise1.8ghz-spice-6ghz and bfr181-spar6ghz-noise1.8ghz-spice-6ghz which include the package
    parasitics. Listed below to save you searching.

    Note that the 181 is a bit of a snowflake in comparison to the 193.
    For example as well as the stability the circuit would need to be
    checked for startup issues to ensure the vbe limit (only 8V) is not
    exceeded. Both of these transistors have the same ft, but apart from
    that...

    .SUBCKT BFR181 11 22 33

    CBEPAR 2 3 7.85E-014
    CBCPAR 2 1 4.949E-014
    CCEPAR 1 3 1.965E-013
    LE 33 30 6.492E-010
    LB 22 20 7.342E-010
    LC 1 11 6.908E-010
    CBEPCK 20 30 1.567E-014
    CBCPCK 20 1 1.975E-013
    CCEPCK 1 30 1.497E-013
    LBI 20 2 6.294E-010
    LEI 30 3 7.211E-010

    Q1 1 2 3 M_BFR181

    .MODEL M_BFR181 NPN( + IS = 9.57E-017 + BF = 131.1 +
    NF = 1 + VAF =
    71.02 + IKF = 0.09745 +
    ISE = 1.618E-015 + NE
    = 1.692 + BR =
    3.287 + NR = 0.959 +
    VAR = 4.081
    + IKR = 0.07617 +
    ISC = 5.969E-015 + NC
    = 1.974 + RB =
    9.72444 + IRB = 3.017E-006 +
    RBM = 6.94667
    + RE = 0.7979 +
    RC = 2.089
    + XTB = 1.303 +
    EG = 1.11 + XTI =
    6.548 + CJE = 8.287E-014 +
    VJE = 0.8281 + MJE
    = 0.7138 + TF =
    1.72653E-011 + XTF = 0.07 +
    VTF = 0.00381019 + ITF
    = 0.027024 + PTF =
    0.397851 + CJC = 8.781E-014 +
    VJC = 0.7715 + MJC
    = 0.7552 + XCJC =
    0.6209 + TR = 1.71536E-008
    + CJS = 0 + MJS
    = 0
    + VJS = 0.309
    + FC = 0.6275 +
    KF = 0 + AF =
    1)
    ***************************************************************

    .ENDS BFR181


    .SUBCKT BFR193 11 22 33

    CBEPAR 2 3 6.251E-013
    CBCPAR 2 1 2.028E-013
    CCEPAR 1 3 2.08E-013
    LEI 3 30 4.62381E-010
    LBI 2 20 8.57656E-010
    CBEPCK 20 30 1.567E-014
    CBCPCK 20 1 1.975E-013
    CCEPCK 1 30 1.497E-013
    LB 20 22 7.342E-010
    LE 30 33 6.492E-010
    LC 11 1 6.908E-010

    Q1 1 2 3 M_BFR193

    .MODEL M_BFR193 NPN( + IS = 8.17971E-016 + BF = 108.285 +
    NF = 1.003 + VAF
    = 36.5472 + IKF =
    0.273876 + ISE = 6.20271E-016
    + NE = 1.553 +
    BR = 42.76 + NR
    = 0.9775 + VAR =
    28.4
    + IKR = 0.01296 +
    ISC = 3.44E-015 + NC
    = 1.338 + RB =
    1.75232 + IRB = 0.0001443 +
    RBM = 0.967481
    + RE = 0.310454 +
    RC = 0.133617
    + XTB = 1.303 +
    EG = 1.11 + XTI =
    6.548 + CJE = 2.126E-012 +
    VJE = 1.052 + MJE
    = 0.47 + TF =
    1.58062E-011 + XTF = 0.0395 +
    VTF = 0.0696445 + ITF
    = 0.00207564 + PTF =
    0.0817673 + CJC = 7.142E-013 +
    VJC = 0.6315 + MJC
    = 0.3991 + XCJC =
    0.1208 + TR = 1.58463E-008
    + CJS = 0 + MJS
    = 0
    + VJS = 0.75
    + FC = 0.675369 +
    KF = 0 + AF =
    1)
    *************************************************************** .ENDS
    BFR193
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Bill Sloman@bill.sloman@ieee.org to sci.electronics.design on Sat Feb 28 05:29:24 2026
    From Newsgroup: sci.electronics.design

    On 28/02/2026 4:04 am, john larkin wrote:
    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist: >>>>>>>> * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p >>>>>>>>
    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>> infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why >>>>> use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    The usual Spice models don't include package parasitics.

    There no reason why you shouldn't add your own.

    Lead and wirebond inductances encourage oscillation.

    Many RF parts have their wirebonds specifically tuned to optimize
    operation in one RF band. LDMOS for example.

    Ancient TO-5 parts like 2N2219 had LONG wirebonds so a simple emitter follower would often oscillate. The fix is a 100 ohm series base
    resistor to kill the Q.

    Why chose 100R? Cambridge Instruments engineers took to sticking a 10R resistor in series with every DMOS gate, not because 10R was a magic
    value, but because it was lot easier to change the value of that
    resistor than it was to change the printed circuit layout if it hadn't
    got a space for a resistor.
    --
    Bill Sloman, Sydney
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From john larkin@jl@glen--canyon.com to sci.electronics.design on Fri Feb 27 12:05:23 2026
    From Newsgroup: sci.electronics.design

    On Sat, 28 Feb 2026 05:29:24 +1100, Bill Sloman <bill.sloman@ieee.org>
    wrote:

    On 28/02/2026 4:04 am, john larkin wrote:
    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist: >>>>>>>>> * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p >>>>>>>>>
    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>>> infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why >>>>>> use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still >>>>> a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the >>>> amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    The usual Spice models don't include package parasitics.

    There no reason why you shouldn't add your own.

    Sure, if you know what they are. Traces and vias matter too. A
    resistor costs 1 cent, but a week of engineering is expensive.


    Lead and wirebond inductances encourage oscillation.

    Many RF parts have their wirebonds specifically tuned to optimize
    operation in one RF band. LDMOS for example.

    Ancient TO-5 parts like 2N2219 had LONG wirebonds so a simple emitter
    follower would often oscillate. The fix is a 100 ohm series base
    resistor to kill the Q.

    Why chose 100R? Cambridge Instruments engineers took to sticking a 10R >resistor in series with every DMOS gate, not because 10R was a magic
    value, but because it was lot easier to change the value of that
    resistor than it was to change the printed circuit layout if it hadn't
    got a space for a resistor.

    100 usually works, or something like that if it's already on the BOM.
    We try to minimize the number of feeders on the P&P machine.

    If noise or voltage drop is an issue, something else needs to be done.
    A small bead maybe.

    There are transistors with built-in base resistors that could be
    dropped in, in a crisis. RET, digital transistor, something like
    that.






    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics
    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Fri Feb 27 16:05:51 2026
    From Newsgroup: sci.electronics.design

    "Bill Sloman" <bill.sloman@ieee.org> wrote in message news:10nsnqh$2o8ug$2@dont-email.me...
    On 28/02/2026 4:04 am, john larkin wrote:
    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman
    <jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist: >>>>>>>>> * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p >>>>>>>>>
    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass
    right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>>> infineon models and not that simplistic mytotor model). Adding
    resistance or bead to Q2 base will stop it, but you're correct - why >>>>>> use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still >>>>> a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the >>>> amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    The usual Spice models don't include package parasitics.

    There no reason why you shouldn't add your own.

    Lead and wirebond inductances encourage oscillation.

    Many RF parts have their wirebonds specifically tuned to optimize
    operation in one RF band. LDMOS for example.

    Ancient TO-5 parts like 2N2219 had LONG wirebonds so a simple emitter
    follower would often oscillate. The fix is a 100 ohm series base
    resistor to kill the Q.

    Why chose 100R? Cambridge Instruments engineers took to sticking a 10R resistor in series with every DMOS gate, not because 10R
    was a magic value, but because it was lot easier to change the value of that resistor than it was to change the printed circuit
    layout if it hadn't got a space for a resistor.

    I used to do that too with nearly anything which required a power FET.
    Put a few ohms in then I can change the value later if needed.


    --
    Bill Sloman, Sydney


    --- Synchronet 3.21b-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Fri Feb 27 23:26:05 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:grm3qk5las6n5id2qvpg1hhd99hdb1qiij@4ax.com...
    On Fri, 27 Feb 2026 11:19:44 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:5l72qk514j9bd6ggbkl6cdh4ua876i0dtv@4ax.com...
    On Thu, 26 Feb 2026 16:44:48 -0800, john larkin <jl@glen--canyon.com>
    wrote:

    On Thu, 26 Feb 2026 21:20:48 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Thu, 26 Feb 2026 21:54:14 +0100, Jeroen Belleman >>>>><jeroen@nospam.please> wrote:

    On 2/26/26 21:05, JM wrote:
    On Thu, 26 Feb 2026 20:50:50 +0100, Jean-Pierre Coulon
    <coulon@cacas.pam.obs-nice.fr> wrote:

    On Thu, 26 Feb 2026, Jeroen Belleman wrote:

    It's probably screaming at the top of its little voice at a few GHz >>>>>>>>> for a millisecond before giving up the ghost.

    Then I need a digital scope to record the process?

    Show us what you are doing, schematics and layout. We'll have a better
    chance at guessing what happened.

    Since I can't attach a drawing here, this is the SPICE netlist: >>>>>>>> * transimp simple
    vin 20 0 ac 1
    rcalib 20 1 10k
    cphd 1 0 22p
    q1 2 1 0 mytotor
    rtransimp 1 4 4k
    ccomp 2 1 .1p
    vp5 10 0 dc 5
    R9 10 2 5.6k
    Rab 3 4 1.8k
    Ca 3 4 100n
    q2 10 2 3 mytotor
    R11 4 11 10k
    vm5 11 0 dc -5
    .model mytotor npn rb=4 is=1e-14 bf=150 tf=20p cje=.5p cjc=.5p >>>>>>>>
    Bye,

    https://mega.nz/file/gxsERDzD#VMfGH9XRllX1gzkSb7aMcFWRcy8KTx5acn7rbONECsk

    Do you really need 8GHz transistors? There's a hefty RC lowpass >>>>>>right at the input.

    Jeroen Belleman

    It oscillates like buggery with a BFR193 nevermind a BFR181 (using the >>>>>infineon models and not that simplistic mytotor model). Adding >>>>>resistance or bead to Q2 base will stop it, but you're correct - why >>>>>use these in the first place?

    People hardly make slow transistors any more.

    My go-to gumdrop NPN is a BCX70J, which has huge beta, but it's still
    a couple hundred MHz. An emitter follower may well oscillate.

    I mostly use mosfets, like 2N7002 or FDV301.


    John Larkin
    Highland Tech Glen Canyon Design Center
    Lunatic Fringe Electronics

    Well nobody is going to manufacture something if no-one is going to
    buy it. The low frequency discrete transistor market has been
    shrinking for the past 30 years. We were spoiled for choice back in
    the day and never knew it.

    Looking closer at this chap's circuit I assume that cphd is actually
    the sensor, and rcalib is for some sort of system calibration. (It
    would help if people actually stated WTF they are trying to do,
    especially the sensor type they are interfacing too.)

    In that case modelling the input as a current source into Q1 base, the
    amp is stable with a BFR193 but not with a BFR181, which is what he
    observes.

    If you have time can you let me know why I can't see any oscillation with the model below.
    There may, of course, be many reasons. The BFR181 model (which I got from the first
    search engine hit on the subject) may be inadequate or perhaps R1/C3 needs changing
    as you said above.
    Thanks.

    Version 4.1
    SHEET 1 6000 680
    WIRE 608 -416 368 -416
    WIRE 864 -416 608 -416
    WIRE 864 -368 864 -416
    WIRE 368 -336 368 -416
    WIRE 608 -256 608 -416
    WIRE 864 -256 864 -288
    WIRE 272 -224 256 -224
    WIRE 368 -224 368 -256
    WIRE 368 -224 336 -224
    WIRE 368 -208 368 -224
    WIRE 544 -208 368 -208
    WIRE 368 -160 368 -208
    WIRE 608 -128 608 -160
    WIRE 720 -128 608 -128
    WIRE 832 -128 720 -128
    WIRE -16 -112 -80 -112
    WIRE 128 -112 64 -112
    WIRE 208 -112 128 -112
    WIRE 256 -112 256 -224
    WIRE 256 -112 208 -112
    WIRE 304 -112 256 -112
    WIRE 608 -96 608 -128
    WIRE 720 -96 720 -128
    WIRE 128 -64 128 -112
    WIRE -80 -48 -80 -112
    WIRE 368 -32 368 -64
    WIRE 608 0 608 -16
    WIRE 720 0 720 -32
    WIRE 720 0 608 0
    WIRE 208 16 208 -112
    WIRE 432 16 208 16
    WIRE 608 16 608 0
    WIRE 608 16 512 16
    WIRE 128 32 128 0
    WIRE 608 48 608 16
    WIRE 608 176 608 128
    WIRE 608 304 608 256
    FLAG 368 -32 0
    FLAG 864 -256 0
    FLAG 608 304 0
    FLAG 128 32 0
    FLAG -80 -48 0
    SYMBOL npn 304 -160 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR181
    SYMBOL res 352 -352 R0
    SYMATTR InstName R9
    SYMATTR Value 5.6k
    SYMBOL cap 336 -240 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 0.1p
    SYMBOL res 80 -128 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R1
    SYMATTR Value 10k
    SYMBOL voltage 864 -384 R0
    SYMATTR InstName V2
    SYMATTR Value 5
    SYMBOL npn 544 -256 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR181
    SYMBOL res 592 -112 R0
    SYMATTR InstName R2
    SYMATTR Value 1.8k
    SYMBOL res 592 32 R0
    SYMATTR InstName R11
    SYMATTR Value 10k
    SYMBOL voltage 608 160 R0
    SYMATTR InstName V3
    SYMATTR Value -5
    SYMBOL cap 704 -96 R0
    SYMATTR InstName C2
    SYMATTR Value 100n
    SYMBOL res 528 0 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R3
    SYMATTR Value 4k
    SYMBOL cap 112 -64 R0
    SYMATTR InstName C3
    SYMATTR Value 22p
    TEXT -32 384 Left 2 !.MODEL BFR181 NPN(IS=1.0519E-18 ISE=1.2603E-14 ISC=1.1195E-17 XTI=3 BF=96.461\n+ NR=0.87757 IKF=0.12146
    IKR=0.24951 XTB=1.5 VAF=22.403 VAR=5.1127 VJE=0.73155\n+ VJC=1.1633 RE=2.1372 RC=2.2171 RB=9.9037 RBM=6.6315 IRB=0.00069278
    CJE=1.8168f\n+ CJC=3.1969E-13 XCJC=0.082903 FC=0.99768 NF=0.90617 BR=16.504 NE=1.7631 NC=1.6528\n+ MJE=0.43619 MJC=0.30013
    TF=1.7028E-11 TR=2.7449n ITF=0.0010549 VTF=0.12571\n+ XTF=0.33814 EG=1.11 VCEO=20 ICRATING=20m MFG=SIEMENS)
    TEXT -80 232 Left 2 !.tran 1 startup
    TEXT -80 264 Left 2 !.options plotwinsize=0



    You should really use the measured S-params and use MO/ADS etc.
    (simulated together wit the layout). But there are spice models in the
    files Infineon files bfr193-spar6ghz-noise1.8ghz-spice-6ghz and bfr181-spar6ghz-noise1.8ghz-spice-6ghz which include the package
    parasitics. Listed below to save you searching.

    Note that the 181 is a bit of a snowflake in comparison to the 193.
    For example as well as the stability the circuit would need to be
    checked for startup issues to ensure the vbe limit (only 8V) is not
    exceeded. Both of these transistors have the same ft, but apart from
    that...

    .SUBCKT BFR181 11 22 33

    CBEPAR 2 3 7.85E-014
    CBCPAR 2 1 4.949E-014
    CCEPAR 1 3 1.965E-013
    LE 33 30 6.492E-010
    LB 22 20 7.342E-010
    LC 1 11 6.908E-010
    CBEPCK 20 30 1.567E-014
    CBCPCK 20 1 1.975E-013
    CCEPCK 1 30 1.497E-013
    LBI 20 2 6.294E-010
    LEI 30 3 7.211E-010

    Q1 1 2 3 M_BFR181

    .MODEL M_BFR181 NPN( + IS = 9.57E-017 + BF = 131.1 +
    NF = 1 + VAF =
    71.02 + IKF = 0.09745 +
    ISE = 1.618E-015 + NE
    = 1.692 + BR =
    3.287 + NR = 0.959 +
    VAR = 4.081
    + IKR = 0.07617 +
    ISC = 5.969E-015 + NC
    = 1.974 + RB =
    9.72444 + IRB = 3.017E-006 +
    RBM = 6.94667
    + RE = 0.7979 +
    RC = 2.089
    + XTB = 1.303 +
    EG = 1.11 + XTI =
    6.548 + CJE = 8.287E-014 +
    VJE = 0.8281 + MJE
    = 0.7138 + TF =
    1.72653E-011 + XTF = 0.07 +
    VTF = 0.00381019 + ITF
    = 0.027024 + PTF =
    0.397851 + CJC = 8.781E-014 +
    VJC = 0.7715 + MJC
    = 0.7552 + XCJC =
    0.6209 + TR = 1.71536E-008
    + CJS = 0 + MJS
    = 0
    + VJS = 0.309
    + FC = 0.6275 +
    KF = 0 + AF =
    1)
    ***************************************************************

    .ENDS BFR181


    .SUBCKT BFR193 11 22 33

    CBEPAR 2 3 6.251E-013
    CBCPAR 2 1 2.028E-013
    CCEPAR 1 3 2.08E-013
    LEI 3 30 4.62381E-010
    LBI 2 20 8.57656E-010
    CBEPCK 20 30 1.567E-014
    CBCPCK 20 1 1.975E-013
    CCEPCK 1 30 1.497E-013
    LB 20 22 7.342E-010
    LE 30 33 6.492E-010
    LC 11 1 6.908E-010

    Q1 1 2 3 M_BFR193

    .MODEL M_BFR193 NPN( + IS = 8.17971E-016 + BF = 108.285 +
    NF = 1.003 + VAF
    = 36.5472 + IKF =
    0.273876 + ISE = 6.20271E-016
    + NE = 1.553 +
    BR = 42.76 + NR
    = 0.9775 + VAR =
    28.4
    + IKR = 0.01296 +
    ISC = 3.44E-015 + NC
    = 1.338 + RB =
    1.75232 + IRB = 0.0001443 +
    RBM = 0.967481
    + RE = 0.310454 +
    RC = 0.133617
    + XTB = 1.303 +
    EG = 1.11 + XTI =
    6.548 + CJE = 2.126E-012 +
    VJE = 1.052 + MJE
    = 0.47 + TF =
    1.58062E-011 + XTF = 0.0395 +
    VTF = 0.0696445 + ITF
    = 0.00207564 + PTF =
    0.0817673 + CJC = 7.142E-013 +
    VJC = 0.6315 + MJC
    = 0.3991 + XCJC =
    0.1208 + TR = 1.58463E-008
    + CJS = 0 + MJS
    = 0
    + VJS = 0.75
    + FC = 0.675369 +
    KF = 0 + AF =
    1)
    *************************************************************** .ENDS
    BFR193

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    Version 4.1
    SHEET 1 6000 1140
    WIRE 1248 -800 368 -800
    WIRE 1600 -800 1248 -800
    WIRE 368 -752 368 -800
    WIRE 1600 -752 1600 -800
    WIRE -256 -640 -496 -640
    WIRE 368 -640 368 -672
    WIRE 368 -640 -192 -640
    WIRE 1600 -640 1600 -672
    WIRE 368 -608 368 -640
    WIRE 704 -608 368 -608
    WIRE 368 -576 368 -608
    WIRE 1248 -528 1248 -800
    WIRE 704 -464 704 -608
    WIRE 96 -448 32 -448
    WIRE 368 -448 368 -496
    WIRE 368 -448 160 -448
    WIRE 976 -400 912 -400
    WIRE 1248 -400 1248 -448
    WIRE 1248 -400 1040 -400
    WIRE 240 -384 192 -384
    WIRE 368 -384 368 -448
    WIRE 368 -384 304 -384
    WIRE 384 -384 368 -384
    WIRE 496 -384 384 -384
    WIRE 640 -384 496 -384
    WIRE 1120 -336 1072 -336
    WIRE 1248 -336 1248 -400
    WIRE 1248 -336 1184 -336
    WIRE 1264 -336 1248 -336
    WIRE 1376 -336 1264 -336
    WIRE 1520 -336 1376 -336
    WIRE 384 -304 384 -384
    WIRE 496 -288 496 -384
    WIRE -496 -272 -496 -640
    WIRE -112 -272 -496 -272
    WIRE 16 -272 -32 -272
    WIRE 32 -272 32 -448
    WIRE 32 -272 16 -272
    WIRE 80 -272 32 -272
    WIRE 192 -272 192 -384
    WIRE 192 -272 160 -272
    WIRE 192 -256 192 -272
    WIRE 320 -256 192 -256
    WIRE 1264 -256 1264 -336
    WIRE 1376 -240 1376 -336
    WIRE 704 -224 704 -384
    WIRE 768 -224 704 -224
    WIRE 896 -224 848 -224
    WIRE 912 -224 912 -400
    WIRE 912 -224 896 -224
    WIRE 960 -224 912 -224
    WIRE 1072 -224 1072 -336
    WIRE 1072 -224 1040 -224
    WIRE 1072 -208 1072 -224
    WIRE 1200 -208 1072 -208
    WIRE 192 -128 192 -256
    WIRE 240 -128 192 -128
    WIRE 368 -128 304 -128
    WIRE 384 -128 384 -208
    WIRE 384 -128 368 -128
    WIRE 496 -128 496 -224
    WIRE 496 -128 384 -128
    WIRE -768 -112 -832 -112
    WIRE -624 -112 -688 -112
    WIRE -544 -112 -624 -112
    WIRE -496 -112 -496 -272
    WIRE -496 -112 -544 -112
    WIRE 368 -96 368 -128
    WIRE 16 -80 16 -272
    WIRE 1072 -80 1072 -208
    WIRE 1120 -80 1072 -80
    WIRE 1248 -80 1184 -80
    WIRE 1264 -80 1264 -160
    WIRE 1264 -80 1248 -80
    WIRE 1376 -80 1376 -176
    WIRE 1376 -80 1264 -80
    WIRE -624 -64 -624 -112
    WIRE -832 -48 -832 -112
    WIRE 1248 -48 1248 -80
    WIRE 896 -32 896 -224
    WIRE 16 16 16 -16
    WIRE 368 16 368 -16
    WIRE 368 16 16 16
    WIRE -624 32 -624 0
    WIRE 368 32 368 16
    WIRE 512 32 368 32
    WIRE 640 32 640 -384
    WIRE 640 32 576 32
    WIRE 368 64 368 32
    WIRE 896 64 896 32
    WIRE 1248 64 1248 32
    WIRE 1248 64 896 64
    WIRE 1248 80 1248 64
    WIRE 1392 80 1248 80
    WIRE 1520 80 1520 -336
    WIRE 1520 80 1456 80
    WIRE 1248 112 1248 80
    WIRE 368 208 368 144
    WIRE 1248 272 1248 192
    WIRE 1360 272 1248 272
    WIRE 1600 272 1360 272
    WIRE 1632 272 1600 272
    WIRE 1248 304 1248 272
    WIRE 1360 304 1360 272
    WIRE 1248 400 1248 384
    WIRE 1360 400 1360 368
    WIRE 1360 400 1248 400
    WIRE -544 416 -544 -112
    WIRE 432 416 -544 416
    WIRE 1248 416 1248 400
    WIRE 1248 416 512 416
    WIRE 1248 448 1248 416
    WIRE 1248 576 1248 528
    WIRE 1248 704 1248 656
    FLAG 1600 -640 0
    FLAG 1248 704 0
    FLAG -624 32 0
    FLAG -832 -48 0
    FLAG 368 208 0
    FLAG 1600 272 osc
    SYMBOL res 352 -768 R0
    SYMATTR InstName R9
    SYMATTR Value 5.6k
    SYMBOL cap -192 -656 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 0.1p
    SYMBOL res -672 -128 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R1
    SYMATTR Value 10k
    SYMBOL voltage 1600 -768 R0
    SYMATTR InstName V2
    SYMATTR Value 5
    SYMBOL res 1232 288 R0
    SYMATTR InstName R2
    SYMATTR Value 1.8k
    SYMBOL res 1232 432 R0
    SYMATTR InstName R11
    SYMATTR Value 10k
    SYMBOL voltage 1248 560 R0
    SYMATTR InstName V3
    SYMATTR Value -5
    SYMBOL cap 1344 304 R0
    SYMATTR InstName C2
    SYMATTR Value 100n
    SYMBOL res 528 400 R90
    WINDOW 0 0 56 VBottom 2
    WINDOW 3 32 56 VTop 2
    SYMATTR InstName R3
    SYMATTR Value 4k
    SYMBOL cap -640 -64 R0
    SYMATTR InstName C3
    SYMATTR Value 22p
    SYMBOL npn 320 -304 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR181
    SYMBOL cap 304 -144 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBEPAR1
    SYMATTR Value 7.85E-014
    SYMBOL cap 304 -400 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBCPAR1
    SYMATTR Value 4.949E-014
    SYMBOL cap 480 -288 R0
    SYMATTR InstName CCEPAR1
    SYMATTR Value 1.965E-013
    SYMBOL ind 352 48 R0
    SYMATTR InstName LE1
    SYMATTR Value 6.492E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 352 -592 R0
    SYMATTR InstName LC1
    SYMATTR Value 6.908E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 64 -256 R270
    WINDOW 0 32 56 VTop 2
    WINDOW 3 5 56 VBottom 2
    SYMATTR InstName LBI1
    SYMATTR Value 6.294E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind -128 -256 R270
    WINDOW 0 32 56 VTop 2
    WINDOW 3 5 56 VBottom 2
    SYMATTR InstName LB1
    SYMATTR Value 7.342E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 352 -112 R0
    SYMATTR InstName LEI1
    SYMATTR Value 7.211E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL cap 0 -80 R0
    SYMATTR InstName CBEPCK1
    SYMATTR Value 1.567E-014
    SYMBOL cap 160 -464 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBCPCK1
    SYMATTR Value 1.975E-013
    SYMBOL cap 576 16 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CCEPCK1
    SYMATTR Value 1.497E-013
    SYMBOL npn 1200 -256 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR181
    SYMBOL cap 1184 -96 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBEPAR2
    SYMATTR Value 7.85E-014
    SYMBOL cap 1184 -352 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBCPAR2
    SYMATTR Value 4.949E-014
    SYMBOL cap 1360 -240 R0
    SYMATTR InstName CCEPAR2
    SYMATTR Value 1.965E-013
    SYMBOL ind 1232 96 R0
    SYMATTR InstName LE2
    SYMATTR Value 6.492E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 1232 -544 R0
    SYMATTR InstName LC2
    SYMATTR Value 6.908E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 944 -208 R270
    WINDOW 0 32 56 VTop 2
    WINDOW 3 5 56 VBottom 2
    SYMATTR InstName LBI2
    SYMATTR Value 6.294E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 752 -208 R270
    WINDOW 0 32 56 VTop 2
    WINDOW 3 5 56 VBottom 2
    SYMATTR InstName LB2
    SYMATTR Value 7.342E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL ind 1232 -64 R0
    SYMATTR InstName LEI2
    SYMATTR Value 7.211E-010
    SYMATTR SpiceLine Rser=1m
    SYMBOL cap 880 -32 R0
    SYMATTR InstName CBEPCK2
    SYMATTR Value 1.567E-014
    SYMBOL cap 1040 -416 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CBCPCK2
    SYMATTR Value 1.975E-013
    SYMBOL cap 1456 64 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName CCEPCK2
    SYMATTR Value 1.497E-013
    SYMBOL res 688 -480 R0
    SYMATTR InstName R4
    SYMATTR Value 0
    TEXT -504 296 Left 2 !.tran 0 0.0001 0 0.00000001 startup
    TEXT -504 328 Left 2 !.options plotwinsize=0
    TEXT -520 560 Left 2 !.MODEL BFR181 NPN(IS = 9.57E-017 BF = 131.1 NF = 1 VAF = 71.02 IKF = 0.09745\n+ ISE = 1.618E-015 NE = 1.692 BR
    = 3.287 NR = 0.959 VAR = 4.081 IKR = 0.07617\n+ ISC = 5.969E-015 NC = 1.974 RB = 9.72444 IRB = 3.017E-006 RBM = 6.94667 \n+
    RE = 0.7979 RC = 2.089 XTB = 1.303 EG = 1.11 XTI = 6.548 CJE = 8.287E-014\n+ VJE = 0.8281 MJE = 0.7138 TF = 1.72653E-011 XTF = 0.07
    VTF = 0.00381019\n+ ITF = 0.027024 PTF = 0.397851 CJC = 8.781E-014 VJC = 0.7715 MJC = 0.7552\n+ XCJC = 0.6209 TR = 1.71536E-008 CJS
    = 0 MJS = 0 VJS = 0.309 FC = 0.6275\n+ KF = 0 AF = 1)
    TEXT 208 -232 Left 2 ;2
    TEXT 400 -360 Left 2 ;1
    TEXT 400 -152 Left 2 ;3
    TEXT 376 176 Left 2 ;33
    TEXT 376 16 Left 2 ;30
    TEXT 0 -288 Left 2 ;20
    TEXT 712 -240 Left 2 ;22
    TEXT 1088 -184 Left 2 ;2
    TEXT 1280 -312 Left 2 ;1
    TEXT 1280 -104 Left 2 ;3
    TEXT 1256 224 Left 2 ;33
    TEXT 1256 64 Left 2 ;30
    TEXT 880 -240 Left 2 ;20
    TEXT 712 -240 Left 2 ;22
    TEXT 384 -640 Left 2 ;11
    TEXT 1264 -552 Left 2 ;11
    TEXT -168 -288 Left 2 ;22


    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Sat Feb 28 04:55:04 2026
    From Newsgroup: sci.electronics.design

    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your
    circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo

    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Sat Feb 28 00:09:17 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself

    Yeah I know but once it's done it's easy to duplicate.
    I'm not getting any response from the mega link, just a continuously spinning graphic.
    I'll try again in the morning.

    - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo



    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Sat Feb 28 09:51:16 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo


    I wasn't able to download any file from the link you provided.


    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Sat Feb 28 19:56:02 2026
    From Newsgroup: sci.electronics.design

    On Sat, 28 Feb 2026 09:51:16 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your
    circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo


    I wasn't able to download any file from the link you provided.



    https://mega.nz/file/Ao1xnSya#RGcDsQxLDNUyaZFKP1d3KW55N9RKRT0CsBFTsAA9VxY
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Sat Feb 28 19:58:16 2026
    From Newsgroup: sci.electronics.design

    On Sat, 28 Feb 2026 19:56:02 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Sat, 28 Feb 2026 09:51:16 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your
    circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo >>>

    I wasn't able to download any file from the link you provided.



    https://mega.nz/file/Ao1xnSya#RGcDsQxLDNUyaZFKP1d3KW55N9RKRT0CsBFTsAA9VxY

    oops, this should be it.

    https://mega.nz/file/11kgHRJY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo

    These links go away after a few hours.
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Sat Feb 28 19:58:59 2026
    From Newsgroup: sci.electronics.design

    On Sat, 28 Feb 2026 19:56:02 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Sat, 28 Feb 2026 09:51:16 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly.
    If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your
    circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo >>>

    I wasn't able to download any file from the link you provided.



    https://mega.nz/file/Ao1xnSya#RGcDsQxLDNUyaZFKP1d3KW55N9RKRT0CsBFTsAA9VxY

    oops, try this

    https://mega.nz/file/11kgHRJY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo


    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Sat Feb 28 20:25:18 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:v6i6qk9spm3dsdvri23815npe310h9rbg1@4ax.com...
    On Sat, 28 Feb 2026 19:56:02 +0000, JM
    <sunaecoNoChoppedPork@gmail.com> wrote:

    On Sat, 28 Feb 2026 09:51:16 -0500, "Edward Rawde" >><invalid@invalid.invalid> wrote:

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:uos4qkdjne2to0cq5e6hc7kd4nr9fbnv5f@4ax.com...
    On Fri, 27 Feb 2026 23:26:05 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    Thanks for that. Here is my first 6GHz oscillator.
    100 ohms for R4 kills oscillation but 10 ohms is not sufficient.
    The ferrite bead with the lowest L in the LTSpice library doesn't kill it but some of the others I tested do.

    You're making life hard for yourself - just change the transistor
    symbol type to a subcircuit (X) and use the subcircuit model directly. >>>> If you change the subcircuit names to numerics you can also step
    through the different transistors - see below where I've changed your
    circuit.

    (I'll upload to mega instead because the netlist will probably need
    manual editing otherwise.)

    https://mega.nz/file/05kVyDZY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo >>>>

    I wasn't able to download any file from the link you provided.



    https://mega.nz/file/Ao1xnSya#RGcDsQxLDNUyaZFKP1d3KW55N9RKRT0CsBFTsAA9VxY

    oops, try this

    https://mega.nz/file/11kgHRJY#an5rHSPesfthy0fJ9W8Hw9kyD8zmMgebGLbh0nZDBIo

    Thank you for the very educational file.
    I was going to ask how you get the attribute list to show up when right clicking a newly placed npn but
    Google just told me it's as simple as pressing Ctrl.

    A subckt to schematic converter would be nice but I can understand why that's hard.





    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From JM@sunaecoNoChoppedPork@gmail.com to sci.electronics.design on Sun Mar 1 22:36:38 2026
    From Newsgroup: sci.electronics.design

    On Sat, 28 Feb 2026 20:25:18 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    A subckt to schematic converter would be nice but I can understand why that's hard.

    There are netlist browsing tools available, but the free ones are
    pretty useless. This link is a view of the BFR193 subcircuit using a commercial tool, to show the type of schematic they generate.

    https://mega.nz/file/d5FkmKgA#Gf53z94th9dgzrnmLNbK8ttmE_Tzbhe65cjxC4EGTkE
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Edward Rawde@invalid@invalid.invalid to sci.electronics.design on Mon Mar 2 11:45:22 2026
    From Newsgroup: sci.electronics.design

    "JM" <sunaecoNoChoppedPork@gmail.com> wrote in message news:7if9qk9r40raqihcfb8n6ao8onbf7bok83@4ax.com...
    On Sat, 28 Feb 2026 20:25:18 -0500, "Edward Rawde"
    <invalid@invalid.invalid> wrote:

    A subckt to schematic converter would be nice but I can understand why that's hard.

    There are netlist browsing tools available, but the free ones are
    pretty useless. This link is a view of the BFR193 subcircuit using a commercial tool, to show the type of schematic they generate.

    https://mega.nz/file/d5FkmKgA#Gf53z94th9dgzrnmLNbK8ttmE_Tzbhe65cjxC4EGTkE

    Perhaps AI will help with that in the future. LCM? (Large Circuit Model).


    --- Synchronet 3.21d-Linux NewsLink 1.2