• Ye olde bootstrappe buffere

    From bitrex@21:1/5 to All on Mon Sep 9 13:28:32 2024
    Looks promising, with a midband gain of about 0.97.

    Use a PNP/JFET cascode as the slower input (as I guess sub-GHz PNPs is
    all we have, now) and fast NPNs on the output.

    It's definitely not expensive, anyway:

    <https://imgur.com/a/MTh2kl7>

    Version 4
    SHEET 1 5120 680
    WIRE 592 -1024 592 -1104
    WIRE 592 -1024 -240 -1024
    WIRE -240 -848 -240 -1024
    WIRE -240 -656 -240 -768
    WIRE 176 -656 -240 -656
    WIRE 800 -656 240 -656
    WIRE -240 -528 -240 -656
    WIRE 592 -416 592 -1024
    WIRE -240 -368 -240 -448
    WIRE 528 -368 -240 -368
    WIRE -240 -256 -240 -368
    WIRE 592 -192 592 -320
    WIRE 592 -192 -192 -192
    WIRE 800 -192 800 -656
    WIRE 800 -192 592 -192
    WIRE -240 0 -240 -160
    WIRE -576 48 -672 48
    WIRE -416 48 -512 48
    WIRE -304 48 -416 48
    WIRE 592 128 592 -192
    WIRE -240 176 -240 96
    WIRE 528 176 -240 176
    WIRE -832 240 -832 176
    WIRE -672 240 -672 48
    WIRE -416 288 -416 48
    WIRE -240 288 -240 176
    WIRE 592 304 592 224
    WIRE -832 464 -832 320
    WIRE -672 464 -672 320
    WIRE -416 464 -416 368
    WIRE -240 464 -240 368
    WIRE 592 464 592 384
    FLAG -240 464 0
    FLAG 592 464 0
    FLAG -416 464 0
    FLAG -672 464 0
    FLAG -832 464 0
    FLAG -832 176 Vcc
    FLAG 592 -1104 Vcc
    SYMBOL njf -192 -256 M0
    SYMATTR InstName J1
    SYMATTR Value 2N3819
    SYMBOL res -256 272 R0
    SYMATTR InstName R1
    SYMATTR Value 1k
    SYMBOL npn 528 -416 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR91A
    SYMBOL npn 528 128 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR91A
    SYMBOL res 576 288 R0
    SYMATTR InstName R2
    SYMATTR Value 470
    SYMBOL res -432 272 R0
    SYMATTR InstName R3
    SYMATTR Value 10k
    SYMBOL cap -512 32 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 100n
    SYMBOL voltage -672 224 R0
    WINDOW 123 24 124 Left 2
    WINDOW 39 0 0 Left 0
    SYMATTR Value2 AC 1
    SYMATTR InstName V1
    SYMATTR Value SINE(0 0.2 1Meg)
    SYMBOL voltage -832 224 R0
    WINDOW 123 0 0 Left 0
    WINDOW 39 0 0 Left 0
    SYMATTR InstName V2
    SYMATTR Value 9
    SYMBOL pnp -304 96 M180
    SYMATTR InstName Q3
    SYMATTR Value MPSH81
    SYMBOL res -256 -544 R0
    SYMATTR InstName R4
    SYMATTR Value 1k
    SYMBOL res -256 -864 R0
    SYMATTR InstName R5
    SYMATTR Value 1k
    SYMBOL cap 240 -672 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C2
    SYMATTR Value 100n
    TEXT -704 488 Left 2 !;tran 0.1
    TEXT -640 368 Left 2 !.ac oct 10 10k 1G
    TEXT 1208 -472 Left 2 !.MODEL BFR91A NPN\n+ IS = 1.32873E-015\n+ BF = 1.02000E+002\n+ NF = 1.00025E+000\n+ VAF = 5.19033E+001\n+ IKF = 8.15511E+000\n+ ISE = 1.39029E-014\n+ NE = 1.51292E+000\n+ BR = 1.76953E+001\n+ NR = 9.94038E-001\n+ VAR = 3.28032E+000\n+ IKR = 1.00000E+001\n+ ISC = 1.04297E-015\n+ NC = 1.18993E+000\n+ RB = 1.00000E+001\n+ IRB = 1.00000E-006\n+ RBM = 1.00000E+001\n+ RE = 7.63636E-001\n+ RC = 9.00000E+000\n+ EG = 1.11000E+000\n+ XTI = 3.00000E+000\n+ CJE = 2.03216E-012\n+ VJE = 6.00000E-001\n+ MJE = 2.90076E-001\n+ TF = 6.55790E-012\n+ XTF = 3.89752E+001\n+ VTF = 1.09308E+001\n+ ITF = 5.21078E-001\n+ CJC = 1.00353E-012\n+ VJC = 3.40808E-001\n+ MJC = 1.94223E-001\n.ENDS
    TEXT 1216 472 Left 2 !.model MPSH81 PNP(Is=10f Xti=3 Eg=1.11 Vaf=100
    Bf=133.8 Ise=1.678p Ne=2.159 Ikf=.1658 Nk=.901 Xtb=1.5 Var=100 Br=1
    Isc=9.519n Nc=3.88 Ikr=5.813 Rc=7.838 Cjc=2.81p Mjc=.1615 Vjc=.8282
    Fc=.5 Cje=2.695p Mje=.3214 Vje=.7026 Tr=11.32n Tf=97.83p Itf=69.29
    Xtf=599u Vtf=10 Vceo=20 Icrating=50m mfg=Fairchild)

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From piglet@21:1/5 to bitrex on Mon Sep 9 19:44:18 2024
    On 09/09/2024 6:28 pm, bitrex wrote:
    Looks promising, with a midband gain of about 0.97.

    Use a PNP/JFET cascode as the slower input (as I guess sub-GHz PNPs is
    all we have, now) and fast NPNs on the output.

    It's definitely not expensive, anyway:

    <https://imgur.com/a/MTh2kl7>

    Version 4
    SHEET 1 5120 680
    WIRE 592 -1024 592 -1104
    WIRE 592 -1024 -240 -1024
    WIRE -240 -848 -240 -1024
    WIRE -240 -656 -240 -768
    WIRE 176 -656 -240 -656
    WIRE 800 -656 240 -656
    WIRE -240 -528 -240 -656
    WIRE 592 -416 592 -1024
    WIRE -240 -368 -240 -448
    WIRE 528 -368 -240 -368
    WIRE -240 -256 -240 -368
    WIRE 592 -192 592 -320
    WIRE 592 -192 -192 -192
    WIRE 800 -192 800 -656
    WIRE 800 -192 592 -192
    WIRE -240 0 -240 -160
    WIRE -576 48 -672 48
    WIRE -416 48 -512 48
    WIRE -304 48 -416 48
    WIRE 592 128 592 -192
    WIRE -240 176 -240 96
    WIRE 528 176 -240 176
    WIRE -832 240 -832 176
    WIRE -672 240 -672 48
    WIRE -416 288 -416 48
    WIRE -240 288 -240 176
    WIRE 592 304 592 224
    WIRE -832 464 -832 320
    WIRE -672 464 -672 320
    WIRE -416 464 -416 368
    WIRE -240 464 -240 368
    WIRE 592 464 592 384
    FLAG -240 464 0
    FLAG 592 464 0
    FLAG -416 464 0
    FLAG -672 464 0
    FLAG -832 464 0
    FLAG -832 176 Vcc
    FLAG 592 -1104 Vcc
    SYMBOL njf -192 -256 M0
    SYMATTR InstName J1
    SYMATTR Value 2N3819
    SYMBOL res -256 272 R0
    SYMATTR InstName R1
    SYMATTR Value 1k
    SYMBOL npn 528 -416 R0
    SYMATTR InstName Q1
    SYMATTR Value BFR91A
    SYMBOL npn 528 128 R0
    SYMATTR InstName Q2
    SYMATTR Value BFR91A
    SYMBOL res 576 288 R0
    SYMATTR InstName R2
    SYMATTR Value 470
    SYMBOL res -432 272 R0
    SYMATTR InstName R3
    SYMATTR Value 10k
    SYMBOL cap -512 32 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C1
    SYMATTR Value 100n
    SYMBOL voltage -672 224 R0
    WINDOW 123 24 124 Left 2
    WINDOW 39 0 0 Left 0
    SYMATTR Value2 AC 1
    SYMATTR InstName V1
    SYMATTR Value SINE(0 0.2 1Meg)
    SYMBOL voltage -832 224 R0
    WINDOW 123 0 0 Left 0
    WINDOW 39 0 0 Left 0
    SYMATTR InstName V2
    SYMATTR Value 9
    SYMBOL pnp -304 96 M180
    SYMATTR InstName Q3
    SYMATTR Value MPSH81
    SYMBOL res -256 -544 R0
    SYMATTR InstName R4
    SYMATTR Value 1k
    SYMBOL res -256 -864 R0
    SYMATTR InstName R5
    SYMATTR Value 1k
    SYMBOL cap 240 -672 R90
    WINDOW 0 0 32 VBottom 2
    WINDOW 3 32 32 VTop 2
    SYMATTR InstName C2
    SYMATTR Value 100n
    TEXT -704 488 Left 2 !;tran 0.1
    TEXT -640 368 Left 2 !.ac oct 10 10k 1G
    TEXT 1208 -472 Left 2 !.MODEL  BFR91A   NPN\n+              IS =
    1.32873E-015\n+              BF = 1.02000E+002\n+              NF =
    1.00025E+000\n+             VAF = 5.19033E+001\n+             IKF =
    8.15511E+000\n+             ISE = 1.39029E-014\n+              NE =
    1.51292E+000\n+              BR = 1.76953E+001\n+              NR =
    9.94038E-001\n+             VAR = 3.28032E+000\n+             IKR =
    1.00000E+001\n+             ISC = 1.04297E-015\n+              NC =
    1.18993E+000\n+              RB = 1.00000E+001\n+             IRB =
    1.00000E-006\n+             RBM = 1.00000E+001\n+              RE =
    7.63636E-001\n+              RC = 9.00000E+000\n+              EG =
    1.11000E+000\n+             XTI = 3.00000E+000\n+             CJE =
    2.03216E-012\n+             VJE = 6.00000E-001\n+             MJE =
    2.90076E-001\n+              TF = 6.55790E-012\n+             XTF =
    3.89752E+001\n+             VTF = 1.09308E+001\n+             ITF =
    5.21078E-001\n+             CJC = 1.00353E-012\n+             VJC =
    3.40808E-001\n+             MJC = 1.94223E-001\n.ENDS
    TEXT 1216 472 Left 2 !.model MPSH81 PNP(Is=10f Xti=3 Eg=1.11 Vaf=100
    Bf=133.8 Ise=1.678p Ne=2.159 Ikf=.1658 Nk=.901 Xtb=1.5 Var=100 Br=1 Isc=9.519n Nc=3.88 Ikr=5.813 Rc=7.838 Cjc=2.81p Mjc=.1615 Vjc=.8282
    Fc=.5 Cje=2.695p Mje=.3214 Vje=.7026 Tr=11.32n Tf=97.83p Itf=69.29
    Xtf=599u Vtf=10 Vceo=20 Icrating=50m mfg=Fairchild)

    What does the JFET get you that an NPN BJT in that position wouldn't do?

    piglet

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From john larkin@21:1/5 to bitrex on Mon Sep 9 12:34:48 2024
    On Mon, 9 Sep 2024 13:28:32 -0400, bitrex <user@example.net> wrote:

    Looks promising, with a midband gain of about 0.97.

    Use a PNP/JFET cascode as the slower input (as I guess sub-GHz PNPs is
    all we have, now) and fast NPNs on the output.

    It's definitely not expensive, anyway:

    <https://imgur.com/a/MTh2kl7>


    The p-to-n emitter-to-emitter connection, a variant on the cascode,
    is intersting, as is the hybrid nfet/pnp source-to-emitter connection.

    Your input could have been into the jfet, for hi-z.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From bitrex@21:1/5 to piglet on Mon Sep 9 16:32:13 2024
    On 9/9/2024 2:44 PM, piglet wrote:
    On 09/09/2024 6:28 pm, bitrex wrote:
    Looks promising, with a midband gain of about 0.97.

    Use a PNP/JFET cascode as the slower input (as I guess sub-GHz PNPs is
    all we have, now) and fast NPNs on the output.

    It's definitely not expensive, anyway:

    <https://imgur.com/a/MTh2kl7>


    What does the JFET get you that an NPN BJT in that position wouldn't do?

    piglet


    An NPN there seems to cause some appreciable peaking around the HF
    cutoff, though I haven't dug into this topology far enough to understand exactly why..

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)