Inspired by the late Jim Thompson's autobias design that we saw in the thread "Power Amplifier for 100kHz" I thought his use of a bang-bang comparator in setting bias was non-optimal but I liked his idea of running the pa in pure class B for large signals and setting a minimum quiescent class A current for small signals. My idea uses JT's way of sensing output device current falling below threshold but instead of pumping up a "rubber Vbe diode" I pull the bias point apart with resistors and a pair of current mirrors. As an IC designer JT was liberal with current mirrors so I hope he would approve. I expect over the decades many clever ways have been devised at eliminating crossover distortion and autobiasing class A/B so I have no shame in adding another. Does this look interesting or has it been done already? Here is a pdf schematic: <https://www.dropbox.com/scl/fi/x1tw0nj4d44pgmutxw36p/classAB_rubber_mirrors_autobias.pdf?rlkey=0xr3fv4l2qswyhrombqtanwvo&raw=1> Here is the LT Spice wirelist: Version 4 SHEET 1 2520 916 WIRE 720 -768 720 -816 WIRE 528 -752 528 -816 WIRE 1088 -752 1088 -800 WIRE -512 -704 -512 -800 WIRE 528 -640 528 -672 WIRE 720 -640 720 -688 WIRE 1088 -640 1088 -672 WIRE 640 -592 592 -592 WIRE 896 -592 784 -592 WIRE 1024 -592 896 -592 WIRE 1888 -576 1888 -624 WIRE 1888 -576 1680 -576 WIRE -512 -560 -512 -624 WIRE -464 -560 -512 -560 WIRE 1680 -528 1680 -576 WIRE -464 -512 -464 -560 WIRE 1616 -480 1536 -480 WIRE 640 -464 640 -592 WIRE 896 -464 896 -592 WIRE 896 -464 640 -464 WIRE 1088 -464 1088 -544 WIRE 1088 -464 896 -464 WIRE -512 -432 -512 -560 WIRE 1888 -432 1888 -576 WIRE 1088 -416 1088 -464 WIRE 1264 -416 1264 -544 WIRE 1680 -384 1680 -432 WIRE 1824 -384 1680 -384 WIRE 1088 -320 1088 -336 WIRE 1264 -320 1264 -336 WIRE -512 -272 -512 -352 WIRE 1024 -272 928 -272 WIRE 1888 -272 1888 -336 WIRE 1888 -272 1328 -272 WIRE 1888 -240 1888 -272 WIRE 1168 -224 1088 -224 WIRE 1264 -224 1168 -224 WIRE 928 -192 928 -272 WIRE 1168 -176 1168 -224 WIRE 1680 -112 1680 -384 WIRE 1888 -112 1888 -160 WIRE 2080 -112 1888 -112 WIRE 2208 -112 2080 -112 WIRE 1888 -64 1888 -112 WIRE 1168 -32 1168 -96 WIRE 928 64 928 -112 WIRE 1888 64 1888 16 WIRE 1888 64 928 64 WIRE 720 128 720 -544 WIRE 864 128 720 128 WIRE 1536 128 1536 -480 WIRE 1536 128 864 128 WIRE 720 160 720 128 WIRE -112 240 -112 176 WIRE -496 256 -624 256 WIRE -304 256 -416 256 WIRE -144 256 -304 256 WIRE 864 256 864 128 WIRE 224 272 -80 272 WIRE 720 272 720 240 WIRE 720 272 224 272 WIRE 2208 272 2208 -112 WIRE -144 288 -224 288 WIRE 720 304 720 272 WIRE 1888 304 1888 64 WIRE 1680 352 1680 -32 WIRE 1824 352 1680 352 WIRE -224 368 -224 288 WIRE -112 384 -112 304 WIRE 1680 384 1680 352 WIRE -624 432 -624 256 WIRE 528 432 528 -544 WIRE 624 432 528 432 WIRE 720 432 720 384 WIRE 864 432 864 320 WIRE 864 432 720 432 WIRE 1616 432 864 432 WIRE 2208 464 2208 352 WIRE 528 480 528 432 WIRE 720 480 720 432 WIRE 1680 512 1680 480 WIRE 1888 512 1888 400 WIRE 1888 512 1680 512 WIRE 624 528 624 432 WIRE 624 528 592 528 WIRE 656 528 624 528 WIRE -304 560 -304 256 WIRE -48 560 -304 560 WIRE 224 560 224 272 WIRE 224 560 16 560 WIRE 528 608 528 576 WIRE 720 608 720 576 WIRE -624 656 -624 512 WIRE 1888 704 1888 512 WIRE 528 736 528 688 WIRE 720 736 720 688 WIRE -304 800 -304 560 WIRE -64 800 -304 800 WIRE 2080 800 2080 -112 WIRE 2080 800 16 800 FLAG -512 -800 P15 FLAG -112 176 P15 FLAG 1888 -624 P15 FLAG 1088 -800 P15 FLAG -512 -272 N15 FLAG -112 384 N15 FLAG 1888 704 N15 FLAG 1168 -32 N15 FLAG 2208 464 0 FLAG -624 656 0 FLAG -224 368 0 FLAG -464 -512 0 FLAG 720 736 N15 FLAG 528 736 N15 FLAG 720 -816 P15 FLAG 528 -816 P15 FLAG 1264 -544 P15 SYMBOL voltage -512 -720 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 15 SYMBOL voltage -512 -448 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value 15 SYMBOL voltage -624 416 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V3 SYMATTR Value SINE(0 1 100k) SYMBOL voltage 928 -208 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V4 SYMATTR Value 0.05 SYMBOL res 2192 256 R0 SYMATTR InstName R1 SYMATTR Value 8 SYMBOL pnp 1824 400 M180 SYMATTR InstName Q1 SYMATTR Value 2SAR533P SYMBOL pnp 1616 480 M180 SYMATTR InstName Q2 SYMATTR Value 2N2907 SYMBOL pnp 784 -544 R180 SYMATTR InstName Q10 SYMATTR Value 2N3906 SYMBOL npn 1824 -432 R0 SYMATTR InstName Q3 SYMATTR Value 2SCR533P SYMBOL npn 1616 -528 R0 SYMATTR InstName Q4 SYMATTR Value 2N2222 SYMBOL npn 1024 -320 R0 SYMATTR InstName Q5 SYMATTR Value 2N2222 SYMBOL npn 1328 -320 M0 SYMATTR InstName Q6 SYMATTR Value 2N2222 SYMBOL OpAmps\\UniversalOpAmp2 -112 272 R0 SYMATTR InstName U1 SYMBOL res 1872 -256 R0 SYMATTR InstName R2 SYMATTR Value 0.22 SYMBOL res 1872 -80 R0 SYMATTR InstName R3 SYMATTR Value 0.22 SYMBOL res 1696 -16 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R4 SYMATTR Value 220 SYMBOL current 1168 -176 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName I1 SYMATTR Value 10m SYMBOL res 32 784 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R6 SYMATTR Value 10k SYMBOL res -400 240 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R5 SYMATTR Value 1k SYMBOL cap -48 544 M90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 5p SYMBOL npn 656 480 R0 SYMATTR InstName Q13 SYMATTR Value 2N3904 SYMBOL npn 592 480 M0 SYMATTR InstName Q12 SYMATTR Value 2N3904 SYMBOL pnp 1024 -544 M180 SYMATTR InstName Q9 SYMATTR Value 2N3906 SYMBOL pnp 592 -544 R180 SYMATTR InstName Q11 SYMATTR Value 2N3906 SYMBOL res 1072 -768 R0 SYMATTR InstName R9 SYMATTR Value 22 SYMBOL res 704 -784 R0 SYMATTR InstName R10 SYMATTR Value 22 SYMBOL res 512 -768 R0 SYMATTR InstName R11 SYMATTR Value 22 SYMBOL res 704 592 R0 SYMATTR InstName R13 SYMATTR Value 22 SYMBOL res 512 592 R0 SYMATTR InstName R12 SYMATTR Value 22 SYMBOL res 704 144 R0 SYMATTR InstName R7 SYMATTR Value 220 SYMBOL res 704 288 R0 SYMATTR InstName R8 SYMATTR Value 220 SYMBOL res 1072 -432 R0 SYMATTR InstName R14 SYMATTR Value 1m SYMBOL res 1248 -432 R0 SYMATTR InstName R15 SYMATTR Value 1m SYMBOL cap 848 256 R0 SYMATTR InstName C2 SYMATTR Value 100p TEXT -656 848 Left 2 !.tran 1m TEXT 2128 -728 Left 2 ;EPW SED DEC 2023 TEXT 2128 -672 Left 2 ;CLASS AB AMPLIFIER TEXT 2128 -616 Left 2 ;CURRENT MIRROR ELASTIC BIAS piglet
Yet another Class AB autobias
Started by ●December 14, 2023
Reply by ●December 14, 20232023-12-14
On Thu, 14 Dec 2023 22:40:23 +0000, piglet <erichpwagner@hotmail.com> wrote:>Inspired by the late Jim Thompson's autobias design that we saw in the >thread "Power Amplifier for 100kHz" I thought his use of a bang-bang >comparator in setting bias was non-optimal but I liked his idea of >running the pa in pure class B for large signals and setting a minimum >quiescent class A current for small signals. > >My idea uses JT's way of sensing output device current falling below >threshold but instead of pumping up a "rubber Vbe diode" I pull the bias >point apart with resistors and a pair of current mirrors. As an IC >designer JT was liberal with current mirrors so I hope he would approve. > >I expect over the decades many clever ways have been devised at >eliminating crossover distortion and autobiasing class A/B so I have no >shame in adding another. Does this look interesting or has it been done >already? > >Here is a pdf schematic: > ><https://www.dropbox.com/scl/fi/x1tw0nj4d44pgmutxw36p/classAB_rubber_mirrors_autobias.pdf?rlkey=0xr3fv4l2qswyhrombqtanwvo&raw=1>Most of the audio amp circuits posted here seem to have been designed before the invention of the opamp.
Reply by ●December 14, 20232023-12-14
On 2023-12-14 18:05, John Larkin wrote:> On Thu, 14 Dec 2023 22:40:23 +0000, piglet <erichpwagner@hotmail.com> > wrote: > >> Inspired by the late Jim Thompson's autobias design that we saw in the >> thread "Power Amplifier for 100kHz" I thought his use of a bang-bang >> comparator in setting bias was non-optimal but I liked his idea of >> running the pa in pure class B for large signals and setting a minimum >> quiescent class A current for small signals. >> >> My idea uses JT's way of sensing output device current falling below >> threshold but instead of pumping up a "rubber Vbe diode" I pull the bias >> point apart with resistors and a pair of current mirrors. As an IC >> designer JT was liberal with current mirrors so I hope he would approve. >> >> I expect over the decades many clever ways have been devised at >> eliminating crossover distortion and autobiasing class A/B so I have no >> shame in adding another. Does this look interesting or has it been done >> already? >> >> Here is a pdf schematic: >> >> <https://www.dropbox.com/scl/fi/x1tw0nj4d44pgmutxw36p/classAB_rubber_mirrors_autobias.pdf?rlkey=0xr3fv4l2qswyhrombqtanwvo&raw=1> > > > Most of the audio amp circuits posted here seem to have been designed > before the invention of the opamp. >JT's amp actually holds its bias very nicely in large signal operation. There's enough gain in the loop that the LM311 only needs to fire occasionally near a zero-crossing for it to work fine. (Don't have time to look at your amp very closely just now, Erich--maybe over the weekend.) Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC / Hobbs ElectroOptics Optics, Electro-optics, Photonics, Analog Electronics Briarcliff Manor NY 10510 http://electrooptical.net http://hobbs-eo.com
Reply by ●December 14, 20232023-12-14
piglet <erichpwagner@hotmail.com> wrote:> I expect over the decades many clever ways have been devised at > eliminating crossover distortion and autobiasing class A/B so I have no > shame in adding another. Does this look interesting or has it been done > already?Your circuit has quite visible distortion throughout the range, including at crossover. Here is a version that is simpler and has no visible distortion throughout the range. The transistors are off at zero voltage output, so there is no need to fool with bias. I believe the original was posted by Larkin, but this was a long time ago so I might be mistaken. The op amp needs to have 100MHz bandwidth and slew rate, be able to supply 50 ma output, and handle +/- 15V supplies. I tried searching Octopart, but had no luck. The output current could be handled with Darlingtons, but the supply voltage for 10V output might be more difficult. You should develop the habit of naming every node. This helps troubleshooting immensely. Here is my circuit. Please be sure to download the .PLT file and look at the currents and voltages. Line wrap should not be a problem with this circuit. Version 4 SHEET 1 1140 1108 WIRE -784 -544 -896 -544 WIRE -1008 -480 -1024 -480 WIRE -928 -464 -944 -464 WIRE -896 -464 -896 -544 WIRE -896 -464 -928 -464 WIRE -848 -464 -896 -464 WIRE -720 -464 -720 -496 WIRE -720 -464 -768 -464 WIRE -640 -464 -720 -464 WIRE -624 -464 -640 -464 WIRE -608 -464 -624 -464 WIRE -1056 -448 -1200 -448 WIRE -1008 -448 -1056 -448 WIRE -608 -448 -608 -464 WIRE -1200 -432 -1200 -448 WIRE -720 -432 -720 -464 WIRE -896 -384 -896 -464 WIRE -784 -384 -896 -384 WIRE -1200 -336 -1200 -352 WIRE -1024 -272 -1024 -480 WIRE -640 -272 -640 -464 WIRE -640 -272 -1024 -272 FLAG -1200 -528 0 FLAG -1056 -448 U1P FLAG -928 -464 U1O FLAG -1200 -336 0 FLAG -1072 -528 0 FLAG -1072 -608 VEE FLAG -976 -432 VEE FLAG -720 -336 VEE FLAG -1200 -608 VCC FLAG -976 -496 VCC FLAG -720 -592 VCC FLAG -608 -368 0 FLAG -624 -464 Vout SYMBOL voltage -1200 -624 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 15 SYMBOL voltage -1200 -448 R0 WINDOW 0 49 39 VRight 2 WINDOW 3 64 142 Right 2 SYMATTR InstName V2 SYMATTR Value SINE(0 10 1e5) SYMATTR Value2 AC 1 SYMATTR SpiceLine Rser=2 SYMBOL opamps\\1pole -976 -464 R0 SYMATTR InstName U1 SYMATTR Value2 Avol=1Meg GBW=100Meg Slew=100Meg SYMATTR SpiceLine ilimit=50m rail=0 Vos=0 SYMBOL voltage -1072 -512 R180 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V3 SYMATTR Value 15 SYMBOL npn -784 -592 R0 SYMATTR InstName Q1 SYMATTR Value 2SCR533P SYMBOL pnp -784 -336 M180 WINDOW 0 63 34 Left 2 WINDOW 3 61 66 Left 2 SYMATTR InstName Q2 SYMATTR Value 2SAR553P SYMBOL res -624 -464 R0 SYMATTR InstName R2 SYMATTR Value 8 SYMBOL res -864 -448 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R3 SYMATTR Value 100 TEXT -1048 -752 Left 2 ;'Class B Amplifer Output TEXT -976 -720 Left 2 !.tran 0 20u 0 [Transient Analysis] { Npanes: 4 Active Pane: 2 { traces: 2 {589828,0,"V(vout)"} {524291,0,"V(u1p)"} X: ('�',0,0,2e-006,2e-005) Y[0]: (' ',0,-10,2,10) Y[1]: ('m',0,1e+308,0.004,-1e+308) Volts: (' ',0,0,0,-10,2,10) Log: 0 0 0 GridStyle: 1 PltMag: 1 PltPhi: 1 0 }, { traces: 2 {524290,0,"V(u1o)"} {34603013,1,"I(R3)"} X: ('�',0,0,2e-006,2e-005) Y[0]: (' ',0,-12,2,12) Y[1]: ('m',0,-0.012,0.002,0.012) Volts: (' ',0,0,0,-12,2,12) Amps: ('m',0,0,0,-0.012,0.002,0.012) Log: 0 0 0 GridStyle: 1 PltMag: 1 PltPhi: 1 0 }, { traces: 1 {34603015,0,"Ib(Q2)"} X: ('�',0,0,2e-006,2e-005) Y[0]: ('m',0,-0.014,0.002,0.01) Y[1]: ('m',0,1e+308,0.002,-1e+308) Amps: ('m',0,0,0,-0.014,0.002,0.01) Log: 0 0 0 GridStyle: 1 PltMag: 1 PltPhi: 1 0 }, { traces: 1 {34603014,0,"Ib(Q1)"} X: ('�',0,0,2e-006,2e-005) Y[0]: ('m',0,-0.01,0.002,0.016) Y[1]: ('m',0,1e+308,0.002,-1e+308) Amps: ('m',0,0,0,-0.01,0.002,0.016) Log: 0 0 0 GridStyle: 1 PltMag: 1 PltPhi: 1 0 } } -- MRM
Reply by ●December 14, 20232023-12-14
On Thursday, December 14, 2023 at 6:05:59 PM UTC-5, John Larkin wrote:> On Thu, 14 Dec 2023 22:40:23 +0000, piglet <erichp...@hotmail.com> > wrote: > >Inspired by the late Jim Thompson's autobias design that we saw in the > >thread "Power Amplifier for 100kHz" I thought his use of a bang-bang > >comparator in setting bias was non-optimal but I liked his idea of > >running the pa in pure class B for large signals and setting a minimum > >quiescent class A current for small signals. > > > >My idea uses JT's way of sensing output device current falling below > >threshold but instead of pumping up a "rubber Vbe diode" I pull the bias > >point apart with resistors and a pair of current mirrors. As an IC > >designer JT was liberal with current mirrors so I hope he would approve. > > > >I expect over the decades many clever ways have been devised at > >eliminating crossover distortion and autobiasing class A/B so I have no > >shame in adding another. Does this look interesting or has it been done > >already? > > > >Here is a pdf schematic: > > > ><https://www.dropbox.com/scl/fi/x1tw0nj4d44pgmutxw36p/classAB_rubber_mirrors_autobias.pdf?rlkey=0xr3fv4l2qswyhrombqtanwvo&raw=1> > Most of the audio amp circuits posted here seem to have been designed > before the invention of the opamp.They're a step up from tubes...
Reply by ●December 14, 20232023-12-14
On Fri, 15 Dec 2023 00:45:53 -0000 (UTC), Mike Monett VE3BTI <spamme@not.com> wrote:>piglet <erichpwagner@hotmail.com> wrote: > >> I expect over the decades many clever ways have been devised at >> eliminating crossover distortion and autobiasing class A/B so I have no >> shame in adding another. Does this look interesting or has it been done >> already? > >Your circuit has quite visible distortion throughout the range, including >at crossover. > >Here is a version that is simpler and has no visible distortion throughout >the range. The transistors are off at zero voltage output, so there is no >need to fool with bias. > >I believe the original was posted by Larkin, but this was a long time ago >so I might be mistaken.Probably me. I invented that way back in my youth. It doesn't work as well with mosfets, because it loses the gate turnon voltage, but that's easily fixed. The other fun topology uses the opamp power pins as outputs, to drive booster fets. That one can swing to the rails. The power pins can be used to make an ideal polarity splitter.
Reply by ●December 14, 20232023-12-14
On Thu, 14 Dec 2023 19:25:00 -0800, John Larkin <jl@997PotHill.com> wrote:>On Fri, 15 Dec 2023 00:45:53 -0000 (UTC), Mike Monett VE3BTI ><spamme@not.com> wrote: > >>piglet <erichpwagner@hotmail.com> wrote: >> >>> I expect over the decades many clever ways have been devised at >>> eliminating crossover distortion and autobiasing class A/B so I have no >>> shame in adding another. Does this look interesting or has it been done >>> already? >> >>Your circuit has quite visible distortion throughout the range, including >>at crossover. >> >>Here is a version that is simpler and has no visible distortion throughout >>the range. The transistors are off at zero voltage output, so there is no >>need to fool with bias. >> >>I believe the original was posted by Larkin, but this was a long time ago >>so I might be mistaken. > >Probably me. I invented that way back in my youth. > >It doesn't work as well with mosfets, because it loses the gate turnon >voltage, but that's easily fixed. > >The other fun topology uses the opamp power pins as outputs, to drive >booster fets. That one can swing to the rails. The power pins can be >used to make an ideal polarity splitter.Here's another one. https://www.dropbox.com/scl/fi/lduajoo7dyxy6h97tipes/Another_AB_amp.jpg?rlkey=lcyq6dsw4wpl64vr15gi0cdj2&raw=1 The transfer function is S-shaped but smooth. Just arrange for a few tenths of a volt across each diode with no signal.
Reply by ●December 15, 20232023-12-15
On 2023-12-15 04:47, John Larkin wrote:> On Thu, 14 Dec 2023 19:25:00 -0800, John Larkin <jl@997PotHill.com> > wrote: > >> On Fri, 15 Dec 2023 00:45:53 -0000 (UTC), Mike Monett VE3BTI >> <spamme@not.com> wrote: >> >>> piglet <erichpwagner@hotmail.com> wrote: >>> >>>> I expect over the decades many clever ways have been devised at >>>> eliminating crossover distortion and autobiasing class A/B so I have no >>>> shame in adding another. Does this look interesting or has it been done >>>> already? >>> >>> Your circuit has quite visible distortion throughout the range, including >>> at crossover. >>> >>> Here is a version that is simpler and has no visible distortion throughout >>> the range. The transistors are off at zero voltage output, so there is no >>> need to fool with bias. >>> >>> I believe the original was posted by Larkin, but this was a long time ago >>> so I might be mistaken. >> >> Probably me. I invented that way back in my youth. >> >> It doesn't work as well with mosfets, because it loses the gate turnon >> voltage, but that's easily fixed. >> >> The other fun topology uses the opamp power pins as outputs, to drive >> booster fets. That one can swing to the rails. The power pins can be >> used to make an ideal polarity splitter. > > Here's another one. > > https://www.dropbox.com/scl/fi/lduajoo7dyxy6h97tipes/Another_AB_amp.jpg?rlkey=lcyq6dsw4wpl64vr15gi0cdj2&raw=1 > > The transfer function is S-shaped but smooth. Just arrange for a few > tenths of a volt across each diode with no signal. >Where those diodes could also be emitter followers for more output current... Nice! Arie
Reply by ●December 15, 20232023-12-15
On Friday 15 December 2023 at 03:25:57 UTC, John Larkin wrote:> On Fri, 15 Dec 2023 00:45:53 -0000 (UTC), Mike Monett VE3BTI > <spa...@not.com> wrote: > > >piglet <erichp...@hotmail.com> wrote: > > > >> I expect over the decades many clever ways have been devised at > >> eliminating crossover distortion and autobiasing class A/B so I have no > >> shame in adding another. Does this look interesting or has it been done > >> already? > > > >Your circuit has quite visible distortion throughout the range, including > >at crossover. > > > >Here is a version that is simpler and has no visible distortion throughout > >the range. The transistors are off at zero voltage output, so there is no > >need to fool with bias. > > > >I believe the original was posted by Larkin, but this was a long time ago > >so I might be mistaken. > Probably me. I invented that way back in my youth. > > It doesn't work as well with mosfets, because it loses the gate turnon > voltage, but that's easily fixed. > > The other fun topology uses the opamp power pins as outputs, to drive > booster fets. That one can swing to the rails. The power pins can be > used to make an ideal polarity splitter.That topology was used by Burr Brown a very long time ago for some of their linear opto-isolators. There was an LED in each supply rail and a resistor from the output to ground. Feedback from a photodiode to the input and the other matched photodiode to the output amplifier. There was no hint in the datasheet that it was done this way. I opened one up. John
Reply by ●December 15, 20232023-12-15
Mike Monett VE3BTI <spamme@not.com> wrote:> The op amp needs to have 100MHz bandwidth and slew rate, be able to supply > 50 ma output, and handle +/- 15V supplies. I tried searching Octopart, but > had no luck.I finally figured how to search. The first hit showed exactly what is needed. The TI THS4031 has 100MHz bandwidth and slew rate, +/- 15V supplies, and can drive 90 ma. It is also low noise at 1.6 nV/Hz, $4.39 https://www.ti.com/lit/ds/symlink/ths4031.pdf Further measurements show the difference between the input and output signals is a 10mV sinusoid at 10V output. We don't need no stinking automatic bias supply for class B operation. -- MRM
