diyAudio (
-   Solid State (
-   -   What am I seeing here? (

martyh 11th February 2019 06:12 PM

What am I seeing here?
2 Attachment(s)
The attached plot shows the emitter current of Q2 and voltage at the node between Q2 E and Q4 C. What is going on with the wiggle at the bottom of the current plot? I see the same thing at Q4 E but 180 degrees out of phase obviously and everything looks fine in the load resistor.

indianajo 11th February 2019 06:46 PM

That connection between the emitter of q2 and collector of q4, and the junction of C3 and R3 is weird. If you look at the JohnE Basic50, and the Apex AX6 Retro Amp 50W Single Supply - Page 22 - diyAudio
they run the feedback back from that point to the emitter of the input transistor Q1. The AX6 schematic is on page 1 of that thread or somewhere close. AX6 also runs feedback from the output of the final capacitor to the emitter of Q1 through a resistor & a DC blocking cap.
This schematic is primitive anyway, with no emitter resistors to prevent the output transistors from thermally running away. Two drivers works better also, in a day when driver transistors cost $1 instead of $10 like in 1974.

Mooly 11th February 2019 07:25 PM

Is there something odd about those 2N3055H models? If you look at the waveform you will see it is effectively 'doubling' the frequency. Remove the bootstrap cap and the effective will vanish.

nauta 11th February 2019 07:49 PM

The circuit is missing the RF traps JLH added in the following issue of WW. Also, there is no scale to the waveform, so how wide is the signal swing for your test?

As a guitar amp, that would sound pretty cool - except for the heat :)

martyh 11th February 2019 08:29 PM

Thanks for all of the replies. It works the same way with a couple of 3055 models I have both of them below. The swing at the output is 7.82 volts peak. Taking C3 to ground fixes the current waveform but the distortion at the output goes up considerably. Reducing C3 to 4.7uf fixes the simulation. I have no idea why it works though. I should mention that normal loads in the 4 to 15 ohm range I don't see any of this.

.model 2N3055 NPN(Bf=73 Br=2.66 Rb=.81 Rc=.0856 Re=.000856 CJC=1000P PC=.75 MC=.33 Tr=.5703U Is=2.37E-8 CJE=415P PE=.75 ME=.5 TF=99.52N NE=1.26 IK=1 Vceo=60 Icrating=10 mfg=STMicro)

.MODEL 2N3055H NPN(IS=2.37426e-14 BF=129.119 NF=0.85 VAF=31.1252 IKF=0.990922 ISE=2.47498e-10 NE=1.89002 BR=1.01252 NR=0.924456 VAR=254.624 IKR=2.70227 ISC=2.47498e-10 NC=2.90624 RB=3.66609 IRB=0.1 RBM=0.1 RE=0.000352673 RC=0.0764459 XTB=1.34801 XTI=1.07207 EG=1.206 CJE=9.03089e-08 VJE=0.513954 MJE=0.59999 TF=1e-08 XTF=1.36696 VTF=1.02605 ITF=0.987296 CJC=5e-10 VJC=0.400243 MJC=0.410238 XCJC=0.803124 FC=0.661216 TR=1e-07 VCEO=60 ICRATING=15 MFG=Motorola)

johnego 12th February 2019 03:30 AM

I think that's normal. Question is why the 600?

The top transistor is a CCS so it is always better if you can choose the lower transistor to be the one with higher HFE. Try the 2N3055H at the bottom. The BF=129 (versus 73 of the 2N3055) shows that it is of higher gain.

martyh 12th February 2019 04:51 AM


Originally Posted by johnego (
I think that's normal. Question is why the 600?...

The six hundred is for a pair of headphones. I like the way they sound hooked up directly to a real power amp so I figured I would give the JLH a try. I'm just playing around with the sim to get an idea of what might go wrong and and where to look to fix it if it does. My experience level is low especially with solid state but I have never seen anything like the current distortion showing up here.

johnego 12th February 2019 05:18 AM

I see. But for best result, there is another JLH circuit intended for driving headphone :) Or you can probably use a parallel power resistor to take the load impedance suitable for the amp. Small external module complete with protection... :) BTW, if you happen to have different hfe of the output stage transistors, put the high gain one at the bottom and it might solve your problem.

mjona 12th February 2019 07:25 AM

There was a broadly similar circuit to the JLH1969 for headphone use published in Hi-Fi News & Record Review in January 1979 issue.

You can find some later Class A headphone designs by Linsley-Hood at A Paul Kemble web page - John Linsley Hood preamp designs. I supplied some of these to Paul Kemble but the board layouts are not shown. I have the layouts for those other than that for the Chiara kit version.

I still have the rather complex pre-amplifier published in Wireless World in 1982 which includes one of these headphone amplifier examples - this based around an IC op.amp and a couple of medium power transistors.

This could be built easily enough on project board on headphones the sound was something of an unforgettable surprise in view of the prevailing dislike of IC op.amps back then.

The quoted distortion is <0.01% on all loads down to 8 Ohms up to 3 volts rms output. The supplies for the IC are +/-15 V while the discrete output stage draws 40-50 m.a. from +/- 25 V. The transient response is also smooth both with resistive and reactive loads.

I did make up some boards for the discrete component version published in ETI in June 1984 but I did not bother replacing this IC one to which I had become quite addicted.

martyh 12th February 2019 11:10 PM

Using a higher hfe part at Q4 fixes the current distortion at Q2 but Q4 still shows the distortion. Reducing the size of C3 seems to work well no matter if the outputs are equal or using the higher gain part at Q4. If C3 is too low the harmonic distortion at the output goes way up. (second is raised by 20dB or so at 1uF)

I have still have no idea why it does the weird frequency doubling thing unless I change C3 and maybe it is one of those things that won’t happen IRL.

Thanks for all of the suggestions for tested HPA designs, I know this is kinda the path less traveled as far as these things go but it seems to work and I had many of the parts so not too much of an investment for me to build it and I can always repurpose it later if I can’t get it working at the breadboard stage.

Thanks again,

All times are GMT. The time now is 10:08 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio