diyAudio

diyAudio (https://www.diyaudio.com/forums/index.php)
-   Headphone Systems (https://www.diyaudio.com/forums/headphone-systems/)
-   -   Reparing Class A symetrically driven Headphone Amp kit (https://www.diyaudio.com/forums/headphone-systems/356686-reparing-class-symetrically-driven-headphone-amp-kit.html)

Lawrence.M 30th June 2020 04:14 PM

Reparing Class A symetrically driven Headphone Amp kit
 
1 Attachment(s)
I am trying to repair my Linsley Hood designed Class A symetrically driven headphone amp section of my pre-amp that I use as a high level output to my Power amp.
I am guessing that the output transistors are faulty(after over 10 years of use) so have removed them.These are the TIP 41C and TIP 42C(higher rated than the original design which shows TIP 41B and TIP 42B). My Question is,is it important for these output devices to be matched to have close hfe values,particularly as they get quite warm and I assume they would last longer if they shared the output current evenly.If this is important where can I get matched pairs from?
Alternatively I have a lot of quite closely matched(eg-197 hfe PNP/170 hfe NPN) MJE 15034 and MJE 15035 transistors) to hand which have similar specs to the TIP 41C and TIP 42C but slightly lower maximum current and Power ratings(4 Amps,350V,50 Watts as opposed to 6 Amps,100V, 65 Watts) according to the data sheets.Would these be a better alternative?
I have attached the schematic that came with the kit for reference.
Any advice much appreciated.

rayma 30th June 2020 05:08 PM

The single positive and single negative output transistors automatically have the same DC bias current.
If paralleled devices for each polarity are used, those should be matched.

You didn't mention the actual problem it has.

Lawrence.M 30th June 2020 08:23 PM

Quote:

Originally Posted by rayma (https://www.diyaudio.com/forums/headphone-systems/356686-reparing-class-symetrically-driven-headphone-amp-kit-post6259563.html#post6259563)
The single positive and single negative output transistors automatically have the same DC bias current.
If paralleled devices for each polarity are used, those should be matched.

You didn't mention the actual problem it has.

Thanks Rayma,One channel would cut out every so often,then eventually there was no music and just a buzz/Hum sounding like mains hum.To complicate the issue,when the channel first cut out I found that the off board power supply to the pre-amp had faulty regulators.I replaced these and the problem came back but only when using the high output from the pre-amp which uses the class A headphone circuit.As the output transistors get quite warm and the problem was initially intermittent,I thought they might be the culprits.Also when looking at the board there were no obvious signs of component damage like bulging caps etc.
After taking out the TIP41C's they measured an hfe gain of 55 and 53 which seems low at least compared to the TIP42C's that measured 187 and 203.

Lawrence.M 3rd July 2020 04:04 PM

DC Offset seems way to high
 
Ive just replaced the output transistors but the DC offset is measuring 0.77V and 0.69V from each channel and the trim pots don't seem to make any difference throughout there adjustment range. Any One got any ideas of where I should look next?

PRR 4th July 2020 05:27 AM

1 Attachment(s)
Yank R9. Ground the input. Measure all the key voltages. I've sketched approximate guesses. Yours would be similar, but I expect you will find one/some WAY off. Crayon your readings and post back.

Lawrence.M 4th July 2020 08:54 AM

Testing ciruit
 
Quote:

Originally Posted by PRR (https://www.diyaudio.com/forums/headphone-systems/356686-reparing-class-symetrically-driven-headphone-amp-kit-post6263437.html#post6263437)
Yank R9. Ground the input. Measure all the key voltages. I've sketched approximate guesses. Yours would be similar, but I expect you will find one/some WAY off. Crayon your readings and post back.

Thanks PRR.But before I start what do you mean by Yank R9?

sgrossklass 4th July 2020 07:37 PM

You're supposed to (temporarily) remove it.

Lawrence.M 4th July 2020 09:57 PM

Ah yes I see.
Thanks sgrossklass

Lawrence.M 6th July 2020 05:37 PM

Measurements compared to ideal
 
1 Attachment(s)
Here are the measurements taken for comparison.The Green are one channel and the blue are the other one.They seem to be pretty close.I don't know whether the 1.5V and 1.47V DC offset at the output is reasonable?If so strange that I couldn't adjust this to lower than 0.77V and 0.69V with the trimpots before temporarily removing the resistors for testing.

PRR 9th July 2020 04:22 AM

1 Attachment(s)
Thanks for the detailed measurements. Something is not balanced but I don't see what. I asked The Idiot to help. First crack gave +0.8V output offset, reasonably close to your observation. So the thing is not inherently zero-centered. And there are some large unbalances. Since my simulated resistors are "exact", the only cause is the inherent difference between NPN and PNP (of similar die-size), which these models obey (a bit too well).

There is a conflict. As Rama observes, Q5 Q6 "must" pass the same current because they are series and there should be no stray DC out of that path. However Q3 Q4 also force Q5 Q6 currents, and not-the-same because Q5 Q6 can not (practically) be exact-matched for both Vbe and hFE.
Quote:

Originally Posted by maurihal (https://www.diyaudio.com/forums/headphone-systems/205803-upgrade-pro-ject-head-box-ii-headphone-amp-post6267548.html#post6267548)
Someone said a long time ago, in the seventies: "NPN, PNP transistors are as truly matched as men and women of the same height and weight".

I added-back the offset trim. It can swing over +/-2.5V at the output. So to get say 10mV offset you have to trim to one part in 500. This is incredibly fine adjustment for a single-turn pot. (Altho not that much finer than an AM radio which must be within 2kHz in a 1MHz band.) It seems that (for these sim parts) the feed to the 10k resistor should be 0.591V for near-zero output offset.

And yet you say it won't pull-in at any setting. Is the voltage *across* RV1 just about 1.44V? But even so, trimming all the way to the negative rail should cause a -negative- output offset, and you only have positive output??

I would wonder about replacing Q1 Q2 just as a mindless way to eliminate possible damage. The biasing should not require high-grade transistors, although hiss and other parameters may suggest types like the ones the Designer specified.


All times are GMT. The time now is 07:16 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio

Wiki