• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

S'men A2502 trouble, help please.

Still can't make that sectionalize the quotes thing work. I'm determined if nothing else, surely I'll get it figured out eventually.

Anyway, got to work on the amp today. Did the Deoxit thing to the input gain pot.s and speaker select switches. I should have thought of and done this, anyway.
Study of the internal cabling showed some rearraning that may be fruitful. Now for adjusting the bias.
I had thought of doing it, and got so far as to locating R25 (voltage drop across which is adjusted to .8vdc for correct setting) while I had it apart. Hooked up to R25 on the right side (facing front of amp) power on, no input, read -.62vdc. Hmmm, didn't think to ID polarity before, so I went full range w/the adj. pot to see if knowing polarity was critical. Full range was -.20vdc to -1.4vdc. So, I adjusted for .8vdc, waited a few minutes, trimmed it back to .8vdc. Moved to the left side, hooked up, read 0vdc. Arrgh, wiggled the leads, 0vdc. Pulled leads, checked for continuity-ok. Read DC offset, got correct readings. Hooked up to left side R25, 0vdc. Tried another meter, 0vdc. Double arrgh. Connected amp into system for test, making music both channels. Back to the bench, hook up again, 0vdc. Tried connectring to the right side, 0vdc. Triple arrgh. This has got to be operator error but I'm outta daylight and patience. Better luck next time, I'm hoping.
 
Mazeppa said:
Still can't make that sectionalize the quotes thing work. I'm determined if nothing else, surely I'll get it figured out eventually.

Anyway, got to work on the amp today. Did the Deoxit thing to the input gain pot.s and speaker select switches. I should have thought of and done this, anyway.
Study of the internal cabling showed some rearraning that may be fruitful. Now for adjusting the bias.
I had thought of doing it, and got so far as to locating R25 (voltage drop across which is adjusted to .8vdc for correct setting) while I had it apart. Hooked up to R25 on the right side (facing front of amp) power on, no input, read -.62vdc. Hmmm, didn't think to ID polarity before, so I went full range w/the adj. pot to see if knowing polarity was critical. Full range was -.20vdc to -1.4vdc. So, I adjusted for .8vdc, waited a few minutes, trimmed it back to .8vdc. Moved to the left side, hooked up, read 0vdc. Arrgh, wiggled the leads, 0vdc. Pulled leads, checked for continuity-ok. Read DC offset, got correct readings. Hooked up to left side R25, 0vdc. Tried another meter, 0vdc. Double arrgh. Connected amp into system for test, making music both channels. Back to the bench, hook up again, 0vdc. Tried connectring to the right side, 0vdc. Triple arrgh. This has got to be operator error but I'm outta daylight and patience. Better luck next time, I'm hoping.
Click to expand...

On checking the bias you need to put the leads on each side of R25.
You are reading the voltage drop across the resistor.
So the polarity is relative to where you put the positive lead. In the end all that is important is that it is .8v dc

I hope you put this back to where it was as you can fry the outputs if it is turned up to high.

With power off measure the resistance of the trim pots making sure that they are ok.

It would not hurt to deoxit these pots and faderlube them.

By the way to make it easier to do quotes use the quote button that is in each post.
Then you will see the begin and end tags of the quoted section just copy and paste the begin quote
Mazeppa said:
Still can't make that sectionalize the quotes thing work. I'm determined if nothing else, surely I'll get it figured out eventually.
Click to expand...
and then go copy the end quote tag
 
Sorry about the whine in my previous post.

Since then, I did think to measure the resistance of R25 to make sure my probes were making connection there. Read 177.5 ohm on one side, 177.3ohm on the other side. Still getting 0.00vdc across R25, both sides.

Will try some DeOxit on R13. Thought problem must be elsewhere as I was initially successful adjusting the bias on the right side.

So, an unusually high resistance (from dirty contact for instance) across R13 can cause the voltage drop across R25 to clamp @ 0vdc?

Will heed your counsel and keep the amp offline until bias is set correctly.
 
Last edited:
R13s DeOx-ed, values are .02k-.432k/L, .01k-.469k/R. Voltage drop across
R25 still 0.00vdc both sides, neither responds to changes in R13 value.

Man, I know I got the expected response to R13 adjustments across R25 the first time I tried this, but I'm beginning to doubt myself as it looks more and more as if I've not identified R25's physical location correctly.
I don't relish the idea of pulling the board(s) and starting all over again, but I may have to.

Now that my bias values are unknown, is there danger in powering up for bias adj. tests, or am I ok as long as I'm w/o input and/or connected to load?

Thanks again, goat67. I admire your patience, and appreciate you sticking w/me on this.
 
If you have the bias pots back in the position they started at you can turn the amp back on.

You said the amp still works right?

What I would like to see is the readings relative to ground for Q9 and Q12
So clip the black lead to ground and read and document the following

Q9
Emitter =
Base =
Collector =

Q12
Emitter =
Base =
Collector =
 
Last edited:
Mazeppa said:
R13s DeOx-ed, values are .02k-.432k/L, .01k-.469k/R. Voltage drop across
R25 still 0.00vdc both sides, neither responds to changes in R13 value.

Man, I know I got the expected response to R13 adjustments across R25 the first time I tried this, but I'm beginning to doubt myself as it looks more and more as if I've not identified R25's physical location correctly.
I don't relish the idea of pulling the board(s) and starting all over again, but I may have to.

Now that my bias values are unknown, is there danger in powering up for bias adj. tests, or am I ok as long as I'm w/o input and/or connected to load?

Thanks again, goat67. I admire your patience, and appreciate you sticking w/me on this.
Click to expand...

Well lets make sure we are reading across R25 you will need to trace this out.

As far as are you OK well you may have fried the outputs or the drivers by cranking the Bias all the way up. What is the resistance you have them set to?

Check you meter settings make sure it is not going into overload it may show 0 if it is reading to high of a voltage.
 
Success!

The assumption that I was not on R25 turned out to be true. Tried thinking instead of just reacting, went back over my notes and "found" R25 again.
My best guess is that the first bias adjusment attempt (right side board) was as I remembered, because it was setting very close to where I left it, fine tuned it to .80vdc. Moved to the left side board, found it @ .712vdc, dialed it in to .801vdc. Double check, still on the money, called 'em good.

I'm thinking I got it correct the first time, then when I moved to the other side, got crossed up and repeated the same mistake over and over. My best excuse is that there are several resistors of the same value in the area. Oh, and for those not familiar with these amps, the component I.D.s such as Q9, R25, etc are not marked on the boards. I'm just glad I didn't mess anything up while I was bumbling around in there. Thank You, Lord for keeping my mistakes small.

So, the DC offsets are .016vdc and .019vdc. Bias is set properly. All Electrolytic capacitors have been replaced. I've taken my first steps up the learning curve of electronics work. The amp sounds very good and runs nice & cool.

Mazeppa is a happy camper and again thanks y'all for the help, especially my main Soundcraftsmen man, goat67.
 
Last edited:
Glad to hear that you figured it out. I thought you might be fighting OE ( Operator Error ). :D:D
Enjoy the tunes :music::music:
 
Mazeppa said:
So, the DC offsets are .16vdc and .19vdc. Bias is set properly.
Click to expand...


160 to 190 mV are not small offsets.

The problem with this amp, concerning the offset, is that the input stage is verry unbalanced. The collector current in Q1 is much smaller than the collector current in Q2. If you measure the voltage drop in R7 and R6, you may find a 1:4 ratio

When you have such a difference in the collector currents of a differential stage, and both transistor have the same current gain, the base resistances must be different to reduce the offset.

For this reason, R4 is much larger than R15. This helps to reduce the offset. But in your amp, the difference between R4 and R15 is not sufficient.

If you replace R4 by a larger resistor, say 75K or 100K, you may be able to reduce the offset to near zero.
 
ecluser said:
160 to 190 mV are not small offsets.

The problem with this amp, concerning the offset, is that the input stage is verry unbalanced. The collector current in Q1 is much smaller than the collector current in Q2. If you measure the voltage drop in R7 and R6, you may find a 1:4 ratio

When you have such a difference in the collector currents of a differential stage, and both transistor have the same current gain, the base resistances must be different to reduce the offset.

For this reason, R4 is much larger than R15. This helps to reduce the offset. But in your amp, the difference between R4 and R15 is not sufficient.

If you replace R4 by a larger resistor, say 75K or 100K, you may be able to reduce the offset to near zero.
Click to expand...

Good catch, offsets are in fact 16mv & 19mv, post has been edited. This OE is getting out of hand, I'd better get it together.

An interesting (to me at least) point is raised, however. If a potentiometer of correct type and value were mounted in the R4 position, would it then become the DC offset adjustment found in many amps/receivers? If so, would this be a desirable modification? I'm thinking of long term ownership and ease of maintenance is to be desired provided performance is not compromised.
 
I forgot to mention that it would be better to change R4 in both channel, in order to have the same gain and the same frequency response. So it's a kind of trade-off between the two channels, and the trim pot is not desirable.
 
ecluser
Thanks for the info on the DC offset. DC offset is always an issue for these amps.
The readings that he has is typical, the best I have ever got was 5mv and 9mv in an amp.

If both channels had a trim pot of 500 0r 1000 ohms installed in series with R4 what would be the issue or trade offs
that are mentioned in your previous post that make a trim pot undesirable?

Is there any other way to easily modify this circuit to incorporate an DC offset adjustment?

Thanks just trying to learn as much as I can
 
Last edited:
It would take much more than 1000 Ohm in series with R4 to fix this issue.

R4 and R3 form a resistive voltage divider. If R4 are different for each channel, if you use a trim pot for instance, the left and right gains of the amp will be different also.

I will try to explain the true nature of this offset.

Q3 and Q4 are current source of 1.8mA and 6 mA respectively. Easy to compute, it is ~0.7V / R6 for Q3, and ~0,7V / R17 for Q4. Add this on your schematic.

Ideally, the current from Q3 should be divided in two for Q1 and Q2. We will see this is not the situation.

The current from Q4 is also the current in Q6. You must visualize this. The collector current from Q6 is divided in two (unequal) parts. One part flows through the bias circuit, and the other part flows through the base of Q9. The current comming from the base of Q12 is, in first approximation, the same as the one going into the base of Q9. This current recollects with the current from the bias circuit in Q4. So the collector current from Q6 is verry close to the collector current in Q4, that is ~6mA

6mA X 47 Ohm (R9) = 0.28V, as you see on the schematic.

The voltage drop in the base-emitter junction of Q6 is approximately 0.6V (at most 0.7V) for a collector current of 6mA. So, the voltage drop in R7 is 0.9 ~ 1.0V.

This means the current in R7 is 0.4 ~ 0.45 mA.

To find the collector current in Q1, we must add the base current from Q6. The minimum current gain for the MM4003 is 20, at Ic = 10mA. But it can be higher than this.

If we assume that Vbe of Q6 is 0.6V , thus there is 0.9V in R7. You will have a negative offset if the current gain in Q6 is less than 35, and a positive offset if the current gain of Q6 is higher than 35, with the actual values for R4 and R15

There is more than one solution to fix this offset issue. You may trim R6, or R7 or R9. But this would change the open loop gain of the amp., and it may have consequences on the stability of the amp. In my opinion, the best place to add a trim pot would be at R6, say a 330 Ohm resistor plus a 100 Ohm trim pot. This would change only the current in Q2. If you have a positive offset, you must rise R6. This would reduce the current in Q2 and R15. I don't think the stability would be severely affected by this change.

My suggestion to change R4 has the advantage that it has no effect on the stability of the amp.

If you have a negative offset, you must reduce R4. If you have a positive offset, you must rise R4. So here is the trade-off, you can change R4 in both channel, with the same value for equal gains, until you have a small negative offset in one channel and a small positive offset in the other channel.
 
Thanks so much for the explanation on the operational theory on this amp

I will study this and apply it to the other amps circuitry.
 
Last edited:
ecluser said:
t most 0.7V)

My suggestion to change R4 has the advantage that it has no effect on the stability of the amp.

If you have a negative offset, you must reduce R4. If you have a positive offset, you must rise R4. So here is the trade-off, you can change R4 in both channel, with the same value for equal gains, until you have a small negative offset in one channel and a small positive offset in the other channel.
Click to expand...

I've got the urge to give this a try. Should be able to get from 16mv/19mv to approx. a 3mv span centered around 0mv.
As my offset is positive, I'll be trying values above 47K. Would you care to make recommendations as to resistor type, wattage, and ohmic value steps of increase?
 
As a first trial, go for a 51k, 1/4 Watt

At this point in the process, it is not important to use a metal film resistor.
 
ecluser,

Finally had opportunity to experiment w/different values at R4. Results were as you predicted. I progressed in +/- 4k ohm steps, DC offset steadily approached 0mv as R4 value was increased above the nominal 47K ohm value.

Initial DC offsets measured were:
A channel = 17.0mv
B channel = 22.2mv.
Both slightly higher than I measured before, I was using a different meter.
I stopped at an R4 value of 73.2K ohms (measured) which resulted in DC offsets of:
A channel = -1.0mv
B channel = 5.4mv

Mouser has metal film, 1/2watt resistors in 73.2K ohm, and 75K ohm. I'm thinking of getting both as the 75K ohm should center the span of difference closer around 0mv. Do these results seem consistent w/your assessment of this circuit?
 
Last edited:
Thanks for the post I will file this info away and do this on the amps I work on in the future.

ecluser
On a side note on this amp I have not been able to find the MM4003 transistors do you think the 2N5415 or the 2N5416 is a suitable replacement?
If not what would you suggest as a replacement?

Thanks Jerry
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom