[b]Please Help - I messed up one channel of my AKSA 100[/b]

dgarner · « **on:** 13 Dec 2003, 01:19 pm »

Hi, my aksa 100 had been working great for over a year now and sounding great. I decided to finally move it to a rack mount cabinet and that’s how I messed up the channel; I think the cause was that one of the power supply leads hit the heat sink with some residual charge left. Since the migration the right channel is dead, all I get is some hum when I tested it with some junk speakers the left channel is still perfect. I guess I probably destroyed a transistor or two, but do not know how to tell for sure or identify which one. Any help would be greatly appreciated.

Thanks,
Darin

PSP · « **Reply #1 on:** 13 Dec 2003, 02:33 pm »

Darin,
Sorry, I can't help with your direct problem (I'm sure Hugh will be up soon with specific advice), but I had a similar incident with my AKSA 55... I was poking around on an amp board, getting ready to put in some hyperpart I guess, when I got a big spark between my jewler's screwdriver and the ground tab on one of the transistors (the amp was powered off and had been unplugged for several minutes)... luckily in my case there was no damage.

To reduce the chance of that happening again, I've fitted bleeder resistors on my power supply PCB... to do this you mount a 1/4 - 1/2 watt cheap carbon resistor, about 15k ohms, between each positive (+36v on the 55w AKSA) to the power supply single point ground and a similar resistor between each negative (-36v on the 55w AKSA) to the single point ground... for a stereo amp, that's four resistors. These resistors will bleed the power supply caps over a period of several minutes and help prevent problems like this in the future. There is no sonic penalty (and one poster a while back suggested that there may be sonic benefits)...

Search "bleeder" on this forum for more info.

Good luck on getting your amp fixed. I know what a bummer it is to lose your AKSA for a while!

Peter

AKSA · « **Reply #2 on:** 13 Dec 2003, 08:13 pm »

Hi Darin,

I'm sorry to hear of your problems, and my thanks to Peter for his advice.

First, remove and replace the two fuses on the sides of the pcb.

You can first check each, after removal, with a DMM for continuity.

Sometimes a power supply short will merely take out the fuse, and nothing else.

Try this, and if unsuccessful, I'll give you the next step.......

Cheers,

Hugh

dgarner · « **Reply #3 on:** 13 Dec 2003, 09:17 pm »

Thanks Peter, I will probably do that once I get it working again especially since that will make the LED's go out faster.

AKSA thanks for your response also, I have already checked the fuses and they were good. I removed them and set the trim pot all the way counter clockwise in an attempt to go through the initial testing phase in step 16 of my instructions, but the drop readings are so low across both (R25) & (R26 on drawing listed as R21 on instructions) that I stopped 0.13V. The power supply is putting out 47V.

Thanks,
Darin

PSP · « **Reply #4 on:** 14 Dec 2003, 12:00 am »

Hi Darin,
May I gently suggest that now would be a good time to put in the bleeder resistors... since you are in there doing surgery, can you imagine the pain if you zapped something else due to residual voltage in the PS caps... generate problem #2 while working on problem #1???? Yikes! I could probably open a window and hear you scream....

Just a thought,
Peter

AKSA · « **Reply #5 on:** 14 Dec 2003, 04:34 am »

Hi Darin,

OK we have eliminated one potential problem.

1. Remove both fuses, check both fuse resistors (R21/22) are 100R (and not 10R, measure them!!), hook up the power, and check you have 49V +3V/-1V at EACH rail.

2. Now, check the voltage drop across each of the fuse resistors. With bias wound right back, it should be about 1.3-3 volts. If not, take note, and move on.

3. Measure the voltage at the output of the amp. It should be 0V +/-50mV (with no fuses in place, of course).

4. Now measure the voltage with respect to ground at the base of T1 (connect UNDER the pcb at the amp side of C1), and at the base of T2 (connect at the TOP of R9). These should be around 100-200mV, and should both be within about 2% of one another.

5. Lastly, check the voltage at the junction of R5/R6 with respect to ground. It should be 28-30 volts.

Let me know results, and we'll talk more.....

Cheers,

Hugh

dgarner · « **Reply #6 on:** 14 Dec 2003, 05:53 pm »

Thanks Peter, lucky I have an old Bryston 4B to use while I'm down, but I sure don’t want to make the problem worse.

Hi Hugh, thanks for the help, I'm glad to see that your business is doing well. I'm personally waiting on the Valhalla kits to be available! Well I performed the checks that you requested and here or the results.

1)R21 & R22 are 100R (101R to be exact) and the rail voltage is 46V, it is low by design because of my house having higher than normal voltage (you helped me with that a long time ago!)

2) The drop across the fuse resistors = 0.12v

3) Voltage at the output is way high 43v.

4) The base of T1 is 280mv and the base of T2 is 8v.

5) The voltage at the junction of R5/R6 = 44.3v.

Thanks,

Darin

AKSA · « **Reply #7 on:** 14 Dec 2003, 08:44 pm »

Hi Darin,

It rather looks like one of your outputs has gone south.....

Disconnect power supply leads from the amplifier, and with your DMM check each and every one of the four outputs.

On diode test, check for low resistance EACH WAY across the outside leads; base to emitter. If less than the diode test voltage (use a known good transistor of any type to test base/emitter to find the reading, typically 0.6), check it's the same both ways. If it is, the device is faulty.

To confirm, set DMM to low resistance and measure between collector (center) and emitter (right), and check it's the same both ways, very very low, say less than 100R.

If so, let me know which ones are affected, and I'll send you new matched pairs.

Accidents with leads during surgery are VERY common. You should always short out the power caps (without the power switched on, of course) before rewiring or remounting a power amp. A bleeder resistor certainly overcomes this problem; typically around 15K from each rail to ground.

Cheers,

Hugh

dgarner · « **Reply #8 on:** 14 Dec 2003, 10:35 pm »

Hi Hugh,

I checked it out and the readings were as follows:

T10 - .452 & .484 (good I guess)
T8 - .453 & .484 (good)

T7 - .468 & .468 (Bad??)
T9 - .467 & .468 (Bad)

I performed the second test and T7 & T9 had the same reading both ways.

Do I need to test anything else?

Thanks,
Darin

AKSA · « **Reply #9 on:** 14 Dec 2003, 11:38 pm »

Darin,

Please remove T7 and T9, and test them individually.

I fully expect there to be a short across collector (center) to emitter (right), and that the resistance (use low Res on the DMM) will be the same both ways.

Cheers,

Hugh

dgarner · « **Reply #10 on:** 15 Dec 2003, 02:54 am »

Hi Hugh, took them out and did not get the readings I was expecting, but I'm sure it will make sense to you.

Transistor 9
c to e (does not register appears open both ways)
b to e (.483 one way does not register other way)
b to c (.479 one way does not register other way)

Transistor 7
c to e (does not register appears open both ways)
b to e (.483 one way does not register other way)
b to c (.480 one way does not register other way)

Thanks,

Darin

AKSA · « **Reply #11 on:** 15 Dec 2003, 04:42 am »

Darin,

I'm sorry, these look fine. Can you reinstall, and try the other two outputs?

We'll nail this !@#$ yet!!

Cheers,

Hugh

dgarner · « **Reply #12 on:** 15 Dec 2003, 03:03 pm »

Hi Hugh, I guess these are good too.

Transistor 8
c to e (does not register appears open both ways)
b to e (21.51 one way does not register other way)
b to c (19.50 one way does not register other way)

Transistor 10
c to e (does not register appears open both ways)
b to e (21.79 one way does not register other way)
b to c (19.77 one way does not register other way)

Thanks,

Darin

AKSA · « **Reply #13 on:** 15 Dec 2003, 08:36 pm »

Darin,

I'm at a loss here; the picture I'm getting is somehow not making sense to me.

If the outputs are OK, it's likely that the inputs OR the voltage amp are blown, though I can't imagine why.

See if you can measure the base/emitter voltage of T3, the voltage amplifier (just measure across R4, or the outer leads of T3). Should be 630mV plus or minus 10mV.

Failing this, try measuring the voltage from base to emitter on EACH of the two input devices. Of particular interest is T2, which for reasons of very high offset will have considerable reverse bias on the base/emitter junction. This may damage it.

Another consideration is the power supply. I suppose it's possible you've damaged some diodes. Can you disconnect the amp, fire up the power supply, and measure the two output rails with black probe on star earth?

Also measure this with the AC setting; should be almost zero if the diodes are good.

Hope this helps, this can be a trying business.......

Cheers,

Hugh

dgarner · « **Reply #14 on:** 16 Dec 2003, 03:56 am »

Hi Hugh, I performed the tests and are the values.

T3 - 3.307K both ways
T1 - 23.3 & 22.5
T2 - 26.5 & 23.6

The power supply rails are running at 47v and the diodes seem to be fine, readings are the same for both left and right channel and left channel works fine with either set of feads.

Thanks,
darin

AKSA · « **Reply #15 on:** 16 Dec 2003, 05:32 am »

Darin,

Can you measure those transistors for voltage, not resistance?

We might be getting warm......

Hugh

dgarner · « **Reply #16 on:** 16 Dec 2003, 02:38 pm »

Hi Hugh, here's the readings:

(T1) c - 42.9v
b - .295v
e - .926v

(T2) c - 47.2v
b - 8.63v
e - .921v

(T3) c - 46.5v
b - 43v
e - 47.4v

Thanks for all the help,

Darin

AKSA · « **Reply #17 on:** 16 Dec 2003, 07:54 pm »

Darin,

This looks as though T3 has failed - VERY unusual - but it might also have taken out T1/T2.

I see this fault rarely, but what concerns me is how it happened. I can't see a simple short at the output stage doing this, but no doubt something nasty has happened to the front end.

The best option is to send you a new matched input pair and a T3. Put 'em in, see what happens. It might take some time to get to you, regrettably....

It won't be expensive, just a bit of post. I'll try to get them off today, and will let you know on despatch.

Sorry about all the time delays!

Cheers,

Hugh

dgarner · « **Reply #18 on:** 17 Dec 2003, 02:34 am »

Thanks, Let me know what I owe you. I appreciate all the help, and can't wait for the super duper version to come out!

Thanks,
Darin

dgarner · « **Reply #19 on:** 31 Dec 2003, 03:01 pm »

Hugh, installed the parts you sent and its back up and sounding better than ever. Thank you very much for all the help. Also, if you ever have a request from a potential customer in my area let me know.

Darin