[Builder Brad]

HELP!

I have just removed C21 from my GK 1R, the 1uf Auricap, and have decided to use it to replace the .470uf C1 input cap in my AKSA55. Straightforward job, 1/2 hours work, or so I thought. I checked the of-set before putting the amp back into the system and thought that everything was fine, however as soon as I turned the amp on I noticed that mid range driver on 1 channel was almost popping out of its frame, something was not right!


Taking the amp out of the system and measuring a number of areas indicated in the fault guide gives the following:

Voltage across C4 69.9v
R10 0.09v
R5 0v
R6 0v
Of-set 33.5v My Seas mid range seems to have survived, although I dont know how as it was on for a good 45 seconds.
R1 352mv
R9 3mv
R4  2.07v
T1 T2 emitters 1v
R5/R6 junction to gnd 34.3v
C7 0v
R15 0v

and I measured the voltage drop at 0.7v with the safety fuses removed.

Will it live?

The only thing I can think of is that I have shorted one of the NPN transistors, as I did cause a spark when I Accidently discharged the PS caps somewhere on the amp board with DMM probes.

I cannot believe that I have done this, I was trying to make it better, I guess if it aint broke.............

Brad

[AKSA]

Hi Brad,

Unfortunate - let's fix it....

I think you MAY have blown the input transistor of the long tailed pair, T1, and one or both of the outputs.

Indications are vaguely pointing to the npn output device, the 2SC5200.

I would first start with replacing that device - they should be available at a local electronics stockist, it's a common transistor, and see what happens.  While you are there you might also buy ten 2N5401s, just in case.  They are cheap, and you will need a few to match them for beta on your DVM.

This has happened because of the shorting accident, and shows how vulnerable these amps are when you work on them under power!

Cheers,

Hugh

[Builder Brad]

nice one Hugh! That was quick.

I suspected one of the NPN devices, as this was mentioned in the 55 build manual as a possible cause of the high + of-set.

The power was turned off when the incident occured - theres a fair bit of juice in them caps! and I thought I had purged them before poking about.

My wife is due back tommorow from NY, so the rest of my kit will be safe from my tweaks - idle hands n all that stuff over the last few days I guess.

Brad

[AKSAphile]

Hm....but why SA1943 did not fail too ?

Maybe still some DC protection device will not be bad idea on AKSA/LF amplifiers to protect such DC offset?

[AKSA]

Brad, Andrej,

The NPN almost always fails first, with the pnp then able to maintain offset long enough to turn it off.

This was a happy discovery when I first did destruction tests years ago.  Consequently I felt little need to establish full DC protection, which would have very considerably increased costs.

Happy ending, I'm real pleased!

Cheers,

Hugh

[thunderbrick]

nice one Hugh! That was quick.

I suspected one of the NPN devices, as this was mentioned in the 55 build manual as a possible cause of the high + of-set.

The power was turned off when the incident occured - theres a fair bit of juice in them caps! and I thought I had purged them before poking about.

My wife is due back tommorow from NY, so the rest of my kit will be safe from my tweaks - idle hands n all that stuff over the last few days I guess.

Brad

Better not tweak HER, as you could get seriously burned if you short something out.   

[Builder Brad]

OK,

thank-you for all of your help so far with this.

I did get to replace the 2SC5200 Tosh transistor, and both of the 2n5401s over the new year - and its still giving 30 odd volts at the output. What should I look at next? HELP...........

I really should leave things alone when they are working, better expect another cry for help when those caps come in from the group buy!!!!!!

Brad

[AKSA]

Brad,

I'd be checking the base/emitter voltage of the voltage amplifier transistor, T3.

Switch on the amp, check the voltage across R4 (1K5), which lies directly in front of T3.

Specifically, it's likely that it has a partially broken base emitter junction, and that this is causing this problem.

A new one and you're away......

Cheers,

Hugh

[Builder Brad]

thank-you Hugh,

The amp is not accessible right now, but I did check R4 previously as posted above I measured 2.07v across R4 - could this be it!!

quote"A new one and you're away......"  I really do hope..

Brad

[Builder Brad]

is this the 2sc1819 transistor?

are there alternatives that I could use? I have strugled to get the SC transistors from local off the shelf suppliers.

Brad

[AKSA]

Yes, Brad,

That's it.....  this junction is kaput.

You can use a 2SC3423, and you could also try the philips BF469, but ensure pinout is the same, you might have to turn it around.

Hope this helps,

Hugh

[whubbard]

Hi Brad,

I hope you 55 is fixed. A little of topic question, did you replace C21 with a wire?
If yes, did your GK-1 continue to operate as normal? I had issues with mine with a wire in a C21, and Hugh and I could figure out the problem...replacing C21 did the trick.

Thanks,
West

[Builder Brad]

cheers Hugh,

I can easily get the Phillips part, so this will be the way to go.....I will check the orientation carefully though!

West,

I did place a wire on the pcb where c21 was positioned. I then moved the output caps completely off-board because of their size. I am running the gk1 minus the input caps, without any problems.

I am trying to get to the bottom of a strange noise that the amp occasionally makes - which I am going to have another look at when I open the amp up to fit the parts that are on the way from the group buy you are kindly organising!

Brad

[Builder Brad]

Hugh,

I have an account with Digikey - and I have a credit on my account!

I can normally get everything from Digikey, but I am finding it hard to get any of those transistors you quote. Are there any other parts that would work equally well in the T3 position.

Brad

[AKSA]

Brad,

I shall send you a couple of the 1819s in the post, simple envelope.

I have your address on file;  it will take about a week.  Would that be OK?  No charge, more time involved doing the billing than just giving it to you.......

Cheers,

Hugh

[Builder Brad]

Hugh,

That would be fantastic, YOU are a Gent and scholar!

Brad

[AKSA]

Brad,

Done today!

Hugh

[Builder Brad]

Hugh,

The 1819 came in the post over the weekend - thank-you!

I have been using a chipamp with the LM3886 IC in place of the AKSA on my Orions, as he chipamp has almost identical Dc gain to the AKSA, and  have put up with almost 2v of DC of-set as the amp would not tolerate the litz braided speaker cables that I have been using.

There is NO comparison.

final of-set on the amp was 2mv and -17mv The higher of-set was on the channel that I "broke" and has two new nearly matched transistors, rather that the nicely matched pairs that you originally supplied.

Brad

[AKSA]

Hi Brad,

Good news, the mail worked!

Those C1819s are obsolete now, very difficult to find.  I bought 1600 of them eight years ago on close out, they were a great find.  They've been very successful in this design.

I'm glad you drew attention to the qualities of the AKSA over the chipamps.  I can admire the chipamps for the ease of implementation, but their sound quality is not the same.

Let us know how the rebuild goes, Brad, we want you up and running as soon as possible!

Cheers,

Hugh