Adcom power light goes out fast when powered down?

I have an adcom gfa-5503 that I just bought and the power light should stay on for 30 seconds or so but it goes out in 3 seconds when powered down. I have a 555 that stays on for at least 30 sec. Does this mean the large power capacitors are bad? Also the transformer is rather loud.Loud as in buzz, and the loudness of the buzz is affected by other components being turned on or off. Any help here would be appreciated. Where to start ?thank you so much.mike
HiFi Engine didn't have schematics, so the following things are just SWAGS.
Amongst the causes of a buzzing transformer:
1. Extraordinary load - in which case, something would be getting warm.
2. Diode bridge with an open leg...draws upbalanced current, makes the transformer buzz pretty badly. You can see this with a scope on the filter caps, as the ripple freq would be 60 Hz, not 120 Hz.

If you can measure the voltage across the big caps before, at, and after turnoff, you'll have a lot better than these SWAGs to go-on.

Akitika GT-101 Audio Power Amplifier Kit
Update My Dynaco
The amp has 6-18000uf caps and 12-1000uf caps in the power section. My friend has the same amp and his light does take a long time to go out to. But he said his amp doesn't run nearly as hot. So you could be right about the bias. I will measure the voltage as joffe said and I can also check the bridge rectifiers for open legs.I did find a few resistors that had a bad solder joint and they were actually rocking they were so loose. I am working 65+ hours a week right now so it's hard to find time to tinker with it. I should have next weekend to myself to get busy with it though. I might call a few electronic repair places near me and see if they could test the caps for me if they have a tester. That would rule that out. Thanks for all the help here. Over and out. Mike
Here's a rough calculation.

assume +/-50 Volt rails, with 54000 uF on each rail. If the amp took 1 amp from each rail, the voltage on the caps would drop at the rate of 18.5 Volts/second
(assumes a constant current load).

That would be about 3 seconds for the light (3*18=54).

There's a darn good chance that either:
capacitance is lost, or more likely
bias current is much higher than it should be.

Akitika GT-101 Audio Power Amplifier Kit
Update My Dynaco
It's actually a 3 channel amp @200 watts rms. here is a pic it says 73 volts and 60 volts.


  • image.jpg
    660.2 KB · Views: 119
Ok so I took the big caps out to be tested and the guy said they are good. He tested the esr and they all tested good. I also checked all 3 bridge rectifiers and they all tested good to my knowledge. With my dmm on diode check I got a reading 1 way and nothing the other way. So at this point I am going to clean everything up and put it togeather. I will power it up and check and set the bias and dc offset to spec and just see if that was the problem. The guy at the repair shop said that the bias might be too high just like a few of you did. I might pull the transformer off its mount and see if I could put some extra dampening between the mounts too. Thanks for all the help. Wish me luck. Mike
If the bias is set too high there's a strong chance that the transformer is buzzing because of the high(er) current draw. Setting the correct value may, therefore, help cure it. Adding in some felt, or similar, to the transformer mounting will be useful if the case is resonating with the transformer.
I would like to go through the amp and replace a little more than just the 3 electrolytics on the amplifier pcb. I read a few post on here about some crappy sounding gfa5500 that is basically the same amp with mine having another channel. Mine is also lacking bass and the treble and mods are grainy compared to my 555. If I post a schematic, could someone pick some better components than what was used from the factory? I read about polystyrene caps being some of the best, can I use some of them? I would like to get the upper end to smooth out a little. Also there doesn't seem to be any detail compared to my gfa-555. I wouldn't mind spending a 100$ or so on capacitors and resistors if they get this amp sounding good for 10+ years or more. Ill work on getting the schematic up. The last thing I will upgrade is the big power caps. They supposedly tested good anyhow. Let me know what you think.

Ian Finch

Paid Member
2010-04-11 4:22 am
Coffs Harbour, NSW
Adcom models such as the 555 have different versions with way different bias levels. That sure makes a noticeable difference to sound quality when you have only 15mA bias current per transistor pair on 2 out of 4 known versions. The higher bias models have, IIRC, at least 4 times that amount of current.

The problem is, the thermal sensing is pretty vague and very slow to stabilize because there is no direct thermal contact with any part of the output stage! Attempts to increase the bias consequently start to run away over around 25 mA. The high bias versions obviously have a different thermal arrangement. If you can learn the practical differences so you can use a higher bias safely, you might get lucky and sort out the fundamental SQ issues, if the option is there in your model.
Ok so I replaced the 12 -1000uf power supply caps, and the 3 electros on each amplifier board. I fired it up to set the bias and was successful. I got each set to 50ma it took a while for the bias to settle down to 50 but the amp still runs quite hot. The transformer noise is a lot quieter. when I was setting the dc offset, the left and right channels were no problem to get set at zero. The problem was with the right channel. I think the trim pot might be bad or there is something else going on, it was at like 30mv and I adjusted it as far down as I could but the trim pot started clicking and it wouldn't go below 22mv. So I am assuming it is at the end if its travel. I also noticed that the bias was slightly touchy and the right channel has original pots where the others were replaced with a different brand. So its probably time to replace them. I am going to hook up some speakers and test the sound today so well see how it goes.
Oh is there anything else that would cause this amp to run this hot? And how do I go about measuring the 60 and 73 volt rails, I'm assuming one lead would be clipped to ground? I'm not sure about with a load but without a load hooked to the amp the power light now stays lit for a long time. I will see what happens when there are speakers hooked up though.