Adcom GFA4402

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
ok, I have this amp for repair and been in my shop for quite a while now. couldn't figure out the overvoltage protection as the pcb traces snake under the middle heatsink. lack of time could also have contributed to that. ;)

the amp runs fine, everything is in good order except that when supply voltage goes to about 14.0V it goes into protect and stays there. it runs again once supply voltage drops to about 13.5V and the remote line is cycled off-on.

anyone got a schematic? that would help me a lot figuring this out. thanks in advance.:)
 
Just curious, but what does the current do right before this happens?


A very good question damage ????

I am still looking for the documents I have. I looked today some while digging through two storage areas. And I will look again tomorrow after I get some help to move 15 Audison LRx 5.1k's, and a box full of other amps I have stacked in the way. I am almost positive the docs are in the box under this stack of waiting repairs. < smaller shop space then before :rolleyes: >
 
Just curious, but what does the current do right before this happens?

did you mean the idle current draw of the amp?

nothing. it just stays constant and when supply voltage reaches a certain point, the amp just shuts down and some red led lights up inside the amp. :confused:

if I remember correctly, idle current draw is a little over 2amps. I had to build a new CC/CV powersupply since I don't have an adjustable bench supply that can power this thing. :D

A very good question damage ????

I am still looking for the documents I have. I looked today some while digging through two storage areas. And I will look again tomorrow after I get some help to move 15 Audison LRx 5.1k's, and a box full of other amps I have stacked in the way. I am almost positive the docs are in the box under this stack of waiting repairs. < smaller shop space then before :rolleyes: >

ok, I'll wait for you. :) that's a lot of amps! I had a weird problem with an LRx before. it goes into protect with unequal rail voltages and was fixed by cleaning the PCB. it had some aluminum dust from heatsink polishing. (transistor clamping area, came from the factory that way) :dodgy: but after that, it still goes into protect in random times but when it doesn't it works just fine. just returned it instead to the owner as it seems to only go into protect when the remote line is cycled on/off quickly.
 
did you mean the idle current draw of the amp?

nothing. it just stays constant and when supply voltage reaches a certain point, the amp just shuts down and some red led lights up inside the amp. :confused:

if I remember correctly, idle current draw is a little over 2amps. I had to build a new CC/CV powersupply since I don't have an adjustable bench supply that can power this thing. :D

Now I'm not perry, so consider this more a mind exercise.

2amps seems like too much for idle. I have only two amps that draw that much at idle, and they make way more power.

Where the power supply fets replaced? If so, I'd first recommend looking at the gate resistors. They are probably incorrect for the replacements.

Otherwise, I might suspect a leaky output.
 
The Adcom"s draw a larger then normal idle current, just like some class A designs do. They tend to high bias the outputs for SQ reasons I assume. This thing looks Like Nelson Pass himself designed it inside, except for the AB output stage its just like his class A designs over in the other forums here. And it does have a class A drive stage to the output right out of the Pass handbook IMO...

I know of one adjustment on the lower rail supply control side. The other adjustments are bias idle current per channel and a DC offset fine adjustment per channel.
They modded some of these with a small board mounted on the turn on lead connection inside the amp. This was a two transistor turn on switch cause the early Adcoms had a very high current draw on the turn on lead.

By chance what are the lower rails doing when this apex point is reached ?

The usual problems I have seen on these is failure of the input mosfets on the channels. I have seen these burn the boards up in some cases, other then that flashed power traces on the supply side from excessive current draw through the power supply. That's why I am puzzled by djQUAN problem.... more later today....:soapbox:

as for the LRx issues we can talk more later or in anew thread about that. Some have Atmels inside monitoring and sometimes causing protection issues. just depends on which model. Audison can sell you a new protection board on their later models.
 
Last edited:
The Adcom"s draw a larger then normal idle current, just like some class A designs do. They tend to high bias the outputs for SQ reasons I assume. This thing looks Like Nelson Pass himself designed it inside, except for the AB output stage its just like his class A designs over in the other forums here. And it does have a class A drive stage to the output right out of the Pass handbook IMO...

I know of one adjustment on the lower rail supply control side. The other adjustments are bias idle current per channel and a DC offset fine adjustment per channel.
They modded some of these with a small board mounted on the turn on lead connection inside the amp. This was a two transistor turn on switch cause the early Adcoms had a very high current draw on the turn on lead.

By chance what are the lower rails doing when this apex point is reached ?

The usual problems I have seen on these is failure of the input mosfets on the channels. I have seen these burn the boards up in some cases, other then that flashed power traces on the supply side from excessive current draw through the power supply. That's why I am puzzled by djQUAN problem.... more later today....:soapbox:

as for the LRx issues we can talk more later or in anew thread about that. Some have Atmels inside monitoring and sometimes causing protection issues. just depends on which model. Audison can sell you a new protection board on their later models.

I'm as puzzled as you are. :confused: I never thought about measuring the rails since it was running fine at moderate supply voltages. I can take a peek with the rail voltages this weekend if you like. there is actually no turn off pop sound coming from the speakers when it goes into protection. it just goes silent with a red LED lit inside the amp.

the Lrx I had didn't have a microcontroller, it's all analog. or am I remembering the wrong model number. :D either way, don't mind it, it works anyway, the last time I left it. ;)
 
Thanks! Had a look in the schematic and figured out what to check for the OVP problem.

It turns out the problem was a leaky 15V 1W zener D305 which should trigger an SCR during an overvoltage condition. I have no idea why they had to use a 1W part where a 0.5W part should work just fine. Going to replace it tomorrow and see how it goes.

I feel stupid now lol this amp has been with me for 4 years hahaha Lack of time and motivation (as this is just a favor for a collector friend) could have been the cause of taking this long.

Sorry for bumping this. Does anyone know the rail voltages for this and the 4404 are ? I'm curious what power you can expect these to produce real world.

The rail voltages for the main power stage are listed as +/-24V on the SM but I can check the actual voltages tomorrow if you like.
 
Last edited:
I measured the unregulated rails and saw about +/-22V with 13.2V input at the battery terminals.

I also realized there was an error with the PCB. The 9V regulator and bypass caps (red circle in attached pic) are not connected to the main ground which caused the regulated +/-34V rails to go as high as 45V with high battery voltage. I couldn't find any burnt trace and there aren't any signs that this has been repaired except for a few bodges on the fan circuit which seems to be from factory. A simple wire jumper fixed the problem.

I also replaced the leaky zener with the same 1W part (1N4744) because that's all I had in stock. Now the OVP trips at about 16V whereas it tripped at ~13V with the leaky diode.
 

Attachments

  • 4402.png
    4402.png
    54.5 KB · Views: 48
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.