Damaged Adcom 565 Monoamps

audio_shop

Member
2008-01-02 10:40 pm
I have a pair of these that are damaged, both as a result of the leaking 220uf 25V caps. One has a circuit board that's badly damaged and the other not so bad. Surprisingly enough, the one with the worst damage is the one I am having less problems with.

In any case - having read most of the info in these forums and others, I have replaced all the usual resistors and caps. Cleaned and repaired the PCBs as best I could.

Now my two remaining issues:
1) The acid damage resulted in a problem with a SCR type rectifier. The original part number is 'ADCOM J2 8934M' - it has a TO-92 (3 lead) case. I need a available replacement part number and manufacturer. I usually deal with Mouser.

2) 2nd issue, on the amp with less damage, I have the worst DC offset ever (meaning if you attach it to a speaker - sounds like a 22 pistol going off - after that the voice coil has no more than a minute or two to live. As I said I have replaced all the recommended parts, and using a VTVM have checked about every component and haven't found anything wrong in checking mirrored components and from amp to amp.

I do not have a service manual so I'm a little blind.
 

audio_shop

Member
2008-01-02 10:40 pm
Me 'audio_shop' again. One thing to add that might be helpful in diagnosing the problem. On turn-on, the instantaneous distortion light comes on and stays on, and the Q111 transister (main driver left side of board looking from the the front of the amp) becomes hot to the touch in about 30 seconds. Sounds like a short somewhere - any ideas where to start looking?
 

audio_shop

Member
2008-01-02 10:40 pm
Now that I'm deep into this and still not making much progress (although I did find a source for a service manual), I'm wondering if someone could come up with a design for a simple driver stage that would work with this amp. After reading all the posts on 565 issues (and they are almost all the same), you have to wonder wasn't there a better and more reliable design for this section, and why is DC at the outputs such a big problem with the 565.

I just fixed a HK Citation sixteen, and it was soo much simpler and I really think it sounds better. Of course the HK is not direct coupled, but other wise from a power supply and output stage perspective, the two amps are almost the same piece.

I would be more than happy to make one of these amps into a test bed for an alternative driver stage if the forum members could point me in the right direction.
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi audio_shop,
Sorry I haven't responded until now. Here goes.

1) The acid damage resulted in a problem with a SCR type rectifier.
No, those are precision 7 V regulators.

2nd issue, on the amp with less damage, I have the worst DC offset ever
You should have watched for DC offset. Your fault harks back to ...
Cleaned and repaired the PCBs as best I could.
Not well enough. That's the simple truth - sorry. back out they come. please read carefully the posts of mine you have come across regarding cleaning those boards. I wasn't kidding.

you have to wonder wasn't there a better and more reliable design for this section, and why is DC at the outputs such a big problem with the 565.
Bad assumption. The design is sound, just some bad caps that didn't show up until years after the stuff was in the field.
I would be more than happy to make one of these amps into a test bed for an alternative driver stage if the forum members could point me in the right direction.
See above. I think you would be making a mistake. These are not the best amps ever, but they are pretty good. At that price point, they are great!

-Chris
 

audio_shop

Member
2008-01-02 10:40 pm
Thanks for the input Anatech,

I'm purchasing a set of precision 5W Zener diodes to provide the voltage regulation. In the meantime I have completed my repairs on one of the two amps, specifically the one with the severely damaged PCB. Getting good results so far even with the cheap OP07 Opamp while I wait on a better piece.

Regarding the cleaning of the board, I used a mix of baking soda and water with a small brush to do localized cleaning, followed by clear water, and then alcohol. It all worked very well.

The second amp without the PCB damage is the one with the more severe DC offset problem. Because replacement of the typical resistors and Caps, and the IC still haven't resolved the problem, as it did in the other amp, I'm thinking that Q101, 102, 105, or 106 may be damaged. Would it be best practice too simply replace them all, or is there some way to isolate which it might be.
BTW, the DC offset problem is positive.
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi audio_shop,
Those zeners are not correct. You need to get the original types made by Linear Technology (I am guessing here. Memory issues). They are 7V each, two in series.

The op amp is for DC servo use only. No audio passes through it. You need a low current type. If the leads are not rotted off them, they are probably good after a proper cleaning.

Note that the stuff you are trying to get rid of is not visible to the naked eye. I wonder if a UV light would show it up? It's also very difficult to get rid of. I personally don't think any of those you have worked on are completely clean. Sorry, but that electrolyte is devilish stuff to remove.

Replace anything you think is damaged. The diff pair must be matched carefully.

-Chris
 

audio_shop

Member
2008-01-02 10:40 pm
And how would you go about testing the diff pair to ensure that their characteristics do match or are at least very close.

The original LT regulators are no longer available. they are listed in the Service manual as zener diodes (although voltage is not specified, I am taking you word for it that the circuit requires 7 V). The pieces I am puchasing have a different form factor, but technically replicate the originals. They are rated 6.8V nom, with an operating range of 6.5 to 7.2 Volts
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi audio_shop,
They are two precision 7V regulators in series to give 14 VDC. It's the low noise they are after here, so regular zeners (0.25W) would work, but the noise level might be higher. I bought some last year, so they should be available at Newark or Digikey I would think. The very last thing I'd use is a large zener. The current through there is not high and you'd be operating on a very soft knee.

What are you getting to replace them? Just curious.

-Chris
 

audio_shop

Member
2008-01-02 10:40 pm
ON Semiconductors Surmetic 40 Zener Diodes. Part no. 1N5342BG. Check out the spec sheet - very stable. Nom. 6.8 volt.

Your saying you were able to get the exact replacement thru digikey or a zener to do the same job. If it was a replacement, you wouldn't happen to have a part number?

BTW, I found a nice relatively inexpensive replacement for the IC101 (ADCOM 2A), its the TI OPA177F. There is also a 177G which I am also buying whose characteristics are most similar to the OPA97ED/LT1012.

The G is about $2, and the F about $6.

The 177F has even lower offset V and is more stable at temperature, although it gain is less at 25 C Nom. I intend to try both. Right now I have one amp working well on a standard OP07 (which I know is just lucky).

- Tom
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi Tom,
I am still looking for those notes I made. The originals were a shunt regulator IC design in a two lead TO-92 case. It's possible you can find the exact part by looking for two leaded shunt regulators. What I bought was definitely not a zener diode, they were shunt regulators.

The exact DC offset is not that important. You will be well below 10 mV even - so who cares beyond that? You aren't going to hear it either. Do check your supply voltage on the op amp. If it draws too much current, the regulators will drop out. That you might hear.

The original op amps might be fine after a really good cleaning.

-Chris
 

easecat

Member
2009-02-28 1:22 am
Now my two remaining issues:
1) The acid damage resulted in a problem with a SCR type rectifier. The original part number is 'ADCOM J2 8934M' - it has a TO-92 (3 lead) case. I need a available replacement part number and manufacturer. I usually deal with Mouser.

hi all

I am looking for the replacement for ADCOM J6 8929M.
could someone provide the message? thanks
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi birdyman,
What more can I say?
For one thing, these are shunt voltage regulators, but not zener diodes. The voltage isn't that critical, but the original parts were low noise shunt voltage references with a nominal 7.5 VDC rating. The op amp is a low quiescent type with low DC offset error. I'm not sure if the original parts are still available these days, but you have enough to go on.

I should also mention that these parts very rarely ever fail in service. Check everything else before you decide these are bad. Zener diodes as a replacement may or may not work. Keep in mind that a zener has a soft knee at lower currents, and they also become for more noisy at these same low currents.

Are you working on an Adcom amplifier, or just curious about the parts?

-Chris
 

birdyman

Member
2010-01-15 2:28 pm
Adcom j6 = lm336 (2.49v shunt reg.)

Hi Chris

I am working on my Adcom 585 amp. I followed your post to repair DC offset. Dc offset is fixed but I still have thump issue and not sure of operational condition of "Ac input/bias time delay" function. I think if my AC input/ bias time delay board is working correctly, I shoud hear delayed "On" state of amp when I turn it on(??), but my amp turns on instantly as I toggle the switch. So I decide to investigate diff pair bias circuit & delay board. When I check conductivity of Adcom j6 anode to cathode, reading changes constantly ( temperature variable??). I found on one web site it says Adcom j6 is same as LM336, but I am not sure of quality of that info. One of your thread posted Adcom j2 is LM329. Based on that info, LM336 does make sense(??). As of now, I am about to test opto isolator (NEC ps2505).
Thanx
Scott
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi Scott,
Firstly, J6 isn't the diode I was thinking of when I answered your post. The part I was referring to was J2. There are 4 of the J2 references to power up the servo op amp. These are the things that are typically asked about. I have never been asked about D115 before, and it's never been a problem either.

D115 sets up the voltage reference that is applied to the base of Q118. This, along with R149, creates a constant current source used in biasing the complimentary differential pairs at the input. It is extremely unlikely to be defective. A problem here would either cause a no sound situation (short = no tail current), or smoke and possibly flames (open = uncontrolled current). Let's ignore that part for now. Make sure the solder connections are good and clean. Take care to avoid overheating the part.

The delayed time relay part of the circuit you are looking at shorts out an anti-surge resistor after a few seconds. It doesn't have much to do with muting the audio output. You may hear the relay pull in after 2 ~ 3 sec after you turn your amplifier on.

-Chris
 

birdyman

Member
2010-01-15 2:28 pm
Hi Chris

My GFA585 amp relay clicks in almost as soon as switch is turned on. Last night while fooling with amp & scope, I found little thump happens as time delay bias (constant current source to long tail pair) turns on. Before I cleaned up, internal wiring of amp was messed up and mistakenly, power connector (J108 plug) was connected to time delay bias (J107) female connector. So raw -82v was fed to Q502, R107 Q117 and counter parts on right channel. Q502,opto coupler, R107, Q117 was not cooked but when I did diode test on D117(Adcom J6), polarity testing was ok but reading was not constant ( Humm, according to Fairchild data sheet, LM336 is "Low temperature coefficient" regulator, just like J2(LM329)

If Adcom J6 is LM336 as one web site suggests, J6 is 2.49v or 5v shunt regulator and was fed to -82v (OMG!!) . My problem is that because LM336 comes with either 2.49v or 5v package and adjustable, if -82v was fed Adcom j6's (D117, D167) and cooked both, I do not have any good channel to verify correct voltage of lm336's.

This is where I need help. 1) Is J6 LM336? 2) What voltage LM336?

I am ordering LM336's (5v &2.49v) from Mouser. I am thinking of replacing one with 2.49v LM336 first and if it reduces gain in a half, I will try 5v Lm336.

Scott
 
Last edited:

birdyman

Member
2010-01-15 2:28 pm
While waiting on J6 replacements, I try to find any faulty parts on time delay/bias current circuit. All the parts except J6 were checked out to be good. Little blip of "thump" I saw on o' scope was positive spike. I am not sure condition of J6 because it passes diode test but value of resistance changes as I hold it to measure. But J2 (LM329) did not change as I hold it to measure. I will see if it is defective when I receive LM336 from Mouser.
Or mild thump on "ON" & "OFF may be normal to Adcom GFA585 (??)
 

anatech

Administrator
Paid Member
2004-06-06 8:31 pm
Georgetown, On
Hi Scott,
I've never once had to replace that vreg. I honestly don;t know what the value is, but I will make a point of testing it the next amp I get open.

If you don't mind, which internet source did you get your information from? They can either be a goldmine, or a package of guesses. I wouldn't mind look at stuff I do know to see if that is accurate.

Now, since you have the data sheet on these parts, you should have read the applications section. The easiest test for shunt type regulators is a pretty simple one. You can use a power supply (or a 9V battery), a 1 K ohm resistor and a voltmeter. Testing with an ohmmeter will only tell you if something has failed short, or a junction is severely out of normal bounds.

So, place the positive of your power source on one end of your 1 K ohm resistor, the other end of that resistor connects to the anode of the unknown device. Connect the positive of your meter here as well. The cathode connects to the negative of your power source as well as the negative terminal of your meter. The meter ought to be turned to the 10 to 20 volt range, and a power supply adjusted to about 10 VDC. Turn the supply on (or connect the battery) and read the voltage across the device. That should be the breakdown voltage. This works with shunt regulators, zener diodes and even reversed E-B transistor junctions (normally about 7 VDC but may range from 5 V to 8 V. Parts like the 2SC2878 are special and have a higher E-B breakdown voltage.). The voltage across the 1 K ohm resistor will tell you how much current you are passing through the device. 1 volt = 1 mA, so you can increase the supply voltage to attain the desired test current. Note that many zener diodes have higher break down voltages (some up to 200 VDC), so keep that in mind before you declare a part is bad because it didn't break down. Also, there are silicon diodes that break down from 12 VDC and on up. RF diodes commonly have low breakdown voltages, common power diodes are rated from 50 VDC (except Schottky which can be lower) on up to 1 KV or more.

If your connections are good, those measurements should be stable. Let us know what you find.

Now, my own feeling is that your part has a 2.5 V breakdown. It's used as the voltage reference in a current source. To find out how much current it's set for, subtract about 0.6 VDC from the breakdown voltage and divide that by the value of the emitter resistor. For example, a red LED and 1,200 ohm emitter resistor using an NPN transistor will generate something close to a 1 mA current. The LED is close to 1.8 V, the E-B junction of the transistor is 0.6 V. So, 1.8V - 0.6V is 1.2 V. Divide that by 1,200 ohms and you get .... 1 mA. Try that exercise with your two possible voltages. Also look at how much power the transistor will dissipate. That is the voltage from collector to emitter, multiplied by the current you figured out. Sometimes it's pretty clear there can only be one answer.

-Chris

Edit: A mild thump is perfectly normal for any amplifier that does not use a speaker relay. So, normal for your Adcom.
 
Last edited: