Help with Direct coupled amp

I would start by removing the output transistors and checking the power supply. It could be the bridge rectifier has gone and is allowing AC into the amp.

ng

re

It could be a driver stage fault. If both transistors are switched on at once that would blow up the output transistors.

You need to check all the driver stage transistors to seee if they are blown

having

sure

DC coupled amps are pigs. You may find, like the Audiolab 8000A (non)repair brief on my file

it is due to high level high f oscillation. I would start by putting some temporary dropper Rs on the replacenment op devices to protct them, while tracking down the causitive problem

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

third reply attempt

Then secondly Looking at the schematic I would change all the electros around the TDA/s

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

Do you have access to a variac ? It makes repair of problems like this, a

*great* deal easier. After replacing output transistors, I always use one to bring up the supply rails gently to a point where I can take measurements as to what is going on, without risking the new ones releasing their magic smoke ...

Arfa

I use high wattage bulbs that closely match the proper load... if something shorts, the incandescent lamp just currents the load into it self and displays bright to indicate a problem. It works great if you have a sudden unexpected short when you think things are up a running. THat wouldn't work so well with the variac I wouldn't think.

--
"I'd rather have a bottle in front of me than a frontal lobotomy"


http://webpages.charter.net/jamie_5"

Then I would have to respectfully suggest that you are perhaps not fully understanding the use of a variac for this type of work ? A variac will allow you to apply power to the unit in a staged and controlled manner. Assuming a linear power supply, as opposed to a switcher, this will result in reduced, and carefully controlled rails being applied to the output stages. So, if your amp uses +/- 50v rails on its outputs, then by applying

10% of the 'normal' input voltage - say 11v or 23v depending which side of the pond you are, your power supply will deliver +/- 5v to the output stages. This is enough to measure what is going on ie collector volts, emitter volts, midpoint offset, base drive volts and so on, without enough current being available to fry the new transistors if there is still a problem.

Once you have established that there are going to be no immediate disasters, you can then ramp up the input volts say another 10%, and take your measurements again. All the time, you can also monitor the output devices for unexpected temperature increase. With 20% input, the preamps will normally be working pretty well, and you can apply signal, and use your 'scope to get an even better idea of what is going on. Controlling the input power in such a simple way, would not be so easy with light bulbs, I venture to suggest ?

Bear in mind also, that a light bulb has a very low filament resistance when it is cold, and will thus be nothing like its expected power rating at that time. It will have the potential to allow a large surge current into the amp for a few mS, which may be long enough to do damage again. I have, however, seen series light bulbs used to great effect to facilitate the repair of SMPSs in TV sets, where it is common to have front-end shorts that knock power line fuses straight out.

I have been repairing this stuff for many years, and have lots of friends and colleagues in the trade, and I would say that amongst pretty much all of them, the 'weapon of choice' for assisting with the repair of DC coupled amps, is the variac.

Arfa

message

a

one

applying

disasters,

input

venture

when

that

amp

however,

of

But not if it was due to intermittant high level, high f oscillation initiated by mains spike say

How would you go about finding such a source of oscillation, in general terms, not necessarily in this thread of Wurlitzer amp ? eg Audiolab 8000A (non)repair brief on my file

Stereo amp and all DC levels on both channels matched after replacement but

3 hours into dummy load test , catastrophic oscillation again , in one channel

just in case I come across another such DC coupled amp

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

You will never see high current surges, as the supply caps will severly limit voltage. I use a lamb box every time I work on amps, and i used to use it on monitors. My lamp box has a three switched tier to select approximate currents. Its a great aid. Of course, sometimes i use both a variac and a lamp box.

greg

Well, I guess that no particular repair aid or technique is going to be appropriate for *every* fault scenario. At this point in the OP's investigation, there is no reasonable indication that the failure is due to a channel hooting at hf, although I would concede that it is possible. I have had cases of output stage failure, which have been attributable to such a functional anomaly, but I reckon that in over 30 years of repairing this stuff, I could probably count on one hand, the number where it has been the cause, and variac repair would not have helped, as opposed to literally hundreds of 'conventional' output failures, where a variac has been appropriate.

As far as finding such a problem, if it is as intermittent as you are indicating on the Audiolab, then I'm not sure that there is any real way of tracking the cause down, other than by experience, and gut feeling. I think that I would probably start by having reduced supply volts and seeing if I could provoke it with a dry finger poked around the back of the board. If not, I would probably next check / replace the components in the Zobel network, and then replace any very low value caps (10 - 1000pF) that might be dotted around, particularly in the earlier preamp stages. I might then start adding caps of 100pF or so across B-E junctions of transistors, or possibly between base and ground. I might also replace any electrolytic bypass caps in transistor emitter circuits, and possibly shunt any such with an additional cap of 470pF or so. Likewise with any rail decouplers.

Beyond that, it is highly likely that the job would become commercially unviable, and I would bow out gracefully ...

Arfa

to

such

the

of

think

with

I have to agree with that DC coupled amp repair "flow chart". I did not want the OP to use variac without being aware of the hf instability problem and blowing another set of devices. I think my only experience of a DC coupled amp repair was that Audiolab, so mentally making a 1 to 1 connection.

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

I'm really surprised at that, since DC coupling rather than AC, has been the norm for 25 years now ??

Arfa

ng

re

Wow, did you draw this yourself? %^)

I would start at the outputs, test them, then work backwards testing all semiconductors, until you stop finding defective ones. Heck, since you've got it apart, test all semi's after C243. Also, if R220 is a output stage bias current adjustment and this thing is old, make sure it isn't "noisy" (make sure it yields a smooth variable resistance moving the wiper end-to-end.) If it isn't smooth and happens to be on a particularly intermittent spot, it could be failing to set bias properly, which could drive everything full on, especially in a direct-coupled amp.

Best of luck.

On Sun, 6 Jan 2008 13:56:09 -0800 (PST), Marra put finger to keyboard and composed:

Why not temporarily insert a 100 ohm 20W resistor (aluminium clad?) into the collector circuit of each output transistor (BD647/BD648) and then scope the various stages to see if any are biased incorrectly?

- Franc Zabkar

--
Please remove one 'i' from my address when replying by email.

I was tearing my hair out trying to chase down a bias problem in an amp, turned out the bias pot itself was bad, in the future I'll always replace that if in doubt.

I fully understand greatly, Did you not see my "unexpected short" comment? ? With a variac you can have it at full operation if you make it that far and let it sit there and then comes along an unexpected short! I am not talking about those shorts/unwanted events that are there as you're ramping up the voltage!..

every one has their own style, I'll stick with mine cause it has served me well thank you.

>

--
"I'd rather have a bottle in front of me than a frontal lobotomy"

"Daily Thought:

  SOME PEOPLE ARE LIKE SLINKIES. NOT REALLY GOOD FOR ANYTHING BUT
  THEY BRING A SMILE TO YOUR FACE WHEN PUSHED DOWN THE STAIRS.
http://webpages.charter.net/jamie_5"

OK. Understood. I *was* talking about the unexpected events as you wind the input up, as in my experience, these are far more common, due to faulty components that have been 'missed' during the diagnostic procedure, than cases where the amp works ok for some time, then suddenly fails again. But I do take your point, that in such instances, a variac would be less use than a light bulb.

Arfa