My Heathkit AR-15 lapses in and out of distortion

[haskellbob]

Does anyone have any suggestions about where to look in investigating the reason why this 40 year old receiver lapses in and out of distortion?

There are some enormous capacitors across the output transistors, as well as a huge one on the power supply. Are they suspects, and how are they tested?

Or could the problem be worn-out transistors?

Clues will be most welcome. I have a sentimental attachment to this unit and want to be able to use it. But I can't deal with having the music get distorted every so often. What's causing it?

Thanks,

Bob H.

A schematic can be consulted at:

http://www.mcmlv.org/Archive/HiFi/Heath_AR15.pdf

Sometimes it's just one channel; sometimes it's both. And the channel involved seems to vary from left to right. Which seems to me to suggest those output capacitors. They are 4000 mfd babies.

The circuit description says "Driver transistors Q204 and Q205, and output transistors Q206 and Q207, form a quasi-complementary output circuit. In this circuit, a positive-going signal applied to the base of Q204 causes it to conduct. This, in turn, causes Q206 to conduct, WHICH CAUSES THE VOLTAGE ACROSS CAPACITOR 223 and the speaker voice coil to increase. Negative-going signals cause Q205 and Q207 to conduct, DECREASING THE VOLTAGE ACROSS CAPACITOR 223 (4000 mfd) and the speaker voice coil. This alternating increase and decrease of voltage across the speaker voice coil converts the electrical signal back to sound."

[vtech]

Bob,

Before condeming any device, best to attempt to narrow it further.
Based on the age & use of mechanical switches/controls, first I would try spray cleaning ALL of them as the metal tarnish can cause all sorts of noise.
--Also how about the speakers & wiring?

Your suspect capacitors are more than likely ok. In general, High value electrolytics tend to lose value/dryout or develop leakage over time & mainly affect the output level--not generate noise or cause distortion?

Are you getting the noise regardless of your source? (tuner/tape/aux). Chances of having cold solder joints, noisy preamp stage &/or driver transistor, power supply or tuner/IF ckt are alot more feasible.

Measurements & scope monitoring are the next step...

[haskellbob]

Speakers and wiring are good. The distortion is there regardless of input - whether internal (tuner) or external...

I will try cleaning all the controls, although none of them produce that scratchy sound associated with dirty controls. It's my understanding that they were all cleaned by the guy who sold it to me on E-bay...

What are the specific implications of possible problems re:

"noisy preamp stage &/or driver transistor, power supply"? I.e., what would I look for?

I don't have a scope, though I do have a multimeter. Do you mean use that and measure voltages? Specifically, the voltages at the driver transistors and in the preamp section? At least I have the schematic, and they're all discrete components...

If I did have a scope, what would I do?

Amateurish questions, I know. It's a kind of psychological challenge for me to get a fully-functional AR-15 going. My dad had one and it's sort of a symbol of adulthood and competence and power!

Also, like all challenges one makes to oneself that are beyond his present knowledge, it's trying and frustrating to me not to just KNOW how to fix it.

But, you learn, right?...
Thanks for your suggestions. I wish I hadn't gone ahead and ordered new caps... can't hurt to put them in, though, I guess...

Bob

[vtech]

It would definitely help to be able to see thru the ckt with scope but measuring DC levels can be helpful.
How often do you hear the distortion & is it signal related? (meaning that if you were to set it to lets say aux & crank the volume with NO source signal; do you notice any noise or distortion?

With power supply, being common to both channels, verify you DC levels (14~21) & make sure none is on the border or perhaps buckling under load. Any of the 'lytics in power supply are suspect.
Also check bias levels around Q201~Q205 (l&R)

Another usefull tool is a can of freeze spray to spot spray on various transistors/caps/diodes/etc that will usually reveal noisy junctions.... rather unlikely to have it on both channels but it is possible.

[haskellbob]

In the Troubleshooting section of the manual, shorted zener diodes across the driver transistors (in each channel there are two, between the emitter of one driver and the base of the other) are listed as a possible cause of distortion at high volumes.

Now, I tried to measure the resistance across all four zeners - there are two for each channel, hooked up in push-pull or opposite directions - hope I'm being clear - and I get NO READING in either direction on the multimeter set for the highest ohms level it reads. The zeners are not conducting at all. Wouldn't an ohmmeter set for 20K ohms or whatever be able to detect current flow across a zener diode in at least one direction?

I think these are burnt out and this is the cause of the distortion.

Again, the schematic can be consulted at http://www.mcmlv.org/Archive/HiFi/Heath_AR15.pdf.

Does that sound reasonable to you?

The only flaw in my reasoning is that maybe zener diodes can't be measured with a traditional ohmmeter? But why wouldn't they?

There must've been a short circuit that caused them buggers to burn out. I found the NTE replacement part number and they're not expensive.

Bob H.

[vtech]

While you sound interested, may want to brush up on the basics a bit; Also , the terms "shorted" & "burned out" are NOT the same. A shorted diode is like a piece of wire (closed circuit) & a burned out diode is like a broken wire(open circuit)

The only function of the back to back zeners in question,(with a cap in series) are strictly to prevent/quench spikes in audio signal(AC). Otherwise, it will act as an open switch & electronically invisible. A zener is like a crowbar switch which will conduct only when the signal across reaches it's rating.
Now if they were shorted, (meaning like a piece of wire--NOT burned out), it could clip the signal & cause distortion. When you are not reading anything with a meter, they are NOT shorted thereby can't possibly cause a problem.

[haskellbob]

Yes, I thought of that. They're not shorted; they're burnt out. And I thought I'd found the problem!

I've got to brush up, as you say, on the basics.

Thanks for the tips.

Bob