What to do when you suspect a relay is faulty?

[leydenjar]

What's the best way to check for a faulty relay?

Like the common Omron types in Peavey and Mesa amps?

Thanks,
jar

[Dave Mac Farlane]

I realize this may be terribly late, but the obvious answer is - replace it even if its only suspect.
They're not expensive, and they have a limited life of x thousand switchings anyway, so it's a bit like asking how long a car will last.
They are very dependable and good these days especially, but my first statement stands.
Dave

[leydenjar]

Maybe I'm cheap but the Omrons cost around 5 bucks here in the states + shipping.

Are there cheap substitutes?

[km6xz]

I am of the firm belief that parts should not be replaced unless proven unsuitable for the task. Shotgunning is bad business, financially and ethically, but even more, it is wrong technically. It introduced more variables that can make diagnosis more difficult and more ambiguous.
A relay is a switch and is tested the same way, measuring milli-ohm "on" values and see if they are within the range expected. A simpler initial test is to just observe input versus output signal, and coil voltage. By connecting one scope lead to the signal out and another to the coil so you can watch if there is noise or bounce at transitions. It is pretty rare that a relay is bad but if is, it is pretty obvious. Either the coil opens or the contacts gradually become pitted or corroded and develop on-resistance that rapidly increased more pitting or oxidation, all resulting in the symptoms of intermittent signal or intermittent distortion at low levels.
If the underside of the pc board is accessible, bridging the suspect contacts can verify that the contacts are intermittent if the distortion or signal loss goes away when the relay is activated.
It is most easily seen with low level signals, a signal of a volt or so is usually enough to punch through an oxide layer of 6 molecules or so thickness(if it can't no contact, relay, pot or switch would be very reliable). An intermittent signal problem with a relay would troublesome with low level signals.
It is much faster to focus on the symptoms and build a diagnosis that accounts for all the symptoms than to swap out parts a lot of parts so the customer pays for both unneeded parts and unneeded bench time.
Good luck!

[Dave Mac Farlane]

Nah, but they probably are cheaper here in Scotland.
I must say, though, if you go to an electrical supplier rather than electronics, you might get a better deal. I've no idea what retail and wholesale means in America, but I always use my trading name for anything I want to buy. You may even have an electrician friend or in the family, who would get it trade discount. If I thought you could save anything, I would post one to you, but that's probably an expensive inroad, what with excise charges etc.
Just look at it this way - what is that cost in relation to the amp's value? Can you solder it?
I agree with km6xz, but if this is an intermittent fault, it forgives a shotgun component, especially if the cost is only the component, and realays are one of the biggest culprits, next to dry loints, of the dreaded intermittent fault. I’m sure in his long career, he can give us examples.
Dave

[Enzo]

Amen,

First off, the question is vague. Faulty relay in what sense? The relay is indeed nothing more than a switch, it just has a little magnet coil inside to move the switch. So do you mean your relay might be unable to pass a signal through its switch contacts? or do you mean the thing is not able to switch from A to B, as in the thing mechanically won;t move. Like asking about your car "won;t go." Won;t start? or won;t go into gear? etc.

The solder side of the board may be hard to get at, but in most cases, every point on a relay connects to something else. If I can;t get to pin 3 of a relay, chances are I can find the resistor it is wired to and stick my meter/scope there. If contacts are closed and signal is on one side but not the other, then that relay contact is not working. And speaker relays sometimes weld themselves shut, so it is in the "ON" position regardless of the magnet coil. An ohm meter will tell you that.

And the coil part? That is easy. The coil either has continuity or not. There is usually a diode across the coil, and that diode will be right next to it. Can't get at the coil pins underneath? Go to the diode. Got relay supply voltage at one end? Good, when the relay is OK, there should then be the same voltage at the other end. In most cases, a relay is wired between a power supply and ground, with a switch in series to turn it off or on. The switch might be a transistor of course.

If I have it narrowed down to the relay, and I just can;t get it to fail, but am pretty sure, OK, I'll go ahead and replace it. I charge by the hour, and parts are cheaper than labor. But if I am trying to figure out why a channel doesn;t work or something, or I "heard on the internet" that it might be a relay, then no, I am not going to throw relays at the thing in hopes it was the problem. I will find out that it is indeed the relay first.

[Dave Mac Farlane]

Absolutely Enzo, but I left a Steel manufacturers site, having replaced what was definitely a faulty three term controller, which I knew to be faulty. The circuit didn't switch once on test, and that incident crept into the background, in the glory of restoring this huge production animal to function. I stayed with the plant for 36 hours, for conditions were different through the night than through the day, and left, feeling satisfied that everything was ok.
Three days later, I had to return to a 'glitch' which had kept extra guys on watch.
I recalled my 'incident' and looked at this area. On examining the suspect relay, It looked as if the contacts were gummed together when asked to release. The relay controlled a circuit which 'looked' at a sporadic condition.
I followed this hunch, and hoped I was not imagining the delay associated with the action I could see through the coil.
End result - success, yet I could not get that relay to fail on the bench - still put the darned thing in the bin.
I always localize the fault - but intermittent? worst nightmare.
Dave

[Enzo]

Well sure. If you are doing field service on an industrial system where downtime costs thousands of dollars a second, you do things different from fixing an amp in the basement. I honed my troubleshooting skills in field service, and I like to think I am pretty resourceful, so I appreciate what you describe. If you know a system well enough to know a part "just has to be" where the problem lies, even if you cannot cause or detect a failure there, then replace it.

Looking at the OP though, the question suggests a basic lack of experience testing the parts/systems. So in this case, the part is more likely suspect because "maybe this" or "somebody said" or something, rather than "just has to be." SO instruction on some test approaches seems appropriate.


And no experienced tech would ever argue that intermittants aren't the toughest nut to crack.

[Steve Conner]

The usual problem with small signal relays in guitar amps is that the silver contacts tarnish from air pollution, causing a distorted signal that eventually cuts out altogether. They really should use hermetic relays with gold contacts, if it weren't for the cost.

Give it an "Enzo Whack" and listen for crackles and intermittent signal. If it has a removable lid, take it off and inspect/clean the contacts (squeeze a piece of paper between them and rub it around)

Test continuity, good relay contacts should be fractions of an ohm.

I didn't know we had any steelworks left in Scotland.

[Dave Mac Farlane]

Absolutement, Enzo, but I thought it would be impolite to suggest our chappy was inexperienced, and I suppose ole' geezers like you or me would have their teeth thoroughly in the problem before we put it forward for diagnostics or antithesis, and as for you Connor, did ye have to show my age? I was there in Ravenscraig before Thatcher closed it.
If anyone makes any other comment about my age - I'll sue for ageism!!!
Hope you all take this as humerous - I don't want thrown out of this forum cos' I like it.
Dave.

[Enzo]

I just checked my resume. As far as I can see, it doesn't say "polite" anywhere on it.


This old geezer has been soldering for something over 55 years now. But I have to say, I really learned to troubleshoot working in relay systems. Relay logic. Talk about old.

[Enzo]

I just checked my resume. As far as I can see, it doesn't say "polite" anywhere on it.


This old geezer has been soldering for something over 55 years now. But I have to say, I really learned to troubleshoot working in relay systems. Relay logic. Talk about old.

[leydenjar]

I think the main problem I have with relays is NOT understanding the way in which they are used in these circuits typically. The relay circuits aren't typically drawn. God forbid there's an optocoupler(s) in the circuit as well. If anyone has a schematic SHOWING how these circuits are wired I'd love to see it. Most schematics that have these devices leave that part of the circuit out or only draw part of it.

I would love some help in this department;

specifically, a schematic that gives a clear understanding of the wiring of a relay/optocoupler switching circuit.

[Enzo]

How about a specific example? Give us a model name so we can all refer to the schematic. Or post the schematic.

Schematics show the electrical relation between parts. A wiring digram shows how all the connections are made. Those are not the same thing. Kinda like the difference betwwen an instruction sheet and a flow chart.

I am unclear on what part you need help on. How the relay control circuits work? Or how the signal path switching works? Sounds more like you wonder qabout the control aspect.

I'd have to disagree that relay circuits are not drawn. Since the only connection between the coil and control part and the signal switching is mechanical, the electrical circuits are unrelated. And so they often, if not usually, are found on different pages of the schematic set. So in some amp we find the signal path switches labeled RY1 and selecting between channels or something. Then two pages later on the power supply page, you will find the relay coil drive circuits. RY1 coil will be driven by some transistor or whatever.

Here is a recent thread, I described on control circuit in a Marshall in post 4.
http://music-electronics-forum.com/t30134/

Is that the sort of area you are talking about?

[Enzo]

How about a specific example? Give us a model name so we can all refer to the schematic. Or post the schematic.

Schematics show the electrical relation between parts. A wiring digram shows how all the connections are made. Those are not the same thing. Kinda like the difference betwwen an instruction sheet and a flow chart.

I am unclear on what part you need help on. How the relay control circuits work? Or how the signal path switching works? Sounds more like you wonder qabout the control aspect.

I'd have to disagree that relay circuits are not drawn. Since the only connection between the coil and control part and the signal switching is mechanical, the electrical circuits are unrelated. And so they often, if not usually, are found on different pages of the schematic set. So in some amp we find the signal path switches labeled RY1 and selecting between channels or something. Then two pages later on the power supply page, you will find the relay coil drive circuits. RY1 coil will be driven by some transistor or whatever.

Here is a recent thread, I described on control circuit in a Marshall in post 4.
http://music-electronics-forum.com/t30134/

Is that the sort of area you are talking about?