Fooling yourself

[Enzo]

Got one of those damned Marshall Mode 4 heads in here - the MF350. Likes to blow up sets of output power modules. I have it finally running. I am very careful to bring it up gently on teh variac al the time so as not to stress anything until i get it squared away.

Last night I finally have it running, and I find the DFX is not running. No reverb. Maybe it only works with the FS? Nah. Stupid DFX is out, I was sure.

But after I bolted it together to burn a while, I noticed the FX were back.

Then I realized the DFX ibeing a digital system does not like a slow variac power up. The power on reset circuit cannot work that way. SO the little computer would not start its program.

Variac - no boot up. Flip on the switch - works fine.

Just goes to show that even the simplest techniques are not foolproof. Never assume.

[Amp Kat]

Hate those damn digital amps like that one and the Cyber Twin and some others. It could be worst though you see I could have been working on it and that happened and then I'd be emailing you wondering what the hell was going on and we'd both be pulling our hair out. Glad it was you and not me

[drewl]

I've been lucky and haven't had to work on one of those yet.......don't think I want to either!

[sportster4eva]

I've seen amps that produce a large DC offset at the output coming up slow on a variac. QSC and Peavey come to mind for some reason. Maybe the current limiter approach is an alternative for this type of situation.

[Enzo]

I still like the variac. If it is going to draw, I can tell at very low mains voltages, and I maintain that is even less stressful than having to conduct the bulb current.

[schraedertr]

Similar problem with AVT150

I was bringing an AVT150 up on the variac after replacing current limiting resistors and voltage regulators in +/- 15V power section. I could get a signal through the "Acoustic" channel but none of the other three.

Since I hadn't found the reason for the problem with the +/- 15V, I was preparing for a long involved session (and a search for a schematic). I had been monitoring voltage across the resistors and nothing looked bad, so I went ahead and brought the voltage on up. I figured I'd see smoke somewhere, but nothing and no excessive current.

Anyway, after bringing the voltage up to normal, the other three channels starting working. I thought at the time it was probably the single preamp tube that wasn't passing signal, but I think you're right in fingering the 5V digital effects.

Anyway, it's up and running and I'm still not sure if I only fixed a symtem and not the underlying problem.

[Euthymia]

There's also the variation that killed my 80W bench speaker a few years ago:

Ampeg bass amp with SS output section, bringing it up on the variac, all is going well, then a click from the PCB and a sizzle from the voice coil of the speaker.

Turns out that there's a relay on the amplifier output that doesn't close until it sees enough voltage, and when it did, there was already enough voltage to roast the output section.

My solution to this (and other design quirks of the SS Crampegs) is to refuse to service SS Ampeg bass amps.

[Alex R]

Yeah those relays don't mix well with the light bulb limiter either. Very hard to chase shorts/test rebuilt amps with one of those around (but see below). Not only ampegs have them, I can't recollect which amps but I've experienced the same problem in other makes of high-watt SS bass amps all right. I guess the answer is to short out the relay and use the limiter...?...though that is only a guess, and my answer has been sometimes to check really carefully around the rebuilt board for shorts then put an ammeter in the mains supply, switch on and watch/listen/sniff hard, with one finger on the off switch and all the others crossed.

What are those relays actually for, by the way?

Personally I much prefer the light bulb to the variac, because it interacts with a big current draw/short in a safe and useful way, and gives you a nice clear warning. Variacs don't limit current in that way, and lower voltages encourage higher current draw, which is why I'm cautious around variacs. If I'm feeling really cautious I start with a low-watt bulb. With a 200 watts of light bulb in, the relays we're talking about will sometimes open. But if anyone can instruct me on the use of the variac to test amps that have current draw problems &/or tiresome relays I'd be willing to learn as always. Does the light bulb PLUS the variac give you the best of both worlds?

Enough questions already.

[R ski]

Another criteria I hate about the digital world. If bringing up the supply slow, some digital memory chips can get screw up from that process.

Easier said than done, I usually watch out for that issue. Can take quiet a while to isolate that voltage.

And last but not least, switch mode power supplies, really can pour out the current on short demand, nasty things happen with defective components.

[Enzo]

The secret is simple, never connect a load to the amp until you KNOW the output is not going DC on you.

Without a load, the relay is just a switch to nowhere. The relay waits a moment after detecting power and then turns on. It is there to prevent the power on transient from launching the speaker cones into the audience.

CLip your scope to the amp side of the relay and then it matters not if it pulls in or not. On the typical SS amp, ther are a bunch of cement ballast resistors between the output devices and the output bus, so clip onto one of those as a handy spot.

Even if there is DC on the output, the thing won't draw current unless a load is there to draw it. If both polarities are conducting, you don't get DC, you just get blown fuses.

I never connect a load to a blown SS amp until I am secure there is no DC.

If you don't have $100 for a variac, the bulb is a nice tool, but I will take a variac and ammeter every day over the bulb. Without an ammeter, the variac is only half a tool. No, it does not limit current, but I can advance the mains up to 10 or 15 volts, and if current is already ramping up, I can back off and nothing was ever stressed. The bulb way means at least the bulb current was flowing through the amp. And a 200 watt bulb on UK mains means a whole ampere was flowing through the unit under test.

SOme amps have a momentary instability at low mains, but not most, and experience tells me if it is that or a problem. I am not sure what other low voltage situations draw excess current.

[Alex R]

Thanks Enzo for the answers. Understood about the relays on the output - not sure whether I remember an amp with one between preamp and amp. There for speaker protection then; ok.

I will experiment some more with the scary ol' variac. Thanks also for the support with the simple maths about the 200w bulb, clearly my dyscalculia is getting worse along with my focal length.

Ignorant question: if the (ss) amp is (say) idling too hot, won't the output section draw current without a load, just straight through the collector-emitter circuits?

[Enzo]

Yes, but it won't send the speaker cones towards the next town.

Excess current in an operating amp is often visible on my current meter. If a SS power amp idles at 150 watts for example, it is working way too hard. I expect under 50 watts. SO right away I look at idle bias.

I am not trying to talk anyone out of the bulb. If you have success using the technique, that is what counts. Just stating my reasons for my preferences.

In fact, idle current is the cheap and dirty way to bias a SS amp. If it is biased too cold, then the signal has to move a bit before it turns on an output. You get crossover distortion. If the bias is too hot, both sides conduct so it gets hot fast. No crossover distortion, but hot. Below the proper bias level, the outputs are off. You can turn it colder and colder, and it won't affect mains draw since the xstrs are off.

SO turn the bias control higher until the mains draw JUST starts to climb. That is the point that both sides start to conduct at the same time. Now back off just a hair. voila, the amp is adjusted.

In fact while servicing amps I sometimes turn the bias way cold just so no other issues upset my game plan. Once the amp is working right, I can tweak the bias back where it belongs in seconds.

No, I have never seen a relay between preamp and power amp, but I have seen preamps with their own output mutes. Keeps them from sending their own loud pulse on to the power amp.

And that is why the cardinal rule in PA sound reinforcement systems is the power amp is LAST ON/FIRST OFF.

[Alex R]

- this is really good advice about putting an ammeter in a solid state amp mains supply. I read it here a while ago and wired plugs for the ammeter into my supply box (the one with the light bulb on top) - the effect is very marked and it's easy to put a good safe quick bias on solid state amps that way & get them idling right. One of those things you quickly get used to what to expect, really speeded me up. I'm sure it was you that told me back then, Enzo, so many thanks for that and the method is thoroughly recommended...