Traynor YGM-3 (early) - Selmer cab

[Barks]

OK, about to put in an order to get the bias trim, fizz cap change and grid resistors etc - so have a few questions.

What should the minimum wattage be for resistors in the amp - or does it depend on where in the circuit they are used?

I've also looked at the two blue electrolytic caps closley -C27 and C28 on the older scheme posted above, to see about replacing them - and it turns out they are not stock as the values are different.
I am assuming they have been replaced around the same time the two 500v caps were replaced.

Anyway, the stated values should be 8mfd 250v and 64mfd 64v. The ones on the amp are 10mfd 300v and 100mfd 63v respectivley. Does this seem ok or should I think about getting ones closer to spec?

thanks

Barks

[martin manning]

Did the adding the 120pF cap solve your problem? if not then you'll need to look elsewhere. If so, it would be good to have a few different values to try there, usually something from 47 to 150pF is used, at a minimum 500V rating.

For the bias supply electrolytics, the higher voltage rating isn't a problem, but higher capacitance is. Larger values will provide better filtering and possibly lower noise (hum), but because the bias and HT circuits are switched on simultaneously, the power valve current will likely spike on start-up. This is because the bias caps will take longer to charge through the 150k resistor coming off the HT. I don't think having a 100uF in there instead of a 63uF is anything to worry about though, and you have years of operation at higher anode voltage to back that up.

Do you know how long the existing electrolytic caps have been in it? I would play it as-is until you decide to replace all of them.

Resistor power ratings are selected for the particular circuit location, but in general 1/2W is fine for everything except power supply dropping resistors and screen resistors. Larger ratings are often selected simply because the parts are physically more robust. If you want to experiment with adding screen resistors I would use 2W wire wound, and something between 100 and 470 ohms.

For the bias trimmer, look for a Piher 15mm round type (probably the largest physical size available), and you may have to get a little creative to mount it to the board, using the resistor lead to short the wiper to one leg, and to bridge the gap to the unused eyelet.

[Barks]

Many thanks. I think the caps were replaced in the last 15 years at the least. There is more hum induced by the single coills than in the amp itself!

Trying the little 120 Silver Mica this afternoon to see if it makes a difference to the fizz - hope so!
Will report back.

Afterward making the order for everything so far with correct ceramics etc.

[Barks]

couple quick vids.

Stock with 47pf in the PI



With the 120pf in the PI

[rp]

I can hear a pretty big difference even w/ mp3s and MBP speakers. The latter might be more pleasant, but I like the bright vicious snarling punk rock sound of the first. Might have to build a +400V EL84 eater one day.

[martin manning]

I'll have to listen later. Is the second clip 47+120 or just the 120? Did the fizz you were hearing go away?

[Barks]

No, second clip was just 120.
It did seem a little smoother.

I'm now trying to trace a fault that I've induced from taking it apart too many times! Grrr.

My wife is very patient.

[Barks]

Was messing with the bright cap on the volume pot and somehow managed to plop a tiny bit of solder into the pot itself which caused an intermittent short. Amp was 3rd volume and very bright but would fatten when you hit the strings hard.
Sorted after complete strip and rebuild. Long day for tired eyes.

[Barks]

Ok. Bias trim components and grid/screen resistors on order.

Now to some diagnostic help if possible?

The amp intermittently now "farts" even at half volume on the lower notes. Sometimes when I attack a note, sometimes a little when decaying. It seems to occur around the low E string on the G or A note.

At first I thought it may have been an EL84 valve base issue being triggered by structural vibration at these resonant frequencies as one really crackled badly when you moved it - but that does not seem to have corrected it.

I don't think it's a cold solder joint as it does not do it when you whack the top of the amp when sustaining a note or a chord, even at volume.

Any ideas?
Is it possible this is related to the high bias?

thanks again!

[martin manning]

Might be an EL84 going south, with some resonant shaking of the electrodes. I'd say try another set of valves, but I think I'd wait until you get the bias adjusted.

[rp]

This is the nature of old things including people and why at 55 I always turn down most "cures" from doctors. Try and fix one damn thing and you cause an other problem somewhere else

Hard to know at this point as you did more than just swap the PT. Best one thing at time, test, and give it a few hours playing then mess w/ something else. Tweak just one stage at a time then test. If you make a mistake you'll at least know the stage, which is knowing a lot especially for us sophomores, at the very worst you can shotgun the few parts in that stage. Never move wires around in an amp that is working well too, try and keep everything the same. You took before and after picts which is helpful. I've never been able to save a damaged pot and have it last for more than a brief while, but I doubt that's the problem. The bright cap across the pot has less effect as you turn up, so if it's too bright or fizzy when cranked tweaking that won't help.

2XEL84s always sound fizzy and raspy to me. Low voltage ~280-300 helps a bit. The only way to get rid of that thin raspiness is to use 4XEL84. On my Rocket I was going to split the cathode resistor to allow pulling a pair, then I thought best to have it exist only as a 4XEL84 amp.

Farting could be a tube not sitting well in a socket or oxidized pins, work them in and out a few times but I'm guessing you tried that. Noval minis seem to sit tight but el84s are too heavy and they really need spring clips. If sockets are not riveted you should add these to your next order of e-parts: https://www.tubesandmore.com/products/S-H144

I'm really amazed those tubes are running 420V at ~100% with no screen resistors and not shooting into the air like bottle rockets. Good tubes.

FYI Here's some talk on running EL84s hot:
http://www.diyaudio.com/forums/tubes-va ... i-amp.html

Hey Martin that pict with the drawn in trimmer and resistor is worthy of TAG employ of the month award!

[martin manning]

Thanks, rp. The EL84's have spring clips on them, see the pictures in the first post in this thread. I'm with you on their holding up under these harsh conditions, not to mention they were previously running 30V higher on the plates and 7.5V on the heaters!

Seems like the sound has improved some, and it will probably get a bit better with a more reasonable bias and the addition of screen resistors. Just guessing, but there might be some ringing in the power stage that can be smoothed.

Barks, you'll have to come up with some way to support the screen resistors, which you will want to solder close to the screen pin on the EL84 sockets (pin 9).

[Barks]

Thanks.

Yes valves have spring clips. This thing is pretty robust and has had heavy use and a lot of lugging in and out of cars etc. As for the robustness of the tubes, I never had a failure until recently. When I was using it 24/7, I only knew it was time to replace after about 9 months when the edge/clarity would compress more at high volume. It was always Sovtek back then.

Parts should arrive today - so stage one will be setting up the bias trim circuit. That'll be my next learning curve.
I've just got some breadboarding circuit board, so I may devise a standoff on that to secure the trim pot, so it stands above the current 22k bias resistor.

I'll try another pair of older valves that I know work fine - and see if the farting/crackle/temp issue comes back.

These TADs are new, and it won't be the first time one has gone a bit weird from a new matched pair. Last time was JJ's within half a day last year, when I went back to an old pair of Sovtek's I had kicking about.

[Reeltarded]

Not a tube. The place where you soldered last, and the time before that.
I can't tell any different in fizz caps on a phone, but there are better plce to do such things..

Some of that fizz is bite. That is the reason to play a tube amp, you mook.

lol

[Barks]

Mook!

Good, point.

I'll check closely around the PI 47pf cap in case there is a dry joint there.
Already gone over the vol control and brite cap.