36W Lite ValvePower cage prototype crackling, thoughts ?

[stevesuk]

I have put a 36W Lite prototype together intending to fit a VVR when it is working nicely without. At the moment it has a 'crackling' break up when pushed. Normally I would look for dry joints or faulty components, it is that sort of noise. GZ34, 4 X EL84's and two ECC83's (12AX7's).
I have produced new transformers OT and PT specifically for up to 50W amps, the OT with 4K/8K/3.25K/6.5K to 4 to 32 ohm output impedance taps.
I have increased the B+ filtering to 2 X 47uF and two 22uF caps.
That helped with the crackling a lot, but it is still there when pushed.
It is very loud and a full sweetness at 2 on the volume but with a LP it breaks into crackling early, a strat not so bad. Channel 1 (low) is not as bad.
It is intended for the 18W cage customers who want two amps, the 18W for duos and small venues and also for a choice of cleanish and crunch in the same amp, once the VVR is installed. Also I have some Muso's who do everything from TV/Radio advert sessions through duo's to different bands playing lots of variations from JJ Cale to Blues to Rockabilly, Alt Country Blah Blah. They love the 18W sound when they need it, but would like to just cart one amp to get the other sounds when they need it. The 36W is ideal for this purpose.

With Graydon I was the first to finish the 36W many years ago and have built many since. But not for a few years.

There isn't really a true Lite IIb 36W schematic around with only two 12Ax7's and vol/tone so once I have cracked the crackling on this I will put the schematic up.

Any ideas ? I need to get back to building more 18W cages



Steve UK
www.valvepower.co.uk

[zaphod_phil]

In my experience crackling sometimes was a symptom of a faulty power valve. Could also be a bad joint. I doubt that it's a design issue. Others have built 36W versions of the Lite IIb without this kind of problem.

[leslie]

Sometimes bad electrolytics can cause this(yes, even if it is new cap).
If it's rather cooking I would search for bad resistor

[stevesuk]

ZP,
I went for valves first as I have heard crackling like this before with a dodgy valve, all ok.
Leslie,
Electrolytics, much more difficult, only way is substituting new for already new caps ? I suppose as adding caps on the power rail improved it I should try it.
Let you know if I come up with anything.

Steve UK
www.valvepower.co.uk

[Plexi]

Steve, i sent you an email..

[stevesuk]

Thanks to Plexi, basically the voltages are too high. Fine for a standard 18W Lite, but evidentally the 36W variant is more sensitive to the voltages that an 18W is perfectly happy with.
I have dropped the voltage with zeners and resistor for proving purposes.
Now just fine tuning to do and another variation lives.
I am particularly pleased with how quiet it is with no hum, buzz or hiss at all.
But it probably means getting 260V HV transformers put together or using my 290/300V 50W PT's and Zeners by the bucket load.

Steve UK
www.valvepower.co.uk

[zaphod_phil]

That's really strange. Is there something different about how you're powering the screen grids compared to your 18W amps?

I once came across a case where a guy had built an amp with a voltage-regulated screen grid supply, to squeeze more power out of the amp. He then got the crackling problem when the amp was cranking. This turned out to be arcing, which was happening when the anode voltage would swing a long way south of the screen voltage.

[Graydon]

I had a similar problem on a 36W build I did a couple years ago. I ended up tweaking the values in the power section but I don't remember exactly what it was. It might have been grid resistors or cathode resistors. Sorry. I haven't built many 36W amps lately.

[justinm]

In my experience with building 36 watt power sections, EL 84s like individual screen resistors. When you double the number of EL84s, they like to be balanced or even known good tubes will crackle.

I have a design that runs the plates at 370 vdc and I drop the screen voltage to 300 vdc and run individual 470 ohm screen resistors. This still allows me to share cathode resistors side to side.

My solution isn't necessarily everyone's however because
I find that the screen current to the EL84s has a direct effect on the mid-range character of the amp.

Lowering the B+ is probably the easiest solution.

Justin

[zaphod_phil]

Lowering the B+ is probably the easiest solution.

But you probably end up with a 30W amp instead of a 36W. Not that it's a huge difference....

[stevesuk]

It has been observed that the crackling can come from the GZ34 arcing.
There is definitely 'something of the night' about the 36W with a tube rectifier.
I don't remember having any of this trouble with the half a dozen 36 Watters I built 7 or 8 years ago, but they were all 2 channel TMB's if I remember, not Lites.

I am going to fit a VVR and try again, perhaps changing the VVR 100K's for another value to limit the voltage. I have tried a variac and that does it, but obviously only for a short time with deference to the heater voltages.

It is suggested the PT HV should be as low as 260V if using a GZ34.

It also seems it is not an unknown issue with a 36W Lite.
Steve UK
www.valvepower.co.uk

[zaphod_phil]

You also wonder why AC30s don't suffer from this problem, as they run at around 360V IIRC. To eliminate the possibility of the problem being caused by GZ34, you could try SS rectifiers with sag resistors.

[justinm]

Lowering the B+ is probably the easiest solution.

But you probably end up with a 30W amp instead of a 36W. Not that it's a huge difference....

True and true.

I've done some with lower B+ by using a second source PT in a pinch and they are more like 30Watts, but they sound really good. In fact, there are on going debates around here which ones sound better!
FYI- those ones get different screen voltage and screen resistor values, but they still need one for each tube.
Also- I use 5U4G rectifiers just because they are easy for me to get from EH and they don't rattle.

RE- AC30. I know- that's a mystery to me, but notice that they don't share screen resistors. Tubes ain't way they used to me either.......

Justin

[stevesuk]

It wasn't as long ago as I thought. I have been looking through my posts and it was 2004.
http://www.18watt.com/modules.php?name= ... sc&start=0
There is a lot of information on those old posts about the problems we had when we made the first 36 watters in the universe.
ZP,
Plexi suggested trying SS rectification, but that means a PT of even lower HV. Although I do have loads of 240V : 240V isolating transformers ?
Not enough room for it on the chassis though.

Steve UK
www.valvepower.co.uk

[stevesuk]

I think I have found the cause of the crackling

I changed the prototype OT for another prototype OT and the crackling is not there anymore.

I put the original back in and added the impedance selector switch switching to the lowest impedance also stops it.

So I think it was/is a mismatched output impedance.

I have never had that with an 18 watt, just less output when switching to the 'wrong' impedance.

I will give this amp and OT a thorough test and then use this prototype as the 'universal' output transformer I have been planning.

It will be useable up to 50W with 8K/4K and 6.5K/3.25K down to 32, 16, 8 and 4 ohm depending on tappings used.
This means it can be used for many 2 or 4 valve combination of EL84's, 6V6's, KT66's, KT88's, 6L6's, and EL34 /5881's and probably a few more obscure tubes.

It will also be the same footprint as the popular 18W mains transformer size.
The PT will be the same with thicker laminations, but I need to decide what voltage to manufacture them to allow a GZ34 to be used.

I am going to take zeners out one at a time from the B+ feed to see if the crackling comes back. If it doesn't, I will settle on 290V HV. Which means the PT can be used with an 18W with 'bells and whistles' because it will be 200mA.

Steve UK
www.valvepower.co.uk