Picked up a 1969 Laney Supergroup 100w. Some questions.

[greengriff]

Purely by chance this evening, I saw advertised, agreed a deal on, and collected a 1969 Laney Supergroup 100w head.

The amp is remarkably clean, looks barely used, and appears to be totally original, bar one of the EL34 valve bases (v5). It even still has really clean 'Partridge' stickers on the transformers and choke, and it came with it's original Brimar 12ax7s and 2 Mullard and 2 Pinnacle (Siemens) EL34s. The owner had put a new set of JJ EL34s in it for some reason, but supplied the old power valves with the amp.

Here's the interesting bit. Since it's over 50 years old I thought I'd give it a look over before plugging in. One of the newly fitted JJs (v4) had a white getter (failed vacuum?) but interestingly so did one of the old Mullard. Is this a coincidence I'm wondering? Could a fault, or hot running, cause a valve to lose vacuum really quickly? The reason I ask about hot running is that the metal ventilation grill in the top of the amp is quite warped, and I can't imagine it would have been that way from new. The previous owner says he hardly used the amp, and it's condition is commensurate with the claim, but the warped grill, and replacement valve base makes me think it might have a problem with extremely hot running?

I don't want to plug it in and risk causing any other problems until I get to the bottom of the issue. It's too late for me to ring the previous owner and quiz him tonight, but I will tomorrow. From the conversation with him I had when collecting the amp I got the impression that he knows nothing about the function of amps, but is a nice, honest guy. As far as I understand it he literally fitted the new valves, turned the amp on to see if it played ok, then put it up for sale. If that's the case what's the chance of vacuum failure in such a short tie period if there wasn't an issue? Small I'd imagine, but I'd certainly welcome any opinions.

He's owned the amp for over 40 years (purchased it in 1979) and said he's never had any work done, so maybe the odd valve base is an earlier replacement.

Also there appears to be nowhere to bias the amp. Am I missing something?

Anyway, any input would be very much appreciated.

[EL-34/JMP]

Bad tube is a bad tube & powering up a tube amp w/a bad power tube can be catastrophic.

[Jon Snell]

It has no separate bias control as a designated winding supplies the bias voltage through a 1N4004 diode and 12k resistor decoupled with a 10uF capacitor and loaded with a 56k resistor.
If the HT rises, the valve is compensated with a rise in negative grid bias voltage, stabilising the emissions.
Broken glass is normally down to rough handling but out of a box off 100 new Chinesium JJ EL34s, that I bought before reading where they were made, four had cracked glass from the factory and two went to air after only two hours use and people don't understand why I don't use JJs.
A common cause of broken glass is also liquid ingress through the ventilation holes at the top of the cabinet. Hot glass and a cold drink = cracked glass!

[Mick Bailey]

You can get early failures of new tubes and I noticed with one JJ it had a tiny crack in the side of the envelope. The getter had already started to show white and after a few days on the bench it had completely gone. Another time recently I had a pair of late-90s NOS Svetlana EL34s that both lost their vacuum after just a single day's use in a Matchless Independence. Replaced them and no further issues. Hot running in itself is not usually a cause of vacuum loss, but sudden thermal shocks are. The metal/oxide/glass 'alloyed' bond where the pins exit the envelope are thermally matched to exceed the maximum operating temperatures of the tube. I've had super-hot tubes where the envelope has melted and collapsed inwards but still maintained a vacuum.

By all means establish that the bias circuit is operating correctly and check the DC voltages,, but coincidences occur all the time and can sometimes lead to inaccurate cause/effect assumptions.

[greengriff]

Thanks guys, what a great forum this is. So, I checked valve by valve and it does seem to be a coincidence. The bias circuit is stock and unmolested as described by Jon above, anode voltages are around 570, but current draw from each valve is low - around 14ma, except for one of the old Siemens/Pinnacle valves which drew around 24ma. The other Siemens is in the same ballpark as the JJs (14ma). I had a similar issue with my old Marshall - with the bias trim pot maxed out it could only get to about 13ma - and the suggestion here was that it was the quality of valves that had changed over the years, yet here is a 51 year old EL34 drawing the same as a brand new one. Is there any reason not to piggyback the 56k resistor to bring the EL34s into the 30ma range? I don't want to do any soldering to to such a lovely, unspoiled old amp, so I could crocodile clip one into place?

Unbelievably all the original Erie electrolytics still read good at 51 years old. You could have knocked me down with a feather. So the amp is good to play, and play I did for a bit this morning. Despite looking very similar to a Marshall Superlead circuit wise, it sounds quite different - thicker and darker. Clean tones are a delight. I connected two 80watt rated attenuators in parallel so I could wind it up to ten, and it produces a thick, juicy crunch, not really Marshally at all. Lovely with a strat, a bit rich/dark with humbuckers for my taste. Certainly no need to jump channels. The bass channel is unusable really, just like a superlead's.

[greengriff]

One correction - the cap in the bias circuit is 15uf in this amp, even though the schematic specifies 10uf.

[Jon Snell]

Work out the dissipation; Formula is Max Anode dissipation divided by Anode voltage multiplied by desired class. A = 70%, HOT. AB = 25% COOL.
Anode voltage = 570.
Current draw = 30mA
That is running in class A and far too hot!
Should be run in class AB = Total anode dissipation 25W
so, 25 / 570 = 44mA for 100% dissipation, we are looking for 30% so 44 X 0.3 = 13mA.

15uF is fine.

[Leo_Gnardo]

Is there any reason not to piggyback the 56k resistor to bring the EL34s into the 30ma range? I don't want to do any soldering to to such a lovely, unspoiled old amp, so I could crocodile clip one into place?

You could parallel the 56K bias load R to make adjustments, no problem. Personally I'd have no problem replacing it with a series combination resistor + trimpot (preset in UK English) but do as you see fit. With 570V on the anode I wouldn't be looking to go 30 mA but you could try it for a while. Also you would expect the B+ to sag a bit as you draw more bias current in the 30 mA range, maybe to 540-550V I'm guessing. Still a challenge to modern EL34s.

FWIW I had to go for KT77 in a late 60's Laney Klipp I worked on a couple years ago, to be more assured the output tubes wouldn't go into self-destruct mode. I used JJ and they worked out OK. Perhaps QC has slipped at JJ in recent times. I've seen their tendency to fail wax & wane over the last 2 decades. And nobody I know wants to spring for the New Sensor so-called Gold Lions. I've seen failures in those too so high price & vaunted reputation have little to do with being supposedly bulletproof.

It would be good to know what's the screen grid voltage. Many of these old amps - Marshall, Orange etc - ran G2 close to the anode voltage, an exercise in brinksmanship with today's selection of new EL34s. I would recommend biasing to the cold side, but I agree 14 mA is just too chill for me also. I'd expect a sweet spot in the low 20's.

Nice find griff, tantalize us with a photo please!

[greengriff]

Here you go:

[greengriff]

You could parallel the 56K bias load R to make adjustments, no problem. Personally I'd have no problem replacing it with a series combination resistor + trimpot (preset in UK English) but do as you see fit. With 570V on the anode I wouldn't be looking to go 30 mA but you could try it for a while. Also you would expect the B+ to sag a bit as you draw more bias current in the 30 mA range, maybe to 540-550V I'm guessing. Still a challenge to modern EL34s.

Yeah, I guess as it's turret board it's a nice easy job and easily reversible later on. I notice the voltage selector is set to 250v, but my wall voltage is around 230. I'm assuming that's why the anode voltage is around 570v rather than the 600 I see quoted elsewhere? I'll aim for low 20s then. No point making it unreliable. To be fair it's not likely to get much use, as I am definitely a Marshall guy, but I couldn't resist such a nice thing.

[greengriff]

Work out the dissipation; Formula is Max Anode dissipation divided by Anode voltage multiplied by desired class. A = 70%, HOT. AB = 25% COOL.
Anode voltage = 570.
Current draw = 30mA
That is running in class A and far too hot!
Should be run in class AB = Total anode dissipation 25W
so, 25 / 570 = 44mA for 100% dissipation, we are looking for 30% so 44 X 0.3 = 13mA.

15uF is fine.

Only just saw this reply. Thank you! So it's running just fine then! Forgive me I was thinking 30ma as all I've ever owned is Marshalls.

[Jon Snell]

Only just saw this reply. Thank you! So it's running just fine then! Forgive me I was thinking 30ma as all I've ever owned is Marshalls.

Pleased to help.

[Chuck H]

Boy I really hate those bare HV jumpers! But Laney does that. Just be careful working in there.

[greengriff]

It's all back together now. No need for me to fiddle anymore, just to play it. I have to say, the wiring looks like it was done with more care than some old Marshalls I've had.

[Helmholtz]

Work out the dissipation; Formula is Max Anode dissipation divided by Anode voltage multiplied by desired class. A = 70%, HOT. AB = 25% COOL.
Anode voltage = 570.
Current draw = 30mA
That is running in class A and far too hot!
Should be run in class AB = Total anode dissipation 25W
so, 25 / 570 = 44mA for 100% dissipation, we are looking for 30% so 44 X 0.3 = 13mA.

There is no fixed relation between tube idle dissipation and operation class. Fixed bias class AB amps typically run between 50% and 80% plate dissipation. Cathode biased class AB amps are often biased at 100% or more (e.g. Vox AC30).

The Mullard EL34 datasheet specifies an anode idle current of 25mA with 800V anode voltage, corresponding to Pa= 20W or 80% dissipation.
So I think Iao = 25mA should be safe with Va = 570V, meaning a dissipation of 57%.

Much lower idle current than that while being even safer but will result in class B and crossover distortion..