Tubes that can handle 500VDC on screens, or tame the power supply??

[cerrem]

For all intents and purposes, it is basically a plexi/2203 output section, with one exception. 1k-5W dropping resistors, also choke is in the same area, 115 ohms if I recall correctly. However, the exception here is that I have a 2k plate load, not 1.7 like a typical 100 watt marshall.

Having 2k for plate load is perfectly fine that should not be an issue...
I would suggest confirming the transformer color codes and make sure the solder terminals on the ohm selector line up with what the selector is indicating....
When the EL34 tubes fail, where is the crack located ??

[Gtr0]

Having 2k for plate load is perfectly fine that should not be an issue...
I would suggest confirming the transformer color codes and make sure the solder terminals on the ohm selector line up with what the selector is indicating....
When the EL34 tubes fail, where is the crack located ??

I have not ever been able to find a crack, see the following image...


As you can see the tube has whited out, and there is a burn at the base... this is near pins 7 and 8. Heater to Cathode violation?? In another case, the tube was white and I was unable to find any cracks or burns... but when they go down there is usually an accompanying flashing inside the tube with it. So I am at a bit of a loss on this subject. I am trying to make sense of it, but taking into account the reputation of "modern production tubes" and the fact that my PT is grossly abusing some of those voltage maximums... and while I can not rule out the possibility of a mis-wiring, I have gone over wiring several times not to mention used the exact same scheme in several amps. However one boo-boo I have found is that I was using metal oxide screen droppers... I will be switching those to wirewound.

[Leo_Gnardo]

As you can see the tube has whited out

It's kaput, broke, busted. Doesn't need lightning flashes inside to prove it. Glass need not be cracked in any place you can see it. A popular leakage site is the "header" where metal pins protrude through the glass tube base. Each metal/glass seal must be perfect, and stay that way, or else pfffft (sound of a vacuum leak.) It happens, even to the best of 'em.

[Justin Thomas]

I thought the getter was for trapping residual gas inside the tube. Any possibility that an an internal tube element is melting down or overheating and releasing gases inside the tube? I've always heard that overheated plates release gas... why not other elements? There may not BE a crack in the tube...

Either way, as Leo said, it's dead. If they all are failing the same way, either there's a run of bad tubes, or...
Justin

[Chuck H]

My EL34 tube supply has been stretched beyond limits. Time to stock up again

either there's a run of bad tubes, or...

Justin may be onto something. If you've been experiencing tube failure on a regular basis and pulling from a "stash" of tubes I have two questions...

Is this tube stash a collection of all the same brand/type/batch? In which case those tubes may be faulty or simply not be up to the conditions in your amp. And,..

When you have a tube failure are you simply replacing THAT tube from your "stash"? This could also be significant because in a hundy W if the tubes at one or the other side of the push/pull aren't at least reasonably matched then the higher current tube will be doing most of the work. So if you had a tube failure, then just plugged in a tube and THAT tube draws less current than the other one on that side then that other tube may be the next to fail. Making the problem seem random in nature.

I'm not saying "matched" tubes are all that important for most guitar amp circumstances. But when you start pushing higher voltage and have tubes sharing larger wattage duty, well then yes, you really should have tubes of reasonably similar current draw.

I don't think you need to replace the PT or add a series resistor or otherwise mitigate high voltage. Plenty of old Marshalls and others are notorious for plate voltages over 500 and are working just fine. If you like the sound of the amp you should leave the circuit alone. If you're not married to the tone it has then sure, go ahead and change it for lower voltage and enjoy longer tube life and reliability. But I think the real problem may just be mediocre tubes, unmatched and in an amp with somewhat demanding operating conditions. This is actually the notion you opened the thread with and I think you had it right then.

Ruby selected Shuguang EL34B tubes. Buy them in matched quads. It may seem expensive at first, but it costs less than new transformers and the time it takes to do other modifications and in the long run it's likely that the life of the tubes will more than prove to be a value.

[Gtr0]

Justin may be onto something. If you've been experiencing tube failure on a regular basis and pulling from a "stash" of tubes I have two questions...

Is this tube stash a collection of all the same brand/type/batch? In which case those tubes may be faulty or simply not be up to the conditions in your amp. And,..

When you have a tube failure are you simply replacing THAT tube from your "stash"? This could also be significant because in a hundy W if the tubes at one or the other side of the push/pull aren't at least reasonably matched then the higher current tube will be doing most of the work. So if you had a tube failure, then just plugged in a tube and THAT tube draws less current than the other one on that side then that other tube may be the next to fail. Making the problem seem random in nature.

I'm not saying "matched" tubes are all that important for most guitar amp circumstances. But when you start pushing higher voltage and have tubes sharing larger wattage duty, well then yes, you really should have tubes of reasonably similar current draw.

I don't think you need to replace the PT or add a series resistor or otherwise mitigate high voltage. Plenty of old Marshalls and others are notorious for plate voltages over 500 and are working just fine. If you like the sound of the amp you should leave the circuit alone. If you're not married to the tone it has then sure, go ahead and change it for lower voltage and enjoy longer tube life and reliability. But I think the real problem may just be mediocre tubes, unmatched and in an amp with somewhat demanding operating conditions. This is actually the notion you opened the thread with and I think you had it right then.

Ruby selected Shuguang EL34B tubes. Buy them in matched quads. It may seem expensive at first, but it costs less than new transformers and the time it takes to do other modifications and in the long run it's likely that the life of the tubes will more than prove to be a value.

No problem...I am used to paying 100 - 130 euros for a matched quad.... I have a quad of TAD EL34B-STR (Shuguang), and so far so good - They have been in a Marshal Plexi for the better part of a year now. So those are also great options.

As for the current problems and replacing them... I was convinced it was a batch problem. So much so that I emailed my retailer and complained. Of two tubes that blew... rather fantastically.. they were within 30 days, so I got replacements that were matched for my previous orders... so they were drop in replaceable. Always checking the bias then as well if they fall within 4mA, then I do not worry any further and set bias according to the highest tube. However, having emailed my retailer and asking about this particular brand, they claimed that they were one of the more solid tubes - these were JJ E34L's - I have used them for well over a decade, religiously, without a single failure, until last year. Since the two sets were purchased a month or two apart, I am going to continue under the bad batch / pushing them toward their limits assumption.

My latest tube fiasco (mentioned above), a Tung Sol EL34 that - after about of month of awesome performance - began red plating - I caught it and got the amp turned off within a few seconds... and the amp fired up the next day and ran fine for at least 30 minutes. On a whim I swapped in another set... fired up the amp and the same tube (socket) flashed... This is a recent revelation, but doesn't change the fact of getting bad tubes these days.

So I have not tested any further with the Tung Sols... In fact I ripped the amp apart down to the chassis because it is a roughed out prototype anyway - rebuilding it now.

Of all of the tubes I have... The Tung Sol EL34 was a "to die for" tone - for me. Certainly haven't tried all, or nearly enough, brands
Tung Sol EL34
JJ E34L
Svetlana EL34 (came in my plexi, originals from the late 1990's)
TAD EL34B-STR

That's my list, in preferred order based on sound. The difference between TAD and Svets were small... then E34L has some kind of lower mid juicyness, then the Tung Sol was midrange madness - in a great and musical way. I was seriously disappointed to see the one side red plating.

Of course It is allllll subjective... but even in 6L6GC/5881 area, I found the Tung Sols were way nicer sounding than other brands... again subjective.

Anyway, again, thank you! I am going to continue on bad batches... see if I can get a super solid build together and further experiment from there.

Cheers!

[Justin Thomas]

... but even in 6L6GC/5881 area, I found the Tung Sols were way nicer sounding than other brands... again subjective.

Slightly off topic, but...
I really hope everyone knows that the TS "reissue" 5881s are NOT equivalent to a 6L6GC, but are a 19W tube equivalent to a 6L6GB... It's the Sovtek 5881WXT that is the "super tube" that's equivalent or better than a 6L6GC.

Also, did TS ever make an EL34? Hard to "reissue" something that was never manufactured in the first place. Instead, now we have Mullard 6L6GCs and TS KT66s. Maybe I'm just being petty, but this kind of crap on the part of New Sensor (yeah,I'm calling them out) just really pisses me off, because of the confusion and muddying it causes.

Anyway, carry on.

Justin

[Leo_Gnardo]

Also, did TS ever make an EL34? Hard to "reissue" something that was never manufactured in the first place. Instead, now we have Mullard 6L6GCs and TS KT66s. Maybe I'm just being petty, but this kind of crap on the part of New Sensor (yeah,I'm calling them out) just really pisses me off, because of the confusion and muddying it causes.

Spot on Justin. To the best of my knowledge there was never a real TungSol EL34. It's a marketing gimmick by New Sensor to sell the s.o.s. for an extra couple of bucks. You can call 'em out but they don't care, har har we get the money, ka-CHING!

[cerrem]

Now see more detail into the tube damage...
Very convinced this is Fly-Back arc-over voltage...
Seen this quite a number of times....
On at least several of these situations..I first asked the customer, who are DIY, to double check the impedance selector...I re-ask a thousand more times and they always come back saying it's fine....
Well the amp gets on my bench and I find the impedance sector wired up backwards or off by a few solder tabs..... main reason is that many times some people assume that the solder tab right behind the OHM number is the correct solder tab , when it is actually the opposite end .... or even worse are the Marshall selectors of the 70's with those "Window Selectors" which are more deceiving....
Getting back on point, the amp in many cases has the 4 Ohm wire inadvertently going to the 16 Ohm tab on the selector, thus making for HUGE plate load which made for crazy flyback voltage ratio ....
On strong transients with tight compliance speakers....the back EMF can by significant...this back emf generated from the speaker acting as a generator will get stepped up by the OT ratio... then this fly-back voltage is additive to the amps B+ voltage and any IN-Phase signals on the primary winding... Now you can get some pretty high voltage spikes in the kV range....
There are several paths this high voltage spike could take to arc-over..... Tube sockets can break-down, Air between PINS 3 and 2 can break -down.... or tube internals can break down at the bass pins...
One method I used to prevent the arc-over was to eliminate the ground potential all together.... Since the arcing was from pin 3 to 2 ,,,since PIN 2 is at Ground Potential from the Heater Center-Tap being grounded... I lifted the Heater Center-Tap off ground and floated it.... Used a LARGE film cap to provide the AC imbalance current to flow with a little impedance as possible...I might have paralleled several caps to get the HUM to acceptable level...
Then never blew a tube again

[Leo_Gnardo]

One method I used to prevent the arc-over was to eliminate the ground potential all together.... Since the arcing was from pin 3 to 2 ,,,since PIN 2 is at Ground Potential from the Heater Center-Tap being grounded... I lifted the Heater Center-Tap off ground and floated it.... Used a LARGE film cap to provide the AC imbalance current to flow with a little impedance as possible...I might have paralleled several caps to get the HUM to acceptable level...
Then never blew a tube again

Please forgive if I've mentioned previously in this thread, I know I have somewhere within the recent past, Ampeg's ground float method they used in many 60's guitar amps. Just like you, Ampeg placed a cap between filament winding center tap and ground. They usually employed a 0.1 uF 200V film cap. No harm in upping the voltage say to 600 or 630V to withstand almost any B+, and changing cap value to whatever minimizes hum. I have occasionally been able to use this technique to good effect. Having the filament floating on the cap eliminates it as a ground potential source for arcs from the nearby plate terminal as you said.

If there's been any arcing, tube bases and sockets should be closely examined for carbonized tracks or traces and dealt with accordingly. Replacement is always a prime option. In some cases I've been able to "save" bases or sockets by grinding away the carbonized areas with a Dremel tool.

[cerrem]

When the arc occurs over the air space between tube socket pins... I have used black RTV between the pins with some success..

[Gtr0]

Slightly off topic, but...
I really hope everyone knows that the TS "reissue" 5881s are NOT equivalent to a 6L6GC, but are a 19W tube equivalent to a 6L6GB... It's the Sovtek 5881WXT that is the "super tube" that's equivalent or better than a 6L6GC.

Also, did TS ever make an EL34? Hard to "reissue" something that was never manufactured in the first place. Instead, now we have Mullard 6L6GCs and TS KT66s. Maybe I'm just being petty, but this kind of crap on the part of New Sensor (yeah,I'm calling them out) just really pisses me off, because of the confusion and muddying it causes.

Anyway, carry on.

Justin

Spot on Justin. To the best of my knowledge there was never a real TungSol EL34. It's a marketing gimmick by New Sensor to sell the s.o.s. for an extra couple of bucks. You can call 'em out but they don't care, har har we get the money, ka-CHING!

Not knowing if the US Tung Sol ever made an EL34, it doesn't really matter to me personally... there is no reason that the Russian Tung Sol (New Sensor) can't make a variety of tubes, new or reissue. But saying this I realize there is a HUGE layer of BS to what tube manufacturers - especially New Sensor, do. One thing that disturbs me is that in one thread, maybe in this one or another thread at a different forum, someone said that Tung Sols are basically made in DIY basements. Then of course I have to wonder how this person knew that and if it is a directly known fact or the result of hear-say. Well, either way....

Now see more detail into the tube damage...
Very convinced this is Fly-Back arc-over voltage...
Seen this quite a number of times....
On at least several of these situations..I first asked the customer, who are DIY, to double check the impedance selector...I re-ask a thousand more times and they always come back saying it's fine....
Well the amp gets on my bench and I find the impedance sector wired up backwards or off by a few solder tabs..... main reason is that many times some people assume that the solder tab right behind the OHM number is the correct solder tab , when it is actually the opposite end .... or even worse are the Marshall selectors of the 70's with those "Window Selectors" which are more deceiving....
Getting back on point, the amp in many cases has the 4 Ohm wire inadvertently going to the 16 Ohm tab on the selector, thus making for HUGE plate load which made for crazy flyback voltage ratio ....
On strong transients with tight compliance speakers....the back EMF can by significant...this back emf generated from the speaker acting as a generator will get stepped up by the OT ratio... then this fly-back voltage is additive to the amps B+ voltage and any IN-Phase signals on the primary winding... Now you can get some pretty high voltage spikes in the kV range....
There are several paths this high voltage spike could take to arc-over..... Tube sockets can break-down, Air between PINS 3 and 2 can break -down.... or tube internals can break down at the bass pins...
One method I used to prevent the arc-over was to eliminate the ground potential all together.... Since the arcing was from pin 3 to 2 ,,,since PIN 2 is at Ground Potential from the Heater Center-Tap being grounded... I lifted the Heater Center-Tap off ground and floated it.... Used a LARGE film cap to provide the AC imbalance current to flow with a little impedance as possible...I might have paralleled several caps to get the HUM to acceptable level...
Then never blew a tube again

Interesting... never heard of floating the CT. How would one go about calculating the required value of this capacitor? I started another thread trying to resolve if I picked up some legit R3000 diodes... this should help to a degree as well right? At least in terms of fly back voltage.

Please forgive if I've mentioned previously in this thread, I know I have somewhere within the recent past, Ampeg's ground float method they used in many 60's guitar amps. Just like you, Ampeg placed a cap between filament winding center tap and ground. They usually employed a 0.1 uF 200V film cap. No harm in upping the voltage say to 600 or 630V to withstand almost any B+, and changing cap value to whatever minimizes hum. I have occasionally been able to use this technique to good effect. Having the filament floating on the cap eliminates it as a ground potential source for arcs from the nearby plate terminal as you said.

If there's been any arcing, tube bases and sockets should be closely examined for carbonized tracks or traces and dealt with accordingly. Replacement is always a prime option. In some cases I've been able to "save" bases or sockets by grinding away the carbonized areas with a Dremel tool.

As far as I can tell, there has not been any carbon scoring. BUT... as said I ripped the amp down to the chassis... rebuilding it now including new sockets. AND (@cerrem) while of course it is entirely possible that I had wired the ohm selector backwards, though I do not think I did, I have no way of knowing now that I tore it apart (though perhaps I have a photo and in such a case I am extremely curious)... in addition I am wiring impedance taps direct to their own/own pair of switchcraft jacks this time to eliminate the switcher.

EDIT: Okay, I did confirm via photos that I did wire it correctly. HOWEVER... there was a time that I had one of my cabinets wired wrong and the amp was set for 16ohm and cab was lower. Its hard to remember if these things occurred at the same time... I wonder.....

Of course I realize, as usual, 99% of the problem is the person sitting at the amp, user error.

[Gtr0]

Slightly off topic, but...
I really hope everyone knows that the TS "reissue" 5881s are NOT equivalent to a 6L6GC, but are a 19W tube equivalent to a 6L6GB... It's the Sovtek 5881WXT that is the "super tube" that's equivalent or better than a 6L6GC.

Also, did TS ever make an EL34? Hard to "reissue" something that was never manufactured in the first place. Instead, now we have Mullard 6L6GCs and TS KT66s. Maybe I'm just being petty, but this kind of crap on the part of New Sensor (yeah,I'm calling them out) just really pisses me off, because of the confusion and muddying it causes.

Anyway, carry on.

Justin

Spot on Justin. To the best of my knowledge there was never a real TungSol EL34. It's a marketing gimmick by New Sensor to sell the s.o.s. for an extra couple of bucks. You can call 'em out but they don't care, har har we get the money, ka-CHING!

Not knowing if the US Tung Sol ever made an EL34, it doesn't really matter to me personally... there is no reason that the Russian Tung Sol (New Sensor) can't make a variety of tubes, new or reissue. But saying this I realize there is a HUGE layer of BS to what tube manufacturers - especially New Sensor, do. One thing that disturbs me is that in one thread, maybe in this one or another thread at a different forum, someone said that Tung Sols are basically made in DIY basements. Then of course I have to wonder how this person knew that and if it is a directly known fact or the result of hear-say. Well, either way....

Now see more detail into the tube damage...
Very convinced this is Fly-Back arc-over voltage...
Seen this quite a number of times....
On at least several of these situations..I first asked the customer, who are DIY, to double check the impedance selector...I re-ask a thousand more times and they always come back saying it's fine....
Well the amp gets on my bench and I find the impedance sector wired up backwards or off by a few solder tabs..... main reason is that many times some people assume that the solder tab right behind the OHM number is the correct solder tab , when it is actually the opposite end .... or even worse are the Marshall selectors of the 70's with those "Window Selectors" which are more deceiving....
Getting back on point, the amp in many cases has the 4 Ohm wire inadvertently going to the 16 Ohm tab on the selector, thus making for HUGE plate load which made for crazy flyback voltage ratio ....
On strong transients with tight compliance speakers....the back EMF can by significant...this back emf generated from the speaker acting as a generator will get stepped up by the OT ratio... then this fly-back voltage is additive to the amps B+ voltage and any IN-Phase signals on the primary winding... Now you can get some pretty high voltage spikes in the kV range....
There are several paths this high voltage spike could take to arc-over..... Tube sockets can break-down, Air between PINS 3 and 2 can break -down.... or tube internals can break down at the bass pins...
One method I used to prevent the arc-over was to eliminate the ground potential all together.... Since the arcing was from pin 3 to 2 ,,,since PIN 2 is at Ground Potential from the Heater Center-Tap being grounded... I lifted the Heater Center-Tap off ground and floated it.... Used a LARGE film cap to provide the AC imbalance current to flow with a little impedance as possible...I might have paralleled several caps to get the HUM to acceptable level...
Then never blew a tube again

Interesting... never heard of floating the CT. How would one go about calculating the required value of this capacitor? I started another thread trying to resolve if I picked up some legit R3000 diodes... this should help to a degree as well right? At least in terms of fly back voltage.

Please forgive if I've mentioned previously in this thread, I know I have somewhere within the recent past, Ampeg's ground float method they used in many 60's guitar amps. Just like you, Ampeg placed a cap between filament winding center tap and ground. They usually employed a 0.1 uF 200V film cap. No harm in upping the voltage say to 600 or 630V to withstand almost any B+, and changing cap value to whatever minimizes hum. I have occasionally been able to use this technique to good effect. Having the filament floating on the cap eliminates it as a ground potential source for arcs from the nearby plate terminal as you said.

If there's been any arcing, tube bases and sockets should be closely examined for carbonized tracks or traces and dealt with accordingly. Replacement is always a prime option. In some cases I've been able to "save" bases or sockets by grinding away the carbonized areas with a Dremel tool.

As far as I can tell, there has not been any carbon scoring. BUT... as said I ripped the amp down to the chassis... rebuilding it now including new sockets. AND (@cerrem) while of course it is entirely possible that I had wired the ohm selector backwards, though I do not think I did, I have no way of knowing now that I tore it apart (though perhaps I have a photo and in such a case I am extremely curious)... in addition I am wiring impedance taps direct to their own/own pair of switchcraft jacks this time to eliminate the switcher.

EDIT: Okay, I did confirm via photos that I did wire it correctly. HOWEVER... there was a time that I had one of my cabinets wired wrong and the amp was set for 16ohm and cab was lower. Its hard to remember if these things occurred at the same time... I wonder.....

Of course I realize, as usual, 99% of the problem is the person sitting at the amp, user error.

[Justin Thomas]

I don't deny New Sensor the right to make tons of different tube types. But you can't reissue something that never existed in the first place.

My bigger point was to NOT simply think of the TS "reissue"5881s as equivalent to a 6L6GC. They never were. The 5881 is a completely different tube. Sure, it might work in the same sockets, but it still only dissipates 19W to the GC' s 30. Except for the Sovtek 5881WXT, which is as tough as they come.

Justin

[Mick Bailey]

Having the filament floating on the cap eliminates it as a ground potential source for arcs from the nearby plate terminal as you said.

I'm not convinced. An arc will still develop if the voltage is sufficient and then the capacitor actually intensifies the arc by causing an oscillatory circuit and can reach a much higher, self-sustaining voltage that will short out the cap unless it's one designed to stand tens of Kv. I've used 30Kv series capacitors in a successful ignition system for agricultural engines to intensify the spark during startup for low-flashpoint fuels. The capacitor increases the arc energy enormously by storing and releasing energy.