• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

The Limitations of an EL34 in UL

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm not sure if everyone here reads Tubelab as intently as myself, but I just saw the addition of the "SimpleSE" amp to the sight, which is very similar to an amp that I am working on myself. Tubelab talks about pumping 410v into a 6l6 in UL, which would scare me if I did the same, So I am going to go with an El34 as it has a higher voltage tolerence than the 6L6.

So my question to all of you is very simple. How much voltage do you guys think an El34 can take both on the plate, and on G2, when wired in Ultralinear without becoming a small explosive device? I have a 200ma power Tranny, and I'm not afraid to use all 200ma of it.
I use cathode bias in that amp (schematic coming soon) with a 560 ohm resistor. The plate supply voltage is 450 when using SS rectifier, and 430 with tube rectifier. This puts 400 to 420 volts across the tube. I have used many different tubes in that amp including JJ and Svetlana EL-34. Of all the different tubes the EL-34 operates closest to (but within) its maximum dissipation ratings.

The original Mullard data sheet gives plate voltage max at 800 volts, and G2 max at 500 volts. Plate dissipation max at 25 watts, and G2 max of 8 watts. I am running about 60 to 65 mA total tube current with 400 to 420 volts across the tube. This works out to 27.3 watts total (plate and screen) dissipation.

One of the first things to consider when designing a SE amplifier is how hard to push the tubes. Each tube has a maximum power dissipation rating. You look this up, and then decide how close to this limit you want to operate. Many people use the 80% rule of thumb, some go more conservative, especially with expensive tubes.

I tend to test the tubes that I want to run in the circuit, and find the point where the tube shows distress. There is a picture of this testing on the SimpleSE web page. In that case the 6L6 was running at 44 watts (way over spec). The interesting information here is that the red glow on the plate was evenly distributed (no hot spots) and even though the plate was very red there is no glow on the screen grid wires. This tells me that I would have no problems running this tube at its published max of 30 watts (the fact that they are $8 helps). It also tells me that even though the plate is glowing, the screen grid has no problem with 400 volts.

I performed the same test with the JJ and Svetlana EL-34's, after finding one bad Svetlana EL-34 the others were stable and showed a slight redness (room lights off) at 35 watts. 27 watts should be no problem, but will work the tubes hard.

The problem child of the bunch turned out to be the Chinese KT-88's. These started to glow at 35 watts, and there were hot spots visible on the plates, not even heating. This will kill a tube relatively quickly. These tubes are specced for 35 watts Plate dissipation and 7 watts of screen dissipation. Here they are showing two pronounced red spots at 35 watts total plate and screen dissipation. This should not happen. I got these tubes from a vendor about 5 of 6 years ago because of customer returns. He decided not to carry them any more and sold me his entire stock cheap. I have been using them without issue in P-P amps. I decided to use them at 30 watts anyway and see what happens. I did not torture test the Sovtek or GE tubes since I only have a few, and the GE's are expensive.

While designing the amp I used a variable power supply, adjustable bias, and 2.5, 3.0 5.0 and 6.6 K ohm loads with 6L6's, EL-34's, KT-88's, 6550's and a few others to arrive at the combination that worked the best with all of the available tube types. I tried to find one set of conditions that allowed all of these tubes to work, afforded each tube the ability for its characteristic sound to come through, and sound great at the same time. Most of the published operating conditions for any of these tubes seem to want to run them at a relatively low voltage (250 to 350 volts) and force a higher current swing by using a 2.2 to 3.6 k ohm load. I found far more dynamic sound by raising the supply voltage and lightening the load at the same time. I found best sound with a 5 K ohm load. Others have proposed similar operating conditions with a load as high as 10 K ohms.

Popping 6L6GC's into the same amp actually draws slightly less current, and therefore there is lower dissipation. The 6L6GC is rated for 30 watts, so you are safe here. The book says that a 6L6GB can only handle 19 watts, but I have ran this amp for 2 hours on a pair of old GE's that previously lived a hard life in a Fender Bandmaster. They sounded good with no problems.

I also used Chinese KT-88's, and both Sovtek and GE 6550's. These tubes run at about 30 watts (rated for 42) and offer the best bass of all of the tubes tried.

All of these tubes operate at or below their published maximum ratings, and far below my "tested safe limit" even the 6L6's. It is important to consider tube dissipation in all SE amplifiers, since this is often the limiting factor when deciding how hard to push a tube.

Once you decide on the dissipation that you want to run a tube at, then you need to figure out how to get there. Lets say you want to run a given tube at 30 watts. You could put 250 volts across it and draw 120 mA through it, that is 30 watts. You could use 300 volts and 100 mA or 400 volts and 75 mA. If the tube was rated for 450 volts you could use 450 volts and 66mA. Each of these sets of conditions will result in 30 watts dissipated in the tube (equal stress on the PLATE). Each should be capable of similar power output if the optimum load and drive conditions could be met. Each set of conditions will have an optimum load impedance and drive requirements that can be determined graphically by drawing a load line. If you do this you will find that for low plate voltages the optimum drive conditions can not be met without driving the grid positive. This will limit the maximum power output for some tubes. I have found more power to be available at higher voltages and higher load impedances, it sounds better too.

There is another point to consider. If each of these sets of conditions generate equal stress for the PLATE (dissipation), which set do you think would generate the most stress on the cathode? The job of the cathode is to push electrons out into the space charge cloud, to be sent on their way to the plate. More current = more electrons sucked out of the cathode! To operate the tube at less current, what do you do? You turn UP the plate voltage! I know that this sounds counterintuitive, but it is true. Find a ham radio operator who built linear amplifiers using transmitting tubes and ask him what makes those old Eimac tubes live the longest. They will tell you to operate them at lots of voltage 2.5 to 3.5 KV, and low current!

You didn't say how many volts your 200mA tranny puts out, but if you figure 100 mA per tube and 25 to 30 watts max per tube, you are limited to a maximum voltage of 250 to 300 volts across the tube. This requires a 2.5 to 3 K ohm load to get good power. I tried those operating points and I didn't like the sound, but you might. To me all of the different tube types sounded the same. The differences started to become obvious at 350 volts and above.
Man, what a great reply! I have been studying tubes for about 8 months and have a small fortune in books, but that was good straight forward stuff. I am working on my first DIY, and have just finished the PS. I planned on running EL34's triode mode at 430v, a schematic I found on this forum. However I just came accross some older EI KT 90's. and would really like to hear your thoughts on substituting them. BTW I read an interesting bit on EL34's the other day. It basically stated that when Mullard designed them, the big deal was that the internal wiring matched up with the pins without any crossing over to reduce the risk of flashover at higher voltages. Just thought that was interesting. I've also heard the school of thought that in older days and mass production, the tubes were run at lower voltages to save on iron costs. Seems to me that the cool thing about DIY is you don't have these constraints. In other words, let the big dog eat. Thanks again for the very useful info, Jay
I have no experience with the KT-90 so I can't help here. I remember reading something on one of these forum pages about them being unreliable at high screen voltages. Search these forum pages before you try it.

It is true that transformer and capacitor costs go up with imcreasing voltage. This was more true 40 years ago. It is interesting to note that no one has a problem operating EL-34's or 6L6's at 450 volts of higher in a push pull amp. Many guitar amplifiers routinely run them at 500 volts plate and 450 volts screen.

I have seen a grosly misloaded (16 ohm speakers on the 4 ohm tap) guitar amplifier (one of mine) cause a cheap 6L6 to arc over from pin 3 (plate) to pin 2 (filament) at the base of the tube. In this case nothing bad happened, but a wicked screaming sound came out of the speaker. The potential for arcing in the base is the limiting factor for octal tubes. The 6BG6 is a 6L6GB with the plate attached to a cap on the top of the tube. These can take 700 volts continuous and pulses to 5000 volts.
Wow, Thanks alot tubelab for the great response, it helped me alot in this wide world of tubes.

my 200ma power tranny is 300v, so after rectification it goes up to about 430 max, I am currently runing at 378v which means about 80ma per tube at 30 watts of disipation, I wish it was working well enough to draw 80ma per tube.

just to make sure, these numbers are true both for Triode-strapped mode, and for UL? or just UL?

Thank you all so much, this is the most informative thread ever
OK, Sherri was not home tonight, time to make some noise. I gathered up some CD's that want to be played loud, Dire Straits (rock), Lisa Marie Presley (rock with female vocals), Berlin (techno), Code of Ethics (techno with major bass), Eric Clapton, and of course Jimi. To get the full effect I made some temporary modifications. I put a jumper in place of the filter choke, added a 100 uF ASC motor run cap in parallel with the B+, and selected the solid state rectifier. I even pulled the rectifier tube (10 less watts draw on the power transformer). These modifications allowed the B+ to reach 480 volts. The amp was LOUD. All testing was done in UL mode.

With the Shuguang 6L6's the cathode voltage was 38 volts, the plate voltage was 470 volts (10 volt drop in the opt). This is a cathode current of 67mA and a total tube dissipation of 29 watts. The amp cranked for over an hour with the volume control dimed (guitar amp speak for set on 10). The extra voltage gave the amp enough headroom so that clipping was only audible on a few tracks (visible on the scope more often). During all of this the 6L6's remained calm and showed NO signs of distress. During long periods of silence there was NO glow on the plate or screen grid. Silence causes the worst case dissipation in an SE amp.

Next I tried Svetlana EL-34's. The voltages, currents and dissipation numbers were almost exactly the same. There was no glow from the plate under any condition. I saw a faint glow on a few of the screen grid wires during long periods of silence on ONE of the tubes. I tried another tube, it did not exhibit these symptoms. I bought 4 Svetlanas and 4 JJ EL-34's at the same time, one Svetlana was DOA, and one does not like running near the edge of spec. Two work OK. These got about 1/2 hour of playing time.

Which brings me to the JJ's. I have 4, I tried all 4. These tubes draw slightly less current in the same amp. All 4 tubes were within 1 mA of 63mA. This allowed the B+ to reach 486 volts,plate voltage 475, tube voltage 439. Good thing I used 500 volt caps. Under these conditions the tubes were dissipating about 28 watts. All 4 tubes behaved perfectly, no glow at all, excellent sound. One pair played for over an hour. When I bought the EL-34's they were the same price. Which ones do you think I will buy next time?

I have copied the material from these tests, and this thread to a theory web page, I will upload it on Sunday. Maybe I will have the schematic done by then. There were 430 to 440 volts across these tubes for long periods of time without incident. I would have no problem with the Shuguang 6L6's or the JJ EL-34's at this voltage level. I have not tested other brands yet. I am still amazed at the sound that I am getting from two $18 output transformers powered by a $43 power transformer.
I was posting at the same time you were, so I just read your question. In all of the dissipation calculations, I was talking about STATIC dissipation. How much power is being burned up in the tube with NO input signal. In this case NO power is being produced by the amp, so all of the power is burned up as heat, none is turned into sound. In this case it makes very little difference whether the amp is triode, UL or even pentode. This case is the worst case for an SE amp. As you apply signal to the amp, some of the DC power is turned into sound. A simplification is, if the amp is dissipating 30 watts static and you drive it hard enough for it to put out 5 watts rms, then the amp is dissipating 25 watts. Any amp that we design must be able to idle for long periods of time, so we use the static calculations.
How much voltage do you guys think an El34 can take both on the plate, and on G2, when wired in Ultralinear without becoming a small explosive device?

Sherri decided to go garage sale shopping this morning with her friends, so it is time for some more loud testing. I like to find the limits of my amplifiers before I even think about selling them, so today I took the voltage to new limits. All of the voltage boosting modifications are still in place from last night, so I plugged the whole amp into a Variac and turned it up until the B+ hit 510 volts. I didn't do any other measurements. This is as far as I dared go on 500 volt caps, as they WILL become explosive devices.

The amp has been playing for an hour with the JJ EL-34's. No glow, nothing bad, just clean, LOUD, music. Flamenco guitar at this moment. Later I will try the Shuguang 6L6's and maybe a few others at this voltage level. The ultimate torture test will come when I play my guitar through this thing!

I have been purposely pushing the amp into clipping to capture its transient behavior on a storage scope. It is well behaved. I plan to add all of this information to the web site soon.
My brief addition here, since about all has been said by Tubelab:

Do note that in UL 6L6s are allowed to take 500V on both anode and g2 (that is because at lowest plate voltage, the screen is also below the B+ voltage, thus at somewhat less dissipation. Voltage-wise one can then work at some 540V in the case of cathode bias. But not in SE; the need for class-A operation there, thus a certain minimum current, will mean that the Va, Vg2 will have to be limited in order not to exceed dissipation.

The main destroying characteristic is the dissipation (an indication of how hard the cathode work, apart from heat that must be gotten rid of). In a current amplifier I work 6L6GCs at 600V anode and 500V g2 with fixed bias of 44V on the cathode. Converted that means 556V anode and 456V g2. The electrode voltages are not the main thing, the heat is. Also, these specs are at a room temperature of 25 degrees and "normal" convection cooling. Even with a very mild fan allowed dissipation is increased.

I have also found that in general EL34s can make a "rosy" g2 before 6L6s because of the non-aligned screen grid windings. (Although g2 current varies more for beam tubes, the fact that g2 windings are screened from the cathode by g1 makes for a more stable operation.)

Agreed. Dissipation is almost always the limiting factor in an SE amplifier. The life of the tubes is usually inversely related to the dissipation that the tube is operated at. There are many operating point possibilities to arrive at a given dissipation. For a constant dissipation, the operating point that allows the lowest cathode current will usually afford the longest tube life, provided that maximum voltage ratings are not exceeded.

All tubes have a published set of maximum ratings, however these were generally just copied from the original 50 year old data sheet, and may not be representative of currently manufactured tubes. The point of my extreme testing is to find any weak links in my new amplifier design (none so far), and identify the tubes that can be reliably operated close to (but within) the published specs.

In doing this I have identified some tubes that are not reliable when operated well within the published specifications. I do not want to put these into an SE amplifier that I offer for sale. They may be OK in a push pull design. When I do sell SE amplifiers, they will use the tube types that have passed my torture testing.
Care to share some of those less reliable brands with some of us pecuniary-challenged folks in these times of quality-challenged products? They say that experience is a good teacher, but second-hand is less expensive! If you do not care to do it here on "open" post, kindly e-mail me privately.

I have often thought about what exactly maximum voltage ratings meant (I think we had this discussion elsewhere months ago). Will there be sparks at anything higher; will tube life be affected? As I had to do with many 6L6s, I noticed an air-gap (oops, I think I should say a space!) round the anode rods in the mica mounting discs when the max. voltage spec. increased. But I could not see any difference in the construction of the 7027 (600 Va) compared to the 6L6GC (500 Va). And not anything in the EL34 to warrant 800V compared to the above. And everybody connected with the design must by now be dead.

....except for a Dutch friend of mine who in his youth worked for Philips at Eindhoven, where he helped making tubes. He said that the ratings were indeed not absolute, but a safe guidance to be used together with other ratings. It also depended on the frequency handled. Somewhere into the r.f. field things could become marginal - I cannot recall exactly how. I do remember that we once tested with a non-destructive (ionisation) tester in our lab. There a 6L6GC could take well over 1,5KV. But don't somebody start using that now!

Anybody else with wisdom/experience in this regard?

Damn Tubelab, I like the way you operate! Talking about the higher voltages in P/P, my 70's B+ is over 500, which has never been a problem with the EL34's biased at 25mA. I'm taking the same approach with the SET I'm working on (high voltage-low current). BTW I have nothing but good things to say about the JJ EL34's. They have been IMO the best new prod. I have found. Thanks for all the testing work and sharing your findings in detail. Wish I could have been there for the Jimi, Jay
From these experiments I have found two popular tubes that I didn't like. The Chinese KT-88, were sold to me because the tube dealer decided not to carry these any more due to quality issues. I have seen two of these arc internally in guitar amplifiers. I have been using them in P-P amplifiers in the 35 watt range with no problems. During the steady state dissipation test, 10 out of 10 tubes had visible red spots on the plate at 30 watts on the plate. These are 35 watt tubes, so I tried 35 watts. The plate was glowing brightly and there were hot spots. To be fair, these tubes are about ten years old. It is possible that Shuguang has improved them since then. I will try some new ones when I get the time.

I purchased 4 Svetlana EL34's and 4 JJ EL-34's from Antique Electronics Supply at the same time. All 4 of the JJ's have been tortured without incident. They sound excellent. One of the Svetlana's had a problem as soon as I took it out of the box. A few of the grid wires glow red very brightly at voltages as low as 200 volts. A second tube has a faint glow at 400 Volts across the tube.

I have several other tubes to try in this amp, I will report back here if I find anything unexpected. As soon as I get the boards, I will be building a few amps inorder to write the construction manual. These amplifiers will undergo some serious testing, so I may find some more wimpy parts.
Small world, I just received 2, Svetlanas from a friend by mail today. He gave them to me to try out in the 70. Since there was only two, I stuck them in the right channel and biased at my normal 25mA. I noticed on one tube I had to turn the bias pot quite a bit. After everything warmed up I flipped on the pre to compare channels (I had EH in the left). The svetlanas seemed to have tighter bass. I left the amp that way and later while soldering at my bench I saw a horrible blue purple light start to glow. I threw the switch as quick as I could, but she was plenty hot (the glass was smoking a little). It was the tube I had to turn the pot a lot on. Oh well, I guess thats a no for Svetlana. I broke my JJ's back out and all is good once more. I guess thats another yes for JJ.
Will there be sparks at anything higher; will tube life be affected?

Possibly! I have had two voltage related failures. Both were 6L6's and both were AC voltage related (not DC). The first was a guitar amp that had a miswired load. The two 8 ohm speakers were in series, but connected to the 4 ohm tap. It was a P-P amp that I had built earlier. The owner had put in new speakers, and wired them in series because he said it sounded better. He brought the amp to me because "it screamed at him" sometimes when he played it loud. The screaming sound turned out to be an arc from pin 3 (plate) to pin 2 (grounded filament) INSIDE the base of the "Audio Glassic" blue glass 6L6. I replaced the tube, and after discussing with the owner, decided to run the 16 ohm speakers on the 8 ohm tap. No further issues were seen. That amp runs 430 volts of B+ and about 400 volts on the screen.

I saw another 6L6 arc over between the pins on the outside of the base, eventually carbonizing the tube base and the socket. This was a stereo amplifier that was playing at high volume when the load was accidentally disconnected. Replcaing the output tubes and socket fixed the amp.

I have not experienced and DC voltage related issues with pentodes, or triode wired pentodes. Despite my reputation for extreme testing, I don't really hammer tubes too hard in normal operation.

It is interesting to note that the 807 and the 6BG6 are closely related to the 6L6GB. They have the plate brought out to the top cap. The plate voltage spec for these tubes is considerably higher, indicating that the internal structure is capable of a higher voltage.

I have seen a voltage related runaway condition occur in Sovtek 300B tubes. The specs printed right on the box claim that the tube can handle 450 volts. Some can, some can't. The ones that can't will go into a runaway condition where you can't shut the tube off even with -150 volts on the grid. Upon investigation it seems that these tubes have some filament wires exposed above the end of the control grid winding. This allows a path from the filament to the plate that does not pass through the grid.
tubelab.com said:
Upon investigation it seems that these tubes have some filament wires exposed above the end of the control grid winding. This allows a path from the filament to the plate that does not pass through the grid.

Never heard of this. But more than once I've seen poorly made small signal tubes with long portions of the filaments exposed on the top, I guess that's too over the plate structure to cause any harm.

So it's probably a defect in the assembly of the tube (someone mounted the cathode too down) or it's a design problem?

Are cheap chinese 300B affected by this, or have you seen this on any other power tube?

Thank you George for your experience sharing, you're great! :D
Okay, I have really learned a ton from what you guys have said about Triode (which I currently have set-up), and about UL (which requires a very coveted 40% Center-tap) However now, I want to know how I can wire my el34s into Pentode. I guess its a little on the thread-jacking side, but I feel that its still directly related to the limitations of tubes. what I can't transfer from the UL to Pentode configurations were the simple numbers that tubelab posted about plate dissiapation. So I am going to post what I think is right and everyone can check to see if my thinking is correct

So The maximum plate dissiapation on an El34 is around 30watts, and the maximum grid dissiapation is around 8 watts. So then if I send 410v to both the plate and the grid, the grid will be drawing (8w/410v) ~19ma and then each plate will be drawing around (30w/410v) ~73ma. is this the correct thinking? or do I need to send a smaller voltage to the grid than the 410 going to the plate? Please tell me where I am wrong and why
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.