6L6GC to KT66

[EFK]

I'll keep it short. I'd like to try KT66 in a 6L6GC amp. Info 'out there' seems slightly sketchy as to whether this is possible. If so, the big hangup seems to be the extra heater current on KT66 stressing a transformer (OT? PT?). I have yet to see any explanation as to HOW you would determine if the transformer could handle the extra current draw. This is a Gibson amp - GA42RVT - and Gibson has been no help whatsoever in this particular line of questioning. Anyone? Thanks!

[mooreamps]

My data sheets show the heater current for the 6L6 is 0.9 amps. For the KT66, heater current is 1.3 amps. Otherwise, they seem to be pin-for-pin compatible. Now, I'm not knowing the spec's on your power transformer, if it can handle the additional 400 mills of heater current. You would have to plug the tube in, and see how much it pulls the heater voltage down with it installed. I think anything below 6.0 volts is going to make the power transformer run pretty warm. Anything below 5.7 volts, and that will make the rest of the amplifier unusable. That low of a filament voltage, and it will just kill the tone coming out of the pre-amp.


-g

[EFK]

Thank you for the response. Is there any way to determine the specs of the power transformer if Gibson is not forthcoming? Is the label on it in any way informative? Assuming not, how would one measure the heater voltage - is this too much for a mb description?

[MWJB]

"Is the label on it in any way informative?" Does the label say anything like "6.3VAC 4A"? If so, you really need 5A or more if you're going to keep all the 12AX7s in there (6A+ would be nice).

If your B+ voltage drops by more than 20vdc then I'd assume that the load on the PT was excessive. A true KT66 may also draw significantly more plate current.

Pulling any 'surplus to requirements' preamp tubes (can you live without a channel/ reverb/trem?) will free up 300mA each of heater current.

There is a Russian, so called, KT66 that's really a 6L6 in a KT88 bottle (any decent tube vendor will have details), this has the same heater current draw as a 6L6 and might be the safer option. Whether it sounds like a true KT66 is another matter.

I'd be even more cautious than Mooreamps and exclude anything that pulled your heaters under 6.2VAC and/or pulled down B+ over 20v. You would measure you heater voltage from pin 2 to pin 7 of a 6L6, heeding the usual advice regarding safe practices for working around tube circuitry - read up on these (e.g. power down & unplug the amp from the wall to make/unmake any connections, use voltmeter leads with clips, double check integrity of connections so you don't short against an adjacent tube socket pin, don't power up without a speaker load connected)...a lot of current in that heater winding.

[jrfrond]

They are a pin-for-pin swap aside from the increased heater current. Bias is even similar.

Plug them in, and monitor the transformer heat with your hand (check it with the old 6L6's first). If the transformer becomes substantially warmer and uncomfortable to hold for an extended period, then you will need a separate heater transformer. You CAN try them to see how they sound, and if you even like them, without worrying about transformer damage.

[soundmasterg]

Run them in the amp for awhile and see if the heater voltage drops much below 6.3v. If it does, then the power transformer needs some help and you should add a seperate filament transformer.

Greg

[EFK]

This is very helpful! Thank you all. Power transformer reads "IMAG 7002 / 680VAC / Mexico." I assume this is 6.8 VAC. I can yank #3 and #4 preamp tubes - at least temporarily - as I absolutely never use tremolo and I can live without reverb if in in the right room. Too bad one tube wasn't a dedicated tremolo tube as I could live without that permanently. I think I'll give it a go to try out the sound and see where things stand.

So, is it a safe assumption that if you can hold your hand on the transformer w/o pain, there is no need to worry (as sort of a blanket generalization in this specific situation)?

Can't thank you guys enough!

[mooreamps]

Thank you for the response. Is there any way to determine the specs of the power transformer if Gibson is not forthcoming? Is the label on it in any way informative? Assuming not, how would one measure the heater voltage - is this too much for a mb description?

Yes there is.....

First, you pull all the tubes, and then measure and record the output voltage ; unloaded. Then, you install all the tubes, and then re-measure and record the output voltage. From these two values, plus knowing the value of the load, one can approximate what the rated max load current would be for that device.

-g

[MWJB]

"Power transformer reads "IMAG 7002 / 680VAC / Mexico." I assume this is 6.8 VAC - No, 680VAC most likely refers to B+ secondary winding voltage.

Pulling 2 preamp tubes will pretty well redress the balance.

"So, is it a safe assumption that if you can hold your hand on the transformer w/o pain, there is no need to worry (as sort of a blanket generalization in this specific situation)?" No, it is NOT a safe assumption. As Greg says, measuring voltage drop at the secondaries (more than 20v B+ drop, heaters running at less than 6.2VAC) is a more tangible way to establish safety margins. Some PTs run too hot to touch for years with no ill effects, others run cooler and one day they just crap out. In either case after a set's worth of playing I'd expect your PT to be uncomfortably hot to the touch, even with the stock tubes.

If you have doubts there are still the Russian KT66 (rebottled 6L6s), same current draw as 6L6s.

[EFK]

"If you have doubts there are still the Russian KT66 (rebottled 6L6s), same current draw as 6L6s."

True, but there are much better 6L6s available! Seems silly to pay for a differently-shaped bottle only!

This has been very sobering and helpful. I think I'll pick up a pair of 66s and try them out, and if I like the sound I suspect I had better buy some books and get to reading. I have a very healthy respect for electricity. Thank you all very much! I hope this thread pops up in the search engines - there is a lot of talk about swapping 6L6s for KT66s, all with the caveat, '...make sure your amp can handle them...' but with no specific details on exactly what to do to ensure no long-term damage. This ought to redress that.

[jrfrond]

For the record, there are a lot of underrated transformers in guitar amps, particularly for the way we used them nowadays. Transformers that are consistently run too hot (i.e. too hot to touch after use), will eventually break down and short. It is only a matter of time until the Formvar insulation on the magnet wire breaks down totally. Newer magnet wire that uses Polysol and other similar poly-based lacquers, have a better chance at survival.

[Chuck_Farlee]

There is a Russian, so called, KT66 that's really a 6L6 in a KT88 bottle (any decent tube vendor will have details), this has the same heater current draw as a 6L6 and might be the safer option. Whether it sounds like a true KT66 is another matter.

That would be the Sovtek KT-66. It is the same guts as their 6L6WXT in an ST-shaped envelope.

Compare:

http://thetubestore.com/sovtekkt66.html

http://thetubestore.com/sovtek6l6wxt.html

The KT-66 version is almost twice as much as the 6L6WXT. Too much to pay for a fancier envelope, IMHO.

chuck

[EFK]

Back from the dead! Well I finally got a pair of Groove Tubes KT66-C to try out in this amp. I pulled V3 and V4 (reverb and tremolo drivers) to ensure ample heater current and WOW. I could never go back to 6L6. The sound is spectacular. OR I should say, was.....

I put about 10-15 hours through it with no problems at all. Not so much as a pop, click, ring, nothing. Did I mention how beautiful the sound was? And then, here I am playing along happily, volume at @ 1 o-clock (which is where I generally keep it - I have no neighbors) when suddenly it sounds like the volume gets rolled back to half; thin and weak. ???? I hit a few notes and fiddle with the volume knobs to make sure I didn't bump anything, and then put it on standby. Tubes look fine - no red plate, heater filaments are glowing nicely, faint blue glow in tubes. Alright, turn full power back on and the sound is back. So I play about 5 more minutes give or take, and it happens again. Sounds like half volume. This time it doesn't go away.

Yanked the tubes, they *look* fine although there are a few areas of very, very faint light brown tinge on the otherwise gray plates. Hard to see unless I shine a drop light on them. I popped in a few different sets of my old 6L6GC and all sound just fine (well, they're no KT66 but they sound as they ought). No problems at all.

What is going on here? I have to add - and I'm not making this up - in 25 years of playing tube amps, I have never blown a power tube. IS this what a dying power tube does? Sorry to sound stupid. What would kill these tubes at only 10-15 hours? I keep a pretty close eye on things and never saw any red plating or sign of overheating. Transformer never got too hot to put my hand on it and as I said the amp with the old 6L6GC works just fine.

This is a cathode bias amp (Gibson GA42RVT) and I would imagine 6L6GC and KT66 are supposedly close enough in bias that it ought to self adjust. At least, as I've been led to believe, the bias ought to be 'close enough' that while it may not be optimal, it shouldn't kill a tube/tubes in 10 hours!

Sorry to be so wordy. I'm in an extremely rural area and there aren't any local techs.

Schematic of the amp is here:

Gibson GA42RVT

Is there anything visible in this schematic which would be contrary to KT66 usage?

This guy has the 1X12 version of the same amp, offers some electrical info on his which I can't offer:

"unstablilized" cathode bias voltage

Hope someone can offer some information - thanks!

[MWJB]

"This is a cathode bias amp (Gibson GA42RVT) and I would imagine 6L6GC and KT66 are supposedly close enough in bias that it ought to self adjust. At least, as I've been led to believe, the bias ought to be 'close enough' that while it may not be optimal, it shouldn't kill a tube/tubes in 10 hours!" - You've based this purely on assumptions, which can easily come back and bite you on the ass. In the same circuit a true KT66 wil usually draw more plate curent than a 6L6. Did you check what current you actually had?

[stokes]

I installed a pair of those Sovtek "KT66's" in a Boogie MKII for someone (against my warnings) and they lasted about 5 hours.Although I rebiased the amp (installed a bias pot for just that reason) those tubes cant take the voltage,forget the exact voltage,but it was in the 450 zone at 30ma's,I set the bias so as to be safe for a plain 6L6 as opposed to the higher rated 6L6GC and they still fried in about 5 hours.I suspect the GT's are just re-labeled Russian or Chinese crap.As a comparison,I built an amp for my son around a NOS pair of GEC KT66's and that amp has been run 3-5 nights a week for almost 3 years with the tubes running 62ma's at about 375v's,which is HOT.I recently gave the amp a once over,and the tubes are still fine,sound great and test the same on my testers as the day I installed them.I dont know what they are putting in these current production KT's,but it aint a real KT66.

[EFK]

Well you're right it probably did bite me on the ass and I'm thankful it didn't burn out something more important than a cheap groove tube. I took the chassis out today and began looking around in there. Not touching anything though. If someone can tell me what to measure and how to measure it- plate current, B+, bias? -, I'll see if I can come up with some numbers which may shed more light on exactly what is going on. I'd surely appreciate the help. BTW Stokes, *supposedly* the GT KT66 HP, which are no longer being made, can handle a lot more juice. I might try to find a set of those. NOS are pretty high $$$ right now!

MWJB - back to the heater voltage, you stated that: "You would measure your heater voltage from pin 2 to pin 7 of a 6L6..." That is under load, correct? Do I set the meter to AC or DC? And, which pin gets red, which gets black?

Mooreamps stated: "First, you pull all the tubes, and then measure and record the output voltage ; unloaded. Then, you install all the tubes, and then re-measure and record the output voltage. From these two values, plus knowing the value of the load, one can approximate what the rated max load current would be for that device." Can you tell me how and where to measure this?

I'm trying to learn and I appreciate any help.

I took a couple of pictures of the inside of the chassis. If anyone is interested I'll post them.

[MWJB]

First, to be on the safe side until you are more familiar with what you are doing...

Make all meter connections with the amp unplugged from the wall.

Use meter leads with clips/wire grabbers.

Power down, unplug from the wall when changing connections. Double check integrity of connections before powering (no chance of shorts etc.)

Power tubes, dc volts <600v (or equivalent setting, maybe <750v?), black wire to ground red wire to part you want to check - pin 3 (plate), pin 4 (screen), pin 8 (cathode). Count pins clockwise from the keyway, whilst looking at the top of the tube socket. On a fixed bias amp you would also measure negative voltage at pin 5, but onm a cathode biasedamp there will be 0v dc at pin 5.

Heaters - set to AC volts <20v (you may only have a <200v setting, you're only looking for between 6.3 & 6.9VAC so whatever range suits best). One wire goes to pin 2, the other goes to pin 7 of the power tube...doesn't matter which goes where.

[EFK]

Thanks! With amp on and tubes running (tubes should be in and running, right?):

Pin 3 to ground: 483 VDC
Pin 4 to ground: 481 VDC
Pin 8 to ground: 39.4

CAthode resisitor is shared by both tubes. It reads. "Dale RH-25 25W 330 ohm 1%

I get 38.8 reading across the resistor.

The amp is *supposedly* class A.

Board reads, GA20RVT/40RVT/42RVT
Circuit design by Pyotr Belov
Board Layout by John Suhr and Pyotr Belov

What does all this mean? I'm trying to figure out how to use Weber's bias calculator....

[MWJB]

To use the calculator you really need to actually measure plate current..ideally use a bias probe.

Rule of thumb method is to divide cathode voltage by cathode resistor value, then divide by no. of tubes.

e.g. 39.4/330 = 120mA, divided by 2 tubes = 60mA per tube.

Now multiply by plate voltage, less cathode voltage, so 443.6 * .060 = 26W per tube, less 1W for screen current = 25W...a bit higher than I'd like unless you were exclusively using tubes proven to have a 30W dissipation (sturdy NOS or JJ6L6). KT66 is rated at 25W...if these are your figures with 6L6 then I'd expect a bit more current from the KT66.

FWIW I have never got an accurate plate current from going the math route...may be my math, but I prefer to actually measure current.

I'd replace the 330ohm resistor with a 470ohm or 680ohm, same wattage rating. As plate current drops, voltage will rise but overall dissipation will drop with each larger cathode resistor value.

Forget what class it's supposed to be, focus on what makes it sound good & keeps it reliable.

[EFK]

Thank you for the help! I hate math. I thought I read somewhere that you can measure plate current pin 3 to pin 8? I didn't try it because I wasn't certain and didn;t know if it mattered which was red, which black etc.

I've run TAD 6L6GC-STR which are supposedly 35 watt rated if you can believe that. I had 200 hours on the original Chinese generics, which ran fine w/ no problems but were very ice-picky on some of the higher strings and kind of thin. The TAD's sound very good for 6L6 and I probably have 300 hours on them with not so much as a burp. They are advertised as being very tough and they do seem to be. I do not completely understand all the math but isn't 60mA at @ 440 or so volts pretty darn hot? How does this translate to plate dissipation? I don;t really care what 'class' the amp is supposed to be, but I mentioned it because I thought it affected how the plate dissipation was set. Again, though, I don;t really know what I'm talking about.

I popped in another set of KT66 *just to test it.* They are pulling 471 at pin 3, 471 at 4 and 40.6 at pin 8. So, using the math as per yours above, this translates to 61mA at about 430 working volts, correct? And about 26 1/4 watts per tube, or 25 1/4 if I subtract 1 W as you did. This is WAY too hot, am I right? My goal is to run KT66 in this thing, so the milion dollar question is, what do I need to do to do that? Is is merely changing the shared cathode resistor?

[EFK]

BTW, heaters: I can;t figure this out. I clipped to pin 2 and pin 7. W/ 6L6GC, the VAC when amp is turned on spikes to @ .9, then drops to .3 and stays there whether on or off standby. W/ the KT66, it spikes to 1.4 then drops to .4 VAC. This is w/ tubes in. This doesn't make sense to me????

[stokes]

You are running those KT66's at 100% idle.They are heading for certain death.Like MWJB says you need a bigger cathode resistor.I wouldnt recommend running any tube in or close to class A with 471 volts on the plate.Even the 375 @ 62ma's I mentioned is higher than I would recommend to anyone,I know,do as I say not as I do.But I am using what is probably the best made KT66 ever produced,they can take it,but I still wouldnt tell anybody else to torture their tubes like that.Especially the current production stuff made these days.As to your heater volt measurements,try checking with no tubes and see what you get.If you still get those odd results you are probably doing something wrong,if the readings are good with no tubes,put them back and try again.If you get the readings you saw the first time,maybe you have tubes with a lot H-K leakage.But its more likely you have one of the leads on the wrong pin.

[EFK]

I'll double check the heates again tomorrow. The tube sockets are numbered on the inside AND outside, though, and I'm pretty certain they were on 2 and 7. Would it make sense that two completely different sets of tubes had such leakage? Hmmm.

That aside, using a bigger resistor will drop the cathode current, right? Which will then lower the bias? How will this affect the plate voltage? Not at all or ??? I guess the question is, if plate voltage of 471 is unaffected, what do I need to lower the bias to to drop the current at idle - say 60 or 70%?

What do you all think of these tables:

http://www.diyguitarist.com/GuitarAm...erTubeBias.htm

The KT66 chart suggests that if the plate current or voltage (working current here is about 430, right?) is at 430, class A at I guess 100% should be 55.2 mA or 'High AB' whatever that means would be 43.6 mA. Or should I be using the raw 471 voltage reading? This lower mA number seems a lot safer to me, but again I'm struggling to understand all of this. For example, if I swapped out the 330 ohm resistor for say a 500 ohm, it would drop the mA at idle to about 40 - right? Or would this resistor swap change the cathode current so therefore I'd have to remeasure and recalculate?

Also: I assume the 25-26 watt per tube dissipation is really pushing it w/ KT66? Will this number be lowered also by changing the cathode resistor?

I VERY MUCH appreciate you guys bearing wth me here. I'm trying to figure this all out.

[MWJB]

Larger cathode resistor will push up the voltage...not a problem for the tubes, but it will alsop bring DOWN the current & plate dissipation - which the tubes will like.

Those tables are bollocks, forget yuo ever saw them.

Don't need to go as low as 60-70% plate dissipation, 80-90% (20-22W) is usually OK for cathode bias, 50mA-ish would be a reasonable target.

Your plate voltage is the voltage at pin 3, minus the voltage at pin 8, OR voltage measured at pin 3 with your black meter probe on pin 8 - the same thing.

"Also: I assume the 25-26 watt per tube dissipation is really pushing it w/ KT66? Will this number be lowered also by changing the cathode resistor?" Yes, as Stokes has already stated. Running a 25W tube (KT66 and most current production 6L6) at 25-26W is not a great idea.

Heaters - are you measuring pin 2 to pin 7 on the same socket? Are you measuring with the meter set to AC? Try measuring from pin 2 on one socket to pin 2, then pin 7 on the other power tube socket? If the amp makes a noise you must have heater voltage.

[stokes]

As current increases,plate volts will decrease,when current decreases,plate volts will rise.Its a kind of compramise, unless you have a 10 watt adjustable bias resistor.I like to tweak my cathode biased amps so that is what I use in them,500ohm wire wound adjustables,they cover just about all power tube varieties.If you cant find one get the two MW cited in his response and take it from there.Like he says,80-90% is good and will get you in class A area.Note I said class A "area",so lets not get into the "class A" definitions.

[EFK]

Again I can't tell you guys enough how appreciative I am for the help. I just ordered a bunch of resistors from Newark and I ought to have enough to get a variety of values in the 400-500 ohm range. One question, I can solder these in series can't I? For example, the 330 that's already in there with an added 80 ohm in-line (same 25 W rating) to get 410? Does it in fact work this way?

I'm not sure if it's necessary to bleed off the capacitors before doing this but I think I ought to be on the safe side. The resistor is easy enough to access but I'd like to account for stray fingers. I've read of two methods which would work easily enough for me, (1) running amp at power and then yank power cord while power and standby both 'on,' or (2) jumper from pin one of V1 preamp to ground. Are either of these methods considered acceptable?

[EFK]

Inside chassis. Someone might find this interesting.

[EFK]

Here's another curious point: 6L6 don't use pin 6. In fact, all the 6L6GC tubes I have completely omit pin #6 - none there. On this amp, the wiring is set up so that the wires to the control grid - pin 5 - use the pin 6 terminal as a soldering 'turret:' the wires are soldered to pin 6, then the grid resistor is jumpered from pin 6 over to pin 5.

Here's the thing: on a KT66, I was told that like 6L6, pin 6 also is not used. BUT, my KT66 tubes do have a pin at pin 6. Is this internally (within the tube) tied to anything or is it just a useless pin? If it is not tied to anything then this shouldn't be a problem but if it is tied to something in the tube then this amp's hot pin 6 terminal might cause an issue. ???

[Satamax]

http://www.aikenamps.com/WhatIsBiasing.htm

[Satamax]

Again I can't tell you guys enough how appreciative I am for the help. I just ordered a bunch of resistors from Newark and I ought to have enough to get a variety of values in the 400-500 ohm range. One question, I can solder these in series can't I? For example, the 330 that's already in there with an added 80 ohm in-line (same 25 W rating) to get 410? Does it in fact work this way?
I'm not sure if it's necessary to bleed off the capacitors before doing this but I think I ought to be on the safe side. The resistor is easy enough to access but I'd like to account for stray fingers. I've read of two methods which would work easily enough for me, (1) running amp at power and then yank power cord while power and standby both 'on,' or (2) jumper from pin one of V1 preamp to ground. Are either of these methods considered acceptable?

http://www.aikenamps.com/TechInfo_2.htm

[EFK]

Wow - that's a fantastic site! Thanks very much for the links - some great information there.

[EFK]

Alright if you all aren't too bored with me yet: I added a 100 ohm resistor which brought the total cathode resistance up to 430, and the bias settled at 50 ma with approx. 20W dissipation. Popped in a set of valve art KT66, put a fan at the back of the amp (need to keep the heat down with the 66, not a lot of room there...) and made some noise.

Those tubes lasted about 5 minutes.

50ma, 20W dissipation and @ 482 B+ seem to be well within the range for KT66 in an amp running in class a or hot ab, at least from the tables and info I have been able to find. So what is going on here? Are these new-made tubes THAT weak in comparison w/ the old Genelex? Stokes it would seem that you are 100% on the mark. I'm only hazarding a guess here, but this is a hot amp and I would suspect that it is the fairly high voltage that is killing these?

Dare I ask how hard it would be to lower B+ and screen voltage? Anyone?

[Satamax]

Alright if you all aren't too bored with me yet: I added a 100 ohm resistor which brought the total cathode resistance up to 430, and the bias settled at 50 ma with approx. 20W dissipation. Popped in a set of valve art KT66, put a fan at the back of the amp (need to keep the heat down with the 66, not a lot of room there...) and made some noise.

Those tubes lasted about 5 minutes.

50ma, 20W dissipation and @ 482 B+ seem to be well within the range for KT66 in an amp running in class a or hot ab, at least from the tables and info I have been able to find. So what is going on here? Are these new-made tubes THAT weak in comparison w/ the old Genelex? Stokes it would seem that you are 100% on the mark. I'm only hazarding a guess here, but this is a hot amp and I would suspect that it is the fairly high voltage that is killing these?

Dare I ask how hard it would be to lower B+ and screen voltage? Anyone?

Zenner trick??? Where's the schemo?

http://web.archive.org/web/200201060...com/Zener2.PDF

[MWJB]

Valve Art aren't a manufacturer, just a reseller...some Chinese KT66 could only handle 450v, don't know if this is relevant to you. Otherwise KT66 should easily handle the voltages you have...it's the current that's more dangerous.

Is that 482v measured from plate to cathode (pin 3 to pin 8)? If so, you are still giving the tubes a hammering current wise (0.050 * 482 = 24.1W). Try 680ohms for the cathode resistor.

How are you measuring plate current?

[EFK]

One problem I've run into is that I can't seem to find specs for new-made tubes anywhere. GT I don't believe ever published their tube specs but I can't find any Shuguang tube specs anywhere either, at least, not for the 66.

I just measured pin 3 to ground for B+ so if you subtract the cathode current it's only about 430-435 working voltage correct? This shouldn't kill anything, at least I wouldn't think it would. Voltage on pin 4 was 479. Heck I think JTM45 amps run this hot or hotter so how are the 66 surviving in those? Although, I've read quite a few recommendations for biasing in JTM45 at @ 35-40mA w/ 66s which initially sounds cold. ???

Could there be something else going on in this amp which is contrary to a 66 staying alive? As soon as this pair fizzled out and died (again, NO sign of redplate at all, and I was watching them - sound just 'went' like hitting the standby switch...) I let them cool off, pulled them and installed the old reliable 6L6GC once more. Played for an hour, no problem at all.