My partner Denise and I are tube enthusiasts, and my association with tubed gear goes back more than twenty years. I do all the repairs and servicing for our tube equipment and I treat tubes with great respect. That includes wearing cotton gloves when I handle them.
Late last year (2007) we decided it was time to replace the output tubes in our two power amps. It seemed a good time to buy in some spare phono stage, input and driver tubes as well, although none of the triodes really needed to be replaced.
For about four weeks I trawled the Net for opinions on current production tubes. Despite some bad press on a few forums, JJ’s KT77 tube looked interesting, and, at about 70 U.S. dollars for a matched quad from Bob Pletka at Eurotubes, seemed like good value for money. We decided to give the KT77 a try, and ordered a matched quad, along with a matched quad of KT88 s, a matched and balanced pair of Gold Pin ECC83S s, and an assortment of ECC81 s.
The tubes duly arrived and we plugged all of them in: some in a preamp, some in each of the power amps. I’d noticed immediately that the pins on each of the KT77 s were of different lengths, with some pins square cut, as though they were missing their ‘solder domes’. As well, at least one pin on each tube, and usually more, were noticeably out of vertical. This was in marked contrast with the KT88 s, all of whose pins were precisely vertical and of equal length, with neatly rounded ends: i.e., what I’m used to and what I’d expect in a tube from a reputable manufacturer.
Perhaps I’ve been lucky, but except for these KT77 s I’ve never seen even a single tube with seriously mismatched pins. I’m not a clairvoyant though, so I didn’t read anything sinister into this instance of poor build quality – but perhaps I should have.
After a few hours, one of the KT77 s showed itself to be unstable: a constantly increasing current draw, initially a rise equivalent to about 5 m.v. per hour (at idle across 10 ohms). I kept adjusting the current draw via the trimpot, checking the tube about once every hour, and for a few hours of playing the tube was usable. However, the rise eventually began to accelerate, to the point that the tube finally went from approximately 300 to over 500 m.v. within less than an hour. At this point I discovered that I’d run out of trimpot adjustment, so the tube was, effectively, unusable.
I reported the problem to Bob Pletka, who promptly dispatched a replacement – perhaps by mere coincidence a KT77 with an almost perfect set of pins. I installed this, and everything seemed to be fine. The KT88 s, meanwhile, had continued to behave themselves, as had all the new triodes.
After about 90 hours of use, one of the JJ Gold Pin ECC83S tubes, installed in the phono stage of a preamp, developed intolerable noise. I emailed Bob with the bad news and, again, he was quick to send a replacement. (By mutual consent this was two unmatched standard ECC83 s, since this preamp is quite indifferent to the brand or quality of the tubes it uses.)
A few hours down the track I noticed that two more of the KT77 s had become unstable, in much the same way as the first casualty had. Each was drawing a constantly increasing current, such that I needed to make trimpot adjustments about once an hour. Neither tube had gone into avalanche mode though, and this time I allowed myself to get my hopes up: maybe these tubes would plateau eventually, and well before I ran out of trimpot adjustment.
However, it was not to be. About 100 hours of playing time later, one of the two wayward tubes was no longer controllable by trimpot adjustment; I estimate that the other is only a few hours away from the same fate.
So, three out of five reputedly FACTORY MATCHED JJ KT77 tubes are, for all practical purposes, unusable: a sixty per cent failure rate.
On 8 July 2008 I emailed Bob Pletka: a detailed report on what had been happening to the KT77 s. He said he had no idea what the issue was; he added that he was currently shipping about 200 KT77 s a month with a failure rate of half a per cent. He said he would put me in touch with JJ by forwarding my email to them.
I eventually got a reply from JJ, relayed (and, I believe, translated) by Bob Pletka.
JJ reported that, although the KT77 is a drop-in replacement for the EL34, there is a minute number of amps in which that has proven not to be possible. Their theory, in respect to these exceptions, has to do with the fact that the impedance of an EL34 on average is about 15K and the impedance of a KT77 is about 23K.
The vast majority of JJ’s customer base is guitar players and the majority of their amps are made to use both EL34 s and 6L6GC s, which have an impedance of 22.5K. So in an amp that is designed to run either, the KT77, with its impedance of 23K, very closely matches a 6L6GC.
Even in amplifiers designed exclusively for EL34 s, JJ has found that one amp will be just fine whereas another will oscillate and after some time the bias cannot be stabilized via Ug1.
JJ claims that, as of March this year, it has modified its testing procedure for the KT77 to more closely match its specs rather than the specs of an EL34. Furthermore the electrodes by the original design of the KT77 are much smaller than those of an EL34 and JJ believes that this is what causes the oscillation when the impedance seen by the transformer is far enough away from the higher impedance of the tubes.
I’m a self-taught amateur, not a formally trained technician, but, frankly, I’m not convinced by JJ’s theory. So far I’ve found no trained technician who is either. In any case, keep in mind the following facts.
In the late 90’s I bought an allegedly factory matched quad of KT88 s, to be installed in an amp that is designed for them. Two of them had exactly the same runaway current draw problem as these recent KT77 s. Who made them? Tesla: the predecessor of JJ.
Three of the recent quintet of KT77 s have this current draw problem, but two don’t. Yet the five tubes are supposed to be precisely matched. In a bunch of precisely matched tubes, surely JJ’s impedance mismatch theory should affect them all.
What do I think of JJ tubes? What do I think of JJ?
First, the tubes. It’s early days yet (less that 100 hours of playing), but our JJ KT88 s seem to be a winner. They sound good, build quality is good, and I’m hopeful about their longevity. The ECC81 s, too, sound good and seem to have no problems that will lead to premature death.
The Gold Pin ECC83S I wouldn’t touch again. This tube is at the top of the JJ quality list: yet one of a matched and balanced pair died in under 100 hours. Had this been the only tube to give us trouble, I might have been persuaded that it was an instance of that less than one per cent failure rate that Bob Pletka claims for JJ tubes; but, of course, it wasn’t.
The KT77. Denise and I like the sound of this tube; but who wants to risk such a casualty rate? And the problems with JJ’s KT77 aren’t confined to just wildly mismatched – and for practical purposes uncontrollable – current draw. Other people have reported generally abysmal build quality. Although this seems to have improved a little (assuming our ‘quintet’ is representative), there’s no excuse for variable length, out-of-vertical pins so crudely finished that they must cause some damage to tube sockets. In fact, since JJ can get it right with their KT88, which has much the same pinout, the suspicion is raised that the KT77 s being distributed now are ‘factory seconds’: pieces that didn’t reach JJ’s quality control standards. If that proved to be the case it would mean consumers are being deceived, because, as far as I know, no retailer has ever presented the KT77 as anything other than a premium quality tube.
What about JJ the company? It’s obviously the right of other readers to decide what they think of JJ, but I’m disappointed and I’m suspicious. If JJ knows that even a ‘minute number’ of amps designed for the EL34 can’t use the KT77, why hasn’t this information been made available to potential customers? Imagine the furore if, say, a major auto company said nothing about a significant problem that it already knew existed – until and unless customers reported its existence! I don’t have enough confidence in this manufacturer to buy any of their tubes again. Obviously I won’t recommend them to others either. How much commercially smarter it would have been for JJ to come clean about any problems, as they emerged, and to level with purchasers about what they, the company, are going to do to sort them out.
If you’re a KT77 user, and IF YOU HAVE NO WAY OF CONSTANTLY MONITORING CURRENT DRAW, be very, very wary. If our experience is typical, there’s a good chance that one, even more, of your KT77 s are drawing seventy, eighty, perhaps a hundred milliamps: plenty high enough to lead to premature failure. And don’t assume that you’re safe because your amp is designed for the 6L6 as well as the EL34. JJ would have us believe it is, but why should customers believe a company that has (as far as I know) never once VOLUNTEERED information on ANY problem that it’s come across with the KT77?
I can be contacted at gwoods100@yahoo.com.au
Graham Woods
Melbourne, Australia
Late last year (2007) we decided it was time to replace the output tubes in our two power amps. It seemed a good time to buy in some spare phono stage, input and driver tubes as well, although none of the triodes really needed to be replaced.
For about four weeks I trawled the Net for opinions on current production tubes. Despite some bad press on a few forums, JJ’s KT77 tube looked interesting, and, at about 70 U.S. dollars for a matched quad from Bob Pletka at Eurotubes, seemed like good value for money. We decided to give the KT77 a try, and ordered a matched quad, along with a matched quad of KT88 s, a matched and balanced pair of Gold Pin ECC83S s, and an assortment of ECC81 s.
The tubes duly arrived and we plugged all of them in: some in a preamp, some in each of the power amps. I’d noticed immediately that the pins on each of the KT77 s were of different lengths, with some pins square cut, as though they were missing their ‘solder domes’. As well, at least one pin on each tube, and usually more, were noticeably out of vertical. This was in marked contrast with the KT88 s, all of whose pins were precisely vertical and of equal length, with neatly rounded ends: i.e., what I’m used to and what I’d expect in a tube from a reputable manufacturer.
Perhaps I’ve been lucky, but except for these KT77 s I’ve never seen even a single tube with seriously mismatched pins. I’m not a clairvoyant though, so I didn’t read anything sinister into this instance of poor build quality – but perhaps I should have.
After a few hours, one of the KT77 s showed itself to be unstable: a constantly increasing current draw, initially a rise equivalent to about 5 m.v. per hour (at idle across 10 ohms). I kept adjusting the current draw via the trimpot, checking the tube about once every hour, and for a few hours of playing the tube was usable. However, the rise eventually began to accelerate, to the point that the tube finally went from approximately 300 to over 500 m.v. within less than an hour. At this point I discovered that I’d run out of trimpot adjustment, so the tube was, effectively, unusable.
I reported the problem to Bob Pletka, who promptly dispatched a replacement – perhaps by mere coincidence a KT77 with an almost perfect set of pins. I installed this, and everything seemed to be fine. The KT88 s, meanwhile, had continued to behave themselves, as had all the new triodes.
After about 90 hours of use, one of the JJ Gold Pin ECC83S tubes, installed in the phono stage of a preamp, developed intolerable noise. I emailed Bob with the bad news and, again, he was quick to send a replacement. (By mutual consent this was two unmatched standard ECC83 s, since this preamp is quite indifferent to the brand or quality of the tubes it uses.)
A few hours down the track I noticed that two more of the KT77 s had become unstable, in much the same way as the first casualty had. Each was drawing a constantly increasing current, such that I needed to make trimpot adjustments about once an hour. Neither tube had gone into avalanche mode though, and this time I allowed myself to get my hopes up: maybe these tubes would plateau eventually, and well before I ran out of trimpot adjustment.
However, it was not to be. About 100 hours of playing time later, one of the two wayward tubes was no longer controllable by trimpot adjustment; I estimate that the other is only a few hours away from the same fate.
So, three out of five reputedly FACTORY MATCHED JJ KT77 tubes are, for all practical purposes, unusable: a sixty per cent failure rate.
On 8 July 2008 I emailed Bob Pletka: a detailed report on what had been happening to the KT77 s. He said he had no idea what the issue was; he added that he was currently shipping about 200 KT77 s a month with a failure rate of half a per cent. He said he would put me in touch with JJ by forwarding my email to them.
I eventually got a reply from JJ, relayed (and, I believe, translated) by Bob Pletka.
JJ reported that, although the KT77 is a drop-in replacement for the EL34, there is a minute number of amps in which that has proven not to be possible. Their theory, in respect to these exceptions, has to do with the fact that the impedance of an EL34 on average is about 15K and the impedance of a KT77 is about 23K.
The vast majority of JJ’s customer base is guitar players and the majority of their amps are made to use both EL34 s and 6L6GC s, which have an impedance of 22.5K. So in an amp that is designed to run either, the KT77, with its impedance of 23K, very closely matches a 6L6GC.
Even in amplifiers designed exclusively for EL34 s, JJ has found that one amp will be just fine whereas another will oscillate and after some time the bias cannot be stabilized via Ug1.
JJ claims that, as of March this year, it has modified its testing procedure for the KT77 to more closely match its specs rather than the specs of an EL34. Furthermore the electrodes by the original design of the KT77 are much smaller than those of an EL34 and JJ believes that this is what causes the oscillation when the impedance seen by the transformer is far enough away from the higher impedance of the tubes.
I’m a self-taught amateur, not a formally trained technician, but, frankly, I’m not convinced by JJ’s theory. So far I’ve found no trained technician who is either. In any case, keep in mind the following facts.
In the late 90’s I bought an allegedly factory matched quad of KT88 s, to be installed in an amp that is designed for them. Two of them had exactly the same runaway current draw problem as these recent KT77 s. Who made them? Tesla: the predecessor of JJ.
Three of the recent quintet of KT77 s have this current draw problem, but two don’t. Yet the five tubes are supposed to be precisely matched. In a bunch of precisely matched tubes, surely JJ’s impedance mismatch theory should affect them all.
What do I think of JJ tubes? What do I think of JJ?
First, the tubes. It’s early days yet (less that 100 hours of playing), but our JJ KT88 s seem to be a winner. They sound good, build quality is good, and I’m hopeful about their longevity. The ECC81 s, too, sound good and seem to have no problems that will lead to premature death.
The Gold Pin ECC83S I wouldn’t touch again. This tube is at the top of the JJ quality list: yet one of a matched and balanced pair died in under 100 hours. Had this been the only tube to give us trouble, I might have been persuaded that it was an instance of that less than one per cent failure rate that Bob Pletka claims for JJ tubes; but, of course, it wasn’t.
The KT77. Denise and I like the sound of this tube; but who wants to risk such a casualty rate? And the problems with JJ’s KT77 aren’t confined to just wildly mismatched – and for practical purposes uncontrollable – current draw. Other people have reported generally abysmal build quality. Although this seems to have improved a little (assuming our ‘quintet’ is representative), there’s no excuse for variable length, out-of-vertical pins so crudely finished that they must cause some damage to tube sockets. In fact, since JJ can get it right with their KT88, which has much the same pinout, the suspicion is raised that the KT77 s being distributed now are ‘factory seconds’: pieces that didn’t reach JJ’s quality control standards. If that proved to be the case it would mean consumers are being deceived, because, as far as I know, no retailer has ever presented the KT77 as anything other than a premium quality tube.
What about JJ the company? It’s obviously the right of other readers to decide what they think of JJ, but I’m disappointed and I’m suspicious. If JJ knows that even a ‘minute number’ of amps designed for the EL34 can’t use the KT77, why hasn’t this information been made available to potential customers? Imagine the furore if, say, a major auto company said nothing about a significant problem that it already knew existed – until and unless customers reported its existence! I don’t have enough confidence in this manufacturer to buy any of their tubes again. Obviously I won’t recommend them to others either. How much commercially smarter it would have been for JJ to come clean about any problems, as they emerged, and to level with purchasers about what they, the company, are going to do to sort them out.
If you’re a KT77 user, and IF YOU HAVE NO WAY OF CONSTANTLY MONITORING CURRENT DRAW, be very, very wary. If our experience is typical, there’s a good chance that one, even more, of your KT77 s are drawing seventy, eighty, perhaps a hundred milliamps: plenty high enough to lead to premature failure. And don’t assume that you’re safe because your amp is designed for the 6L6 as well as the EL34. JJ would have us believe it is, but why should customers believe a company that has (as far as I know) never once VOLUNTEERED information on ANY problem that it’s come across with the KT77?
I can be contacted at gwoods100@yahoo.com.au
Graham Woods
Melbourne, Australia
I have heard other stories on the net about folks who have had problems with the JJ stuff at one time or another. Yet there are other folks who seem to have great luck with their product and who never have problems. I wonder if most of this is batch related in that some of their batches of tubes are good and some are not.
My own personal results with their stuff have been very good. I've been using their tubes in guitar amps since 1993 and I can't think of a single failure we've had that's been tube related. But my experience comes from using only one tube type from them the E34L in amps that do not subject them to very much voltage stress and generally run at around 450volts B+. But these amps do get played a lot and moved around to band gigs so I do tend to change the output tubes about every 6months to a year max depending on the type of use they get.
I wonder if the KT77's structure and build just isn't all that well sorted out yet and they will get better over time, I hope so....
I just did an edit to my post because I found some interesting info on the KT77's ratings.
According to the chart in the book which comes with the Groove Tubes bias tool the KT77 is listed as having the same max voltage as the 5881, KT-66 etc and NOT the same as EL34 which is listed as being able to go all the way to 800 volts. I'm wondering if this is one of the reasons some folks have problems with the KT77 in some EL34 circuits and others do not since the ratings are not the same as EL34.
My own personal results with their stuff have been very good. I've been using their tubes in guitar amps since 1993 and I can't think of a single failure we've had that's been tube related. But my experience comes from using only one tube type from them the E34L in amps that do not subject them to very much voltage stress and generally run at around 450volts B+. But these amps do get played a lot and moved around to band gigs so I do tend to change the output tubes about every 6months to a year max depending on the type of use they get.
I wonder if the KT77's structure and build just isn't all that well sorted out yet and they will get better over time, I hope so....
I just did an edit to my post because I found some interesting info on the KT77's ratings.
According to the chart in the book which comes with the Groove Tubes bias tool the KT77 is listed as having the same max voltage as the 5881, KT-66 etc and NOT the same as EL34 which is listed as being able to go all the way to 800 volts. I'm wondering if this is one of the reasons some folks have problems with the KT77 in some EL34 circuits and others do not since the ratings are not the same as EL34.
From what Jim McShane says (and he's a knowledgeable and trustworthy guy), it's very dependent on what tube you're talking about. I have gone through about 20 of their EL84 with zero problems. Ditto the ECC81; not only have all 24 I've bought, tested, and used performed without problems, they also show much better audio performance than any other ECC81/12AT7/6201 I've tried. Jim has reported similar results as mine for those tubes, but also has had issues with some of the other power tubes.
I have found JJ EL84's to be reliable and great sounding, my favorite EL84's of the 3 brands I have tried. I also use their 12AX7 and 12AU7 and EL34S in a PP UL EL34 amp. No issues after ~2 + years. The EL34S has ~525V on the plate, and I suppose similar on the screen, I am running them pretty close to the 25W rating and have no issues with any of them.
Also use the ECC99, no problems there either.
From what I read there appears to be issues with the KT77 for most but not all users, thats why despite what people say about them sounding great I avoid them.
Andrew
Also use the ECC99, no problems there either.
From what I read there appears to be issues with the KT77 for most but not all users, thats why despite what people say about them sounding great I avoid them.
Andrew
SY said:
Hi SY!
Yes, you nailed it. The JJ EL84 is fine. I don't used the ECC81 or ECC82 as I have excellent NOS supplies, so I have no real experience. Their 12DW7 is fine. The ECC99 is okay, not real consistent but low in failures. And surprisingly, their E88CC has improved remarkably; it's a truly fine tube now!
But the JJ octal power tubes and the GZ34 are generally poor quality. And their ECC83/12AX7 family is not good either. I'm not saying you can't find at least a FEW good octals or ECC83s, I'm saying the initial build quality and/or the reliability is so bad I can't afford to carry them.
Also, the reaction from the JJ factory managementwhen I informed them of the quality issues was to insult my intelligence. Compare that to this email I got back from New Sensor's main man, Mike Matthews, when I let him know about the batch of KT-88s that were troublesome (I sent him some links to read too):
"Jim,
I DO appreciate your sending me these sites and for having an interest in us supplying the best quality tubes possible.
I will immediately forward your email to our tube engineers in Russia. You can be assured we will do our best to rectify any problems.
Thanx again,
Mike"
The Genalex reissue KT-77 has been entirely trouble free for me, and I've sold nearly 200 of them. Not a single dud or field failure. The same goes for the KT-66. The KT-88 had a batch with some issues, but those have been taken care of and it's back to being very good quality.
BTW (slightly OT), I guess the 12AT7 family will have to be our secret. IN THE RIGHT CIRCUIT it does things no other small signal tube can do, it's amazing. In the wrong circuit it's not very good.
Andrewbee said:
The problem is that the next set of EL34 tubes you get from JJ may well be horrible. You got lucky, many/most people (like the original poster) aren't as lucky.
I hope not, I bought a quad of their Blue EL34S tubes at the same time (they are so pretty 🙂) that I will use when the present ones eventually wear out.
There is always New Sensor as you say.
Andrew
Andrewbee said:
I'm not saying Mike and New Sensor are saints, but they seem to do a good job all things considered. JJ USED to, but not anymore.
I just have to ask, if you get some brand new octal tubes with crooked/square pins, why would you put them in your amplifier? You can severely damage the pin receptacles in your tube sockets and render the socket(s) useless! (like putting motor oil in your car that comes with metal shavings)
The JJ KT-77 has repeatedly come up as being pretty junky. A small difference in impedance is just not significant enough to cause the drifting bias, that's really only an issue of a poor quality tube (like a bent grid pin that doesn't always contact well).
You're lucky you didn't order the special EF86 for anything, that is an amazingly noisy tube; I think the gold plating on the pins may actually make them noisier.
The JJ KT-77 has repeatedly come up as being pretty junky. A small difference in impedance is just not significant enough to cause the drifting bias, that's really only an issue of a poor quality tube (like a bent grid pin that doesn't always contact well).
You're lucky you didn't order the special EF86 for anything, that is an amazingly noisy tube; I think the gold plating on the pins may actually make them noisier.
Hi there !
I never used KT77 as an EL34 substitute, but let me point at least one difference.
Data sheet states a maximum grid leak resistor of 500K for the EL34 vs 250K for the KT77.
This limit is frequently overlooked by more than one designer.
Do not forget that this is listed in "maximal limits" and that, in a grid biased amplifier, the TOTAL resistance of the bias source MUST be taken into account, not only the grid leak resistor itself.
Moreover, some designs use serie pass regulators in the bias circuitry wich are completly unable to sink any current from the grids ant thus they soar, inducing thermal runaway.
Of course, one batch may be more sensible than another one to this phenomenon while still not being out of specs.
My two pences.
Yves.
I never used KT77 as an EL34 substitute, but let me point at least one difference.
Data sheet states a maximum grid leak resistor of 500K for the EL34 vs 250K for the KT77.
This limit is frequently overlooked by more than one designer.
Do not forget that this is listed in "maximal limits" and that, in a grid biased amplifier, the TOTAL resistance of the bias source MUST be taken into account, not only the grid leak resistor itself.
Moreover, some designs use serie pass regulators in the bias circuitry wich are completly unable to sink any current from the grids ant thus they soar, inducing thermal runaway.
Of course, one batch may be more sensible than another one to this phenomenon while still not being out of specs.
My two pences.
Yves.
The grid leak would be worth looking at, but with gold standard of coupling cap size seeming to fall in the .1uf to .22uf area these days, anything above 250k seems unlikely (although not impossible). With fixed bias, I'd gander that it's generally a bit lower.
audiowize said:
I'm with you on this one. Lower g1 circuit resistance is the way around lots of gas-induced grid current( see type 50 for worst case ). Take a look at 6GB5 for the opposite extreme...🙂
cheers,
Douglas
Bandersnatch said:
The 50 is so extreme in fact that the only safe way to drive it is probably with an IT.
In general modern output tubes seem to be gassier than their earlier counterparts, and designers who deliberately violated grid circuit resistance specs without grid current induced runaway established a precedent that should no longer be considered acceptable due to the quality issues with modern tubes.
My recent experience with JJ KT88 has not been all that great, and I have no trouble believing that QC and manufacturing process issues at JJ are contributing greatly to the current crop of problems with the KT77 even in properly designed circuits.
JJ is one of the few low cost suppliers that is located in a politically stable and relatively free democratic country which is more than I can say for most of its competitors. I'm a bit concerned that in a worst case scenario they could end up as the whole game.
kevinkr said:
A grid choke will serve the same purpose I suspect. Now that I have thought of that, I am missing 3 quads of NOS 8417 that did a slow run away on 150k Ohm grid resistors( fixed bias ). They'd be perfectly fine with the current grid choked amp. talk about a D-Oh! moment...🙂
cheers,
Douglas
I don't know about hi-fi applications for JJ, but I have run them exclusively in my guitar amps for 6 years or so with ZERO problems. I run them pretty hard in my Peavey 5150 combo ( it takes 5 12AX7's and two 6L6GC's) and have never had one fail, even through years of doing gigs, banging the amp around, and the fact that the speakers vibrate the heck out of the amp. In fact I've been running the same tubes for nearly 3 years and they still sound great. I should swap them out, but am loathe to since they sound so good. Maybe it's just certain types they're having issues with?
- Status
- Not open for further replies.