KT120 transformer failures


I know about the warning regarding the KT120 filament current draw posing a possible issue with some amps, but my question is: has anyone actually experienced this issue in a modern amp?
wolf_garcia
Interesting post there Trelja...My Sovtek "brand" kt88s are produced at the same New Sensor factory as my sets of Tung Sol 6550s and kt120s, and they (the kt88s) sound GREAT...upon close physical inspection using my watery and often reddened eyeballs they do seem to have very similar "innards" to the 6550s, but sound and behave differently nonetheless, as well they should. I find it hard to imagine why a factory would make a poorly regarded tube when it also makes very highly regarded tubes (they're the ONLY ones making the universally accepted kt120s) in the same place. From the Tube Store:

From our KT88 tube type listening tests:
"The Sovtek KT88 tube has smooth, warm and clear with excellent imaging. The bass (except for the very bottom end) is tight and the high end is crystal clear. The separation of parts and balance is better than any of the 6550 tube types."

They left out (assumed typo) what IS "smooth, warm and clear"...the glass? The sales staff? The skin on the factory manager's mistress? Mine don't run with "red plates" or explode...although they do have a beautiful blue glow that appears after a while. Also, I just googled 'em and they seem to be more or less accepted by "geekdom". All this aside, for reasons I find hard to explain I am currently preferring the kt120s...if only because they're larger and seemed to become depressed when I put them in their little boxes to try other stuff. The sets of 6550s and kt88s are in a large box together so they can discuss their early days spent on the Volga. I will not let them near any vodka.
Sovtek had made a decent living producing the worst sounding tubes in the industry. The surprise would be if you didn't find you're happier now with something else.

A lot of OEMs, who keep the brand in business, use Sovtek because of the pricing and their general reliability. Of course, these are not the worst as they are a KT88 - 6550s beat them in that regard.
My KT88s were an impulse purchase at a local "DIY" place and they're sets of "matched pair" Svetlanas...I think Svetlana has an OK reputation and in any case they weren't expensive (same Russian factory also). I think the KT120s have a little more clarity or something...bigger sound sort of, but the 88s do sound good (as do the 6550s). I guess I'm good for YEARS of power tubes now. Good luck with the mods, and it feels good that another Jolida tech thinks the 120s are OK as that was a hard issue to nail down. It's interesting how the 502 just keeps impressing me, and now I suppose I'll be looking at 12ax7/at7 rolling...it's endless and like it!
Hey Wolf
I got the KT-120s but my 502p is @ my local Jolida dealer getting upgraded. The guy is a tube expert so I trust him with modding the amp. In another couple of weeks I'll be able to enjoy the music as well as the buring smell you mention from the Jolida & the KT-120s. Thanks for the heads up on the buring smell. I probably would have been worrying about that. Which KT-88s did you try? I have my eye on the Penta KT-88SC tubes just in case if the KT-120s don't work out. Also when I spoke to the dealer/tech he said the Jolida 502p could handle the KT-120s w/ no problem. I can't wait to get up & running w/ everything.
Jedinite24...did you stick the KT120s in yet? Mine keep getting better...not sure if they've peaked yet, but they likely have. Also, I discovered that some new tubes have a "new tube burning smell" that I thought might be the Jolida frying until I tried some new KT88s that had a "burning glass" sort of smell for a day or 2...just like the KT120s.
Hi Johnsonwu

I shall offer my reasoning for posting multiple questions about TCtubes throughout the forum here. It is not as funny as Wolf's though. When trying to price out places that sold the Tung Sol KT-120 and found that they offered a service called: Precision Power Tube Matching

Direct from the TCTubes website:

"Our power tubes are burned in for 48 hours and tested on a laboratory grade Amplitrex AT1000 tube tester. We match tubes for both current draw (mA) and transconductance (Gm) at typical operating conditions (see #3 below) for no extra charge. We can also custom match power tubes at whatever specifications you provide (see #4 below)."

They then ask for amp maker/model, how to bias the amp and finally the plate and screen voltage for a custome match. I'm new to tubes and had never seen this before. I was trying to get information from members but since I never really know what the moderators will put through I have to post questions in different threads.

I'm new to tubes and don't know if this precision matching will really make a difference. I'm trying to gather info about it. I have no affiliation with TCTubes whatsover.

For me it is now down to a cryo treated platinum factory matched quad of KT-120s or a precion matched quad of KT-120s from TCTubes.
The KT88 and 6550 draw 1.6A by their heaters (filaments)
The KT120 draws 1.9A by it's heater (filament)

As you can see the KT120 is higher and will tax the heater (filament) winding of the power transformer if it was not designed to take the extra 300mA. It very well over time cook the heater (filament) winding of the power transformer and burn it out, (fire is a possibility), which will be expensive to fix because the heater (filament) winding is a small part of the mains power transformer that supplies the HT (high tension) as well to the tubes, and usually the largest transformer on your amp.

Cheers George
I wondered the same thing, but went with a standard factory matched quad without "cryo" treatment just to be cheaper...and now I'll never know! Supposedly the fact that you bias each tube individually makes it less important to have them so carefully matched (according to the manual), but I'm no expert. I did read more about the "cryo" treatment later though and it seems interesting. My amp is still sounding fine, and better actually...KT120 tube break in? Maybe get a custom matched set of KT120s and send 'em out for cryo treatment someplace...your local college science lab might do it for free!
What's with all the posts about TCtubes all of a sudden everywhere on every thread that has to do with tubes?
Prices being equal. Would it be better to have tubes custom matched by TC Tubes or cryogenically treated by Cryoset? I saw both for about $200/quad. Factory platinum matched was about $25 less. I don't think either process would affect how the tubes work with the transformers of an amp.
Run the CL120 half power with 4 kt120 each mono block and you can have your kt120s.
Wilson...I looked up "tctubes.com" and now I see why you asked, and I hope you find out and let me know. Although I will feel bad that MY tubes weren't as "custom matched"...damn!
Nope...but you should feel free to call them and bet they'll give you the info you want. I merely described to them the "filament headroom" warnings that I kept noticing in the KT120 descriptions and they said there's plenty of transformer headroom built in. I've been using KT120s for a few weeks now and they sound great with no issues, although I have no idea what the "issues" might be (melting? radioactive waste?). I'm using 2 factory "matched" pairs...otherwise nothing special about them.
Hey Wolf_garcia
in all you talks with the folks at Jolida did you happen to get the specs on the Plate,Screen and Grid Bias voltages. I am looking into getting some kt120's custom matched at tctubes.com.
Damn...that's such sad new for CL-120 users everywhere. I hope they can go on somehow to find meaning in this otherwise tube specific world we share as they stand in the cold shadows with their frostbitten noses pressed against the window hopelessly watching others who in fact can, and do, use KT120s.
Unfortunately, the CL-120 is not on ARC's list of products that can accept the KT-120.
@ Wolf Yes a CL-120 or Classic 120 was a tube amp made by ARC. @Smoffat, I don't believe the top cover will fit if KT-120's are installed.
The Roswell crash proves that interstellar space aliens should NOT be using Russian tubes (back then from OUR future no less) without robust transformer filament windings.

I have no idea what "Smoffat" is, or what a CL-120 is unless you mean the old ARC amp.

I had a 1971 Honda CL450 for years, but that was a double overhead cam "street scrambler" motorcycle...should have kept it. No tubes, except for the tubular frame and shocks.
Wolf, once while speaking with a UFO expert discussing the Roswell crash, according to him the ship that crashed at Roswell was in fact using KT120s in their navigational system....
@Wolf...there's always a first time. @Smoffat if I get a chance I will talk to the factory about the CL-120/KT-120 issue and see what they say. I can tell you that the CL-120 has triode configured output tubes. The Ref 150 has a different tube compliment and does not run the output tubes in triode. I believe it runs the outputs in ultra linear configuration so it will sound different. Also, the CL-120 uses servos to set bias on half the output tubes and the cage must be removed in order to set bias. The Ref 150, I believe, allows user adjustment without going inside the amp.
According to my research (kinda lame...consists of merely asking on these forums and talking to some "tube head" designers and "experts") there has never been a transformer failure in any amp in the history of the world from using KT120s. The entire world...all of it. That's what my research has shown to this date.
Hifigeek, Have you or has ARC (to your knowledge) re-tubed a pair of Classic 120's with the KT120 without modding the tranny. 120's with 120's that should be a match in heaven.
However we're talking 8 output tubes per amp and that could put a major load on the transformer. Could we have a complete meltdown of the transformer. Would this affect the 6FQ7 driving those tubes.
Space is not an issue with mine as i operate them without the cages on.

In all honesty, although those have been modded by Steve Huntley in 2005, i think i would get a better bang for the buck (sonically speaking) if i replace those with a REF150. I need to get a "loaner" from a dealer and do an A/B comparison in my system. A 150 would likely pair better with my REF3 LE and REF2 Phono and Verity Parsifal speakers than the 120's.
Have been an ARC fan for 25 years and have had the 120's for 22 years, replacing a CL60. Analog front end? Well, too many to count. If i had kept them all, my house would have turned into an ARC Museum.
Luckily there seem to be NO transformer issues with the 502P. Not yet anyway, and I assume if it was gonna die it would show some signs of impending failure. My trusty warranty is there just in case. The CJ "SE" is on a price level that begs comparison to "things that interest me" like the Cary 120SII, Audio Research VS 115 (now with 8 KT120s!) McIntosh MC275, and other stuff, but the Jolida sounds so damn good and provides me with the requisite space heater and tube glow features that apparently I now need, so I'm content with it. At around a grand new there's not a lot out there that compares.
The newer ARC products have been designed with higher current ratings on the filament winding expressly for the purpose of using the KT-120. A few earlier ARC pieces, by virtue of their robust filament windings can also take advantage of the KT-120's. There are other ARC amps that can never use this tube do to issues where the length of the tube creates issues, i.e. cover will not go on or transformer in the way.
Pretty much every KT120 sales spec/note/description including the "official" one from New Sensor has that "tube heater current filament supply or" warning. I stand warned. Audio Research along with a bunch of other amp builders have modified their transformers to handle this issue and I still had to ask...is there a melted amp someplace? I bias the KT120s to the Jolida 500mv specs (I use a voltmeter even though the Jolida "ez bias" LED thing works fine) and the KT120s sound significantly stronger in every way over the otherwise fine stock (also Tung Sol) 6550s. I even talked to Kevin Deal at Upscale Audio and he said the KT120s are selling like hotcakes and he's never heard a customer mention an actual transformer overload issue...so much for my little research project.
Hey Wolf

I can't help you answer your query but maybe the gents over at Conrad-Johnson can. The CJ Classic Sixty SE Amplifier utilizes KT120 tubes throughout. Maybe they saw the current draw issue you speak of and overcame it. From stuff I've read about them they are really good people over there and who knows they may be willing to chat with you about the KT120 transformer issues.

Info about the Classic Sixty SE.

http://www.conradjohnson.com/It_just_sounds_right/Classicsixtyse.html
I have not heard of KT120 filament failures as you describe them nor any of the warnings you mentioned. I believe an autobias amplifier that does not have adequate plate voltage or current may experience this issue if it cannot supply adequate voltage to power the tube in the socket. Some amplifiers are set to bias with a standard tube range made for that amp, for example, my amp is optimized for KT88's, however, I can use the KT120 with my amp and when I adjust the bias manually, I only adjust for the tolerance built in for the KT88 the amp was designed for. I can, and I do, when I use a bias measurement tool, raise the ma for the KT120's, not by much though. The stock adjustment for the KT88 is around .40 and with the 120's, I have tried .55-.65 with no problems.

I did experiment using KT120's with a .75ma bias setting and that experiment was a disaster. The tubes turned black and one of the bases of the tubes separated. I have since learned to live with the lower ma settings and the stock range of the KT88's. The KT120's are lovely though. They do have a pronounced deeper bass and an even frequency response. I hope my information helps you some.

Ciao,
Audioquest4life