[TheChairGuy]

A PM between Dr. DMason and myself recently prompted me to solicit additional opinions on the following subject of tubes....

Over time I have found reliability of the old tubes to be better than the new stuff.  Tho it's great to have them matched, and tested for low noise and microphony (at fair prices)...new tubes have not faired well with me over time.

I just bought some low noise/low microphonic and matched 12AT7's (JJ's) from Tube Depot...I paid $6 extra per tube for the testing.  Well, either they ain't testing, or the JJ tubes suffer from QA issues, or if Tube Depot tests them they ain't testing for everything....as at least two (of 4) are defective 

They were to go into four 12AU7 slots for higher gain in my amp....they have lower gain than the NOS Sylvania's there now.  Noise? Ugg.....a hum as loud as any I've heard.

I've had problems with Sovtek, EH, Chinese and JJ's now.  NOS, or even gently used tubes - never (and I've probably bought equal amounts over time of each).  So, I think I'm done with newly manufactured tubes as much as I can avoid them. Well, maybe one (fantastic sounding) Raytheon 6922 problem I had once...but I remain convinced it was because of that silver bearing super grease migrated and fried it (I only use Caig now, it's safe for tubes).

I have even bought used tubes (ebay, audiogon or tubes that came with amps and preamps) and the old stuff keeps chugging on until they die their natural death of age and use...and maybe half the new ones go prematurely.  I've already given up on anything but NOS rectifiers...seems the new ones just cannot do the job without blowing.

As for cryo.....not sure if it works or not.  I've never had a cryo'ed and non-cryo'ed tube of the same matched sections, low noise and microphony to pick a winner among them.  So, the benefits of cryo'ing remain elusive to me.

I'm not even delving into the issue of whether NOS or new sound better (feel free to expound on that if you want - I'm staying away from that topic for now).

Does anyone just buy JAN's (Joint Army-Navy) for better reliability on the old stuff?

Anyhow gents - thoughts and perspectives out there?

John

[richidoo]

Hey John! Howyadoin?
I bought the JJ 12AT7s, tube depot, gold pin, matched with balanced triodes, based on raves on asylum and elsewhere about the super low distortion of that particular tube.  Indeed, it blew away the stock EH on my amps, but it didn't last. Windy noises soft and sometimes loud started occuring after a month. Eventually I put the EH back in. Not as smooth and sweet but quiet and reliable. 9 months later I put the JJs back in... Ahhhh so nice!!  Funny thing, they stayed quiet for a month (again) then started up with the same noises. Out they came, EH back in.  There's a reason the mfg chose EH. Just as there's a reason asylum "experts" rave about JJ 12AT17.  Similar experience with JJ KT77. Loud Windy noises and crooked glass, typical JJ. But they weren't even close to the excellent low distortion sound of the AT7.

Another... the stock Groove Tubes E34Ls tubes that came with the amps were great. Quiet and powerful for 2000+ hours as expected. Replacing them with another batch of new, mfg screened, GT screened, matched tubes I had 3 die within 200 hours. Mfg is aware of lousy batch consistency and redesigning tube matching rig to filter out the junk. The GT tubes are also made by JJ under contract.  No other tube sounds as good on my amps. SED is usually very reliable but just doesn't have the love. Once again JJ wins with sound, but loses the sale for quality. 

Nowadays I'm considering building Transcendent THE BEAST but there is only one tube for it, JJ EL519. The tube is gently run in this design and people have reported NO problems with the tubes aside from occasional dud.  But it's something I think about before buying 32 of them!

Last story, I run a pair of AU7s in my minimax. I never even tried the stock EE branded chinese tubes, just left them in the foam while I farted around with vintage AU7s. They all had tubular euphonia or tonal inaccuracy of some kind. I wanted clean and low distortion. By chance I was reminded of the stock tubes and tried them in the minimax - BINGO!  My system came to life, and the vintage AU7 signal tubes are packed in the drawer, while the TungSol rectifier bought from Bill at Morningstar made a huge difference. After 500 hours the Chinese stock EE branded 12AU7s are behaving perfectly. China is making some excellent tubes. A certain 300B made in China is suppsed to be superb, an article in TAS mag featured the tube mfg a few months back. A guy who degreed in tube mfg in 1950s and has been perfecting tube mfg plant ever since, a real craftsman, but he's old - what will happen to the plant and production quality when he is gone?

So I guess you really gotta try and see, but don't be afraid to try Chinese, and don't be afraid to question the wisdom of chasing high vintage tube prices. It ain't always worth it.

ARC uses SED 6550 tubes and has developed tube screening process that is very elaborate to increase the tube quality above mfg standards for their customers. But a retube at the factory is a very expensive proposition! I found these tubes to be extremely reliable and great sounding when I had an amp that used them. With the Russian political situation looking grim, we could be looking at a Russian tube shortage soon. JJ and China will be on the spot to pick up the slack. I hope it never comes to that.

I am looking for a ss amp that sounds as good as tubes (or better) and get me out of this nut race. Some hopeful leads, but not struck gold yet. One thing I've learned it won't be cheap. Til then I put up with them for the great sound.
Rich

[TheChairGuy]

Yeo RichieRich (ie, SuperDad ) -

Thx on the reply.

I never buy super expensive NOS types...I think the most I have ever paid for a NOS (or used) tube is $20 each.  With so many new ones out there for half the price I never thought it was prudent to pay much more for the oldies.  I know - that in itself may not be a smart thought - but like you say, the money pit that is a tube amp (or preamp) can get mighty deep if you buy pricey oldie tubes all the time 

So, you're saying you have had bad luck with some new tubes (as have I) for quality control.  Beyond discussing any sonic properties of any tubes....I just keep find a 50% defect rate on newbie tubes and near 0% on the NOS/oldies.

I'm at a point where I want to not consider buying any more new tubes and only buy NOS ones and take my chances sonically and for low noise/microphonics.  It has nothing to do with snobbiness or sonic primacy...it has to do with my mental health (which always borders on suspect  )

Perhaps a nicely equipped hybrid amp is in our future...one with highly regulated power supplies for longlasting tube life?

That's probably an ersatz ad for van Alstine gear 

btw, I am not against Chinese anything.  I have seen amazing advances in everything they make there over the past 12 years sourcing there.  If ever there is hope for a better new hope in the future that is not costly - it would be China for me. Quality control in Slovakia and Russia seems a foreign concept 

John

[randytsuch]

Hey John
I read somewhere recently, I think at the Antique site, that the guitar amp makers were the driving force in tubes these days, as they buy the majority of tubes, far more than audio people.

They are driving the tube makers to low cost, so it's not surprising that they don't last as long.
I will look for the link later, when I get a chance

Randy

[richidoo]

So, you're saying you have had bad luck with some new tubes (as have I) for quality control.  Beyond discussing any sonic properties of any tubes....I just keep find a 50% defect rate on newbie tubes and near 0% on the NOS/oldies.

I agree, the vintage tubes seem to be more reliable, but my only bad experiences have been with JJs. And yet some JJs are just awesome. New sensor and Chinese new production have been quality and reliable. Quality was a big part of the culture of tube mfgs back in the glory days. But I am not buying any more vintage tubes, the flavor of euphonic distortion in the major brands drives me nuts. For power tubes NOS can be very pricey and newer tubes aren't so bad, some even good.

[Dmason]

is there a concensus with new production EL84, and their Russian variants? What about these TAD bottles that are "different?"

The opinions on the JJ El84 are varied in the extreme. Considering the pricing, 2000+ hours doesnt seem all that unreasonable..

Ebay pricing on this stuff, and especially www.tubeworld.com   are right over the top. I simply cannot JUSTIFY those prices.

[TheChairGuy]

Randy - long time no hear, brutha'    I'd think musicians amps would pound the hell out of tubes and they'd be only more picky as to a tube holding up (who needs a tube going south during a concert or recording?).

I think our salvation will be Chinese tubes...they are not going to leave a market with high priced competitors that sell bad products.  They will, if not already have, sense the opportunity and exploit it I think (at least that has been the case with scores of other prodcuts I have been involved with over time)

Rich
I'm lucky - my output tubes are odd ones - 6CD6's (old TV tubes).  Tho they are not prevalent I have found them for as cheap as $3 a piece.  They are usually no more than $7 a pop.  My other tube amp(s) take 6V6's....also fairly easy to find affordable old tubes for these (the new Tungsol's I bought for them sounded less good than the NOS Sylvania's before them...but at least lasted the couple hundred hours I listened to them in there.  The EH rectifier fried tho...replaced successfully with NOS  )

John

[markC]

I've never really had the opportunity to give the new tubes the longevity test, as I find the sound of vintage tubes much more pleasing. Any tube gear that I have purchased only runs on new production tubes for a couple of weeks b4 I start with rolling in JANS and then more sought after N.O.S. The N.O.S or "gently used" oldies have always sounded better to me, be it in amp, cdp, dac. I am not speaking of power tubes.
Once a year, I pull all my tubes and test them. My pre uses 6N1P, my CDP uses 12ax7, (now a transport only so it is irrelevant although it was stand alone for a year and a half and the Brimars tested as new after a year), my DAC uses 6922, (not yet a year old, so also irrelevant), my amps use 12at7 and 12au7, (actually 5963).
The only tubes I've replaced  are the 6N1P's which are obviously not vintage. I did replace these with claimed '80's vintage rocket logo's, so I'll see next time I test.
I've yet to have a vintage tube die or even sound bad after several thousand hours of use.
Only problem that I can recall was a 12au7 mullard, (cv4003), that gave a slight "pop" after about 30 seconds after turn on.
The 5963's will be in use 2 years next month, so it's time to test again. Last year, when I tested them, they still tested as new.

[jon_010101]

I worry that your problem here is using 12AT7s instead of 12AU7s  ...

The differences go beyond just gain... they will seek a different operating point that may not be optimal.  And, since your amps have feedback (the Dukanes, right?), changing the gain of the amp will also change the feedback factor.  You'd need to redo a lot of the circuit to truly compensate for such a different kind of tube. 

Instead, I would suggest a different path - dumping the 12AU7s for 6CG7s, which is almost always a worthwhile pursuit.  You will need to change the socket wiring slightly, but the results should be good.  The 6CG7 is notably more linear than the 12AU7, and yet is a nearly direct replacement in practice.  They are also cheap in NOS or vintage form.  The catch is that the filament current draw is slightly higher - so if your power transformers run hot now, you might want to reconsider my advice.  For most vintage commercial grade amps, this should not be a problem.

[rpf]

I have not had a tremendous amount of experience with tubes. What I've had is as follows.

SED EL34s and 6550s: reliable and good sounding.

Sovtek 5AR4/GZ34: not reliable; nor nearly as good sounding as any of a half dozen NOS varieties (GE, Mullard, Amperex, and Japanese examples) and NOS relations (Philips and Mullard GZ32s, Mullard GZ37s, RCA 5V4) that so far (a year) have held up well.

Edicron EF86: decent sound, seemingly reliable.

Amperex EF86: very good sound, not in long enough to rate for reliability.

Valve Art KT66: very good sound, poorly constructed (loose bases, guide pins falling off) - not in long enough to rate for long term reliability.

Sovtek KT66 (metal base): very good sound, solidly constructed - not in long enough to rate for long term reliability.

NOS JAN Philips, Tungsol, and Raytheon 5687s: good sounding, last a reasonable amount of time but often go slightly microphonic - not enough to be an issue in my CDP but problematic in a pre-amp (where I formerly used them).

NOS RCA/Cetron 829B: great sounding power tube (48 watt output); not in long enough to rate for longevity but supposedly will last for a long time (I did have one tube blow but, of course, that could have been an exception). Very inexpensive at $10-20 apiece.

 

[DaveC113]

I haven't had tube gear for long, and already had a Sovtek and JJ rectifier go. The JJ fried a resistor and the fuse a couple times because it shorted as it was cooling down, releasing the ps caps to ground through the unfortunate resistor. I unplugged the preamp before the Sovtek did anymore than make some noises. So, I spent $80 on a NOS RCA/Mullard 5ar4 and $20 on a used big bottle Sylvania 5ar4... no more new rectifier tubes for me.

It does concern me a little, I built a SET amp that uses the JJ ecc99 driver tube, which is not compatible w/ NOS variants. It's driving re-issue Tung Sol 6550 power tubes, NOS are very pricey so I hope to not go there, and I hope the ecc99 proves reliable. It is very linear, especially with ccs loading, and I am very happy with these new production tubes so far.   

[JerryM]

Once a year, I pull all my tubes and test them...

Mark,
What method, and tester, do you use to test your tubes? Do you check the match? If so, to what degree? Do you change tubes or bias settings to adjust for any variance? Please elaborate.
Thanks,
Jerry

[TheChairGuy]

I worry that your problem here is using 12AT7s instead of 12AU7s  ...

The differences go beyond just gain... they will seek a different operating point that may not be optimal.  And, since your amps have feedback (the Dukanes, right?), changing the gain of the amp will also change the feedback factor.  You'd need to redo a lot of the circuit to truly compensate for such a different kind of tube. 

Instead, I would suggest a different path - dumping the 12AU7s for 6CG7s, which is almost always a worthwhile pursuit.  You will need to change the socket wiring slightly, but the results should be good.  The 6CG7 is notably more linear than the 12AU7, and yet is a nearly direct replacement in practice.  They are also cheap in NOS or vintage form.  The catch is that the filament current draw is slightly higher - so if your power transformers run hot now, you might want to reconsider my advice.  For most vintage commercial grade amps, this should not be a problem.

Hey Jon - yup, it's the Dukane's I have (good memory )

Being a non-techie sort, I always run the tubes that are indicated in the slots by the manufacturer unless told otherwise by someone far more informed than I.  A helpful tech told me to change out the 12AU7's for 12BH7's, a direct swap, and I did so last year.  The difference was huge - a fantastic upgrade it was.  Cheap too...the 12BH7's (I bought NOS, but there are new out there I understand) were inexpensive (like $5 each).

Another tech, the EE that is re-building/modding an Advent 300 quite for me now, suggested swapping in 12AT7's in the 12AU7 slots for higher gain (the Advent is likely to have a little less gain than my current preamp...disastrous for getting good phono volume out with majority of cartridges). 

That's where I'm having the issue with the JJ 12AT7's...they are neither higher gain (in fact, lower) and they hum (at least one of two, and possibly two of the four bought seem defective in some way).  Even tho, I bought low noise/microphony and matched sections, seems something slipped TubeDepot by with these.

Are you thinking that the issue may be incompatibility with my Dukies rather than outright issues with the tubes? Would that account for vastly lower output that I get with either the 12AU7 or 12BH7 in there versus the 12AT7? 

John

[markC]

Once a year, I pull all my tubes and test them...

Mark,
What method, and tester, do you use to test your tubes? Do you check the match? If so, to what degree? Do you change tubes or bias settings to adjust for any variance? Please elaborate.
Thanks,
Jerry

I use a Hickok 533A tester and test for any sign of shorts and then transconductance. I use all triodes so I check the readings of each side of each tube and if they are out more than 20%, (but still a good tube), I'll put it on the shelf and grab another closer match. I then check the 2 sides of the partnering tube and try to keep all numbers as close as  possible.
All my gear is auto bias for the tubes. Only our Hybrid amps have adjustable bias for the mosfets and while I've got them open I check the bias on the fets. They have yet to change from the original setting, (give or take a milivolt).

[jon_010101]

Are you thinking that the issue may be incompatibility with my Dukies rather than outright issues with the tubes? Would that account for vastly lower output that I get with either the 12AU7 or 12BH7 in there versus the 12AT7? 

I don't have a copy of the schematic, but the 12AT7 is fairly picky about operating points -- it's a somewhat nonlinear tube, but if the circuit is tweaked correctly they perform great.  If not, they might be barely usable.  I would stick to 12AU7 or 12BH7.  If you have used 12BH7 before, I think you can realistically try 6CG7s, since they both use more filament current than the 12AU7.  Maybe use one 6CG7 for the first stage and a 12BH7 for the driver stage.  6CG7 is a 9-pin 6SN7, which is the tube that the 12AU7 tried (and failed) to imitate. 

You'll need to change the filament connection, so it will take a tech a few minutes with a soldering iron to get it set up.

[TheChairGuy]

Jon,

Thanks - amazingly (amazing as these weren't popular hi-fi amps and they are quite old), Triode Electronics has the schematic of my amps bigger brother...the 100 watt 1A475 (I think mine is indicated as 50 watters)

http://www.triodeel.com/du1a475.gif

Don't kno wif that can even be read legibly to know what is going on with my amps.

I sent the JJ's back to TubeDepot yesterday for them to check over again...tho, based on what you are saying, I think there is dual problems at work: some bad tubes AND incompatibility with my amp(s).

Thx, John

[jon_010101]

Thanks for finding that -- Well, looking at the schematic, the design is a bit unconventional, with three cascaded single ended stages.  The plate loads are a bit larger than I expected, too (I wish I could read the voltage values!).  No red flags though ... I like that it uses a cathodyne phase splitter, rather than a paraphase.

A 12BH7 can work for the first stage, but it may not perform as well as a 12AU7 there (will want more current).  If it sounds good, then it's okay, but I would definitely consider the 6CG7 mod to be an improvement over either. 

For the second stage, the 6CG7 or 12BH7 can work, but the latter would again benefit from changing some resistor values around (let me know if you feel this motivated ).  As with the first stage, it is configured for fairly low current operation (~3mA), and the driver stage plate and cathode loads are big-ish - good for 12AU7 tube life, but not so good if you want a 6CG7 or 12BH7 to properly ass-kick the output tubes.   

The 12AT7 will have way too much gain, way too high distortion, and will probably mess badly with the feedback.  I think it would be best to use tubes with similar gain to the 12AU7 here, unless you want to totally re-do the circuit.   

The 12AT7 is a quirky tube, because it has fairly high gain and transconductance, and mediocre linearity on paper, but can perform well in a differential amplifier or with local feedback.  The Citation I preamp uses them properly with of feedback, and achieves awesome solid-state grade distortion figures and bandwidth.  But in a conventional circuit, you can just end up with a bunch of distortion. 

[TheChairGuy]

Jon,

Weird thing is that the 12AT7 sounded pretty good with CD/tuner (besides the all too obvious hum issue)...but the gain was too low for phono.  The 12AU7 is louder....and the 12BH7's are as 'loud' as the 12AU7's, but sound abso-friggin-lutely wonderful in there.

Guess I shouldn't mess with success...but I may have issues with the relatively low gain Advent 300 that is coming   

We'll see tho...thanks so much for your input on this 

John

[SET Man]

Hey!

    To NOS or not to NOS, that is a question...

    I just bought a pair of JJ ECC802S of which are a long plate version of the 12AU7 from Triode Electronic.

   I asked for a matched pair but when I measured them on my tester they are are not matched. One tube is 2075/2125 and the other is 1950/1825  I did request for matched pair but when I checked the bill look like they only charged me for tube only and not the extra $ for matching service. 

   Anyway, the measurement of these two tubes are a bit of a let down. My tester Hickok 6000 stated that "Average new" for 12AU7 is 2200 and minimum for these tube type is 1400 (according to the TV-7 tester) So, as you can see these "new" JJ tubes actually measured below average new tube. I know they are not actually 12AU7 but I don't think that is the problem here.

  I have a pair of Mullard 4003 '84 vintage that I've been using for a while and they still measure as new. And never had any problem with them.

  I also bought 4 used RCA "Cleartop" 12AU7 off eBay sometime ago. One pair is '59 and the other is '67 I think. They are all measured above average new. But sadly after about a year of use one of the '59 vintage develop a popping noise problem and get noisy. Well, these are old use tubes. I do have a another pair of a later vintage NOS JAN cleartop 12AU7 somewhere I've tried them for a couple of hours and they measured better and are quieter than the old used pair. So, maybe these will last longer.

  As for JJ I've also used 12AX7 JJ also in my phono-stage of my pre and two out of 4 developed noise problem after a few months. I took them out and replaced them with Ei 12AX7 Gold Elite. This was couple of years ago and the Ei set still work fine without any problem. I'm not sure how the new Ei tubes are today though.

   I haven't put the JJ ECC802S in my pre yet so I don't know how well they will sound. I know I have had problem with JJ before but I can't help try these ECC802S in my system. I guess I'm a bit intrigued by them with the long plate and all. 

  So, I guess I have to say that a good NOS is less problematic then new tube so far. Well, I just can't understand why can't current tube manufactures can't make tube that are as good as the NOS?

  Now the problem with NOS is that they can be expensive and hard to find. You know what if you tuned your system's sound with a pair of rare $$$$$ NOS. Than a couple of years later you need to replace them but you can't find them anywhere? Or they get way too expensive?

  Anyway, I guess this is one of the downside using tube. The price we tube users pay for the sound of which only tube can provide 

Take care,
Buddy

   

[jrebman]

I think that there are a lot of crappy new production tubes, and some really excellent ones, and from what I've noticed, the trend is that overall things are getting better with some of the new production stuff -- but certainly not all of it.

The stuff from tAD -- though mostly hand picked and screened chinese, soviet, and eastern block stuff that has been relabeled is getting a lot of attention -- mostly the stuff with the "STR" label, which is definitely not everything that TAD offers.

I've heard some people say the JJ EL84s are just awful for the first 100-200 hours but then "turn into Mullards", but as I never ran any of mine that long, I can't personally bouch for that.

I just got a pair of the TAD 5AR4/GZ34 STR rectifiers and while I don't yet have the amp they will be living in, some people who have had these in the same amp say that they like them better than their NOS mullards.  Fortunately I have some of those NOS mullards so I'll be able to compare.  This TAD 5ar4 btw, is a hand picked, thick glass version from shuguang.  And from what I've heard from some reliable and knowledgeable people, the shuguang factory is now under new management and the quality is, and will be going way up now, so I'll definitely be keeping an eye on them.  New Sensor is also making some fine reproductions -- the TS labelled 6550, 6SN7, and now supposedly the 12AX7 are all garnering a lot of respect.

As for Buddy's question -- why can't they make new tubes as good as they used to, well, the answer is that they can, and I believe they will, but only a few select companies and probably only a handful of the most popular types.  Once they figure out that the demand is there, or creat the demand with some speculative ventures (as it appears they are starting to do) I think we'll be seeing even more, and higher quality tubes out of the former soviet block and China.

I'm optimistic, and as much as I love my NOS tubes, the prices are really going insane, so I'm far less likely to buy any tube gear that requires tube types that are not readily available, or stated another way, gear that is voiced to sound good with the more readily available new production stuff -- of which there are more and more examples showing up all the time.

-- Jim