Another 6550 vs. KT88, 90, et. al.?

Kevziek, I appreciate your concern about reliability with the Rubys and can only say that there have been no problems so far (3 weeks). Actually, one of my JJ/Tesla KT88s shorted out recently,frying resistors in the amp so I was motivated to try something new.Eventually I will buy a replacement JJ to complete that pair and make another comparison.BTW the EH 88s were nice too but not on par with the Rubys.

Jkaway, sorry to hear about the JJ tube. Man, I'm getting tired of problems with tubes!

I also had JJ's (EL34's) go bad on me before I tried the Svet's, once taking a resistor. No such problems with Svet's or Sovteks, but the Svetlanas sounded best anyway.

Update: Russ did bring me (Literally brought them to me, 200 miles - Now how's that for service above and beyond the call of audiophile duty? [He had an 'errand' which took him by my neck of the woods] ) six each of his EI KT-90's and NOS Mullard KT-88's. Alas, the results of this noble effort and experiment are going to remain entirely inconclusive for now.

The KT-88's, whether because of their advanced age or something else (such as possibly differing somewhat from the current-day spec on the Svetlana KT-88's, which VTL assures me are a drop-in replacement that they've auditioned but rejected in favor of the Svet 6550C's I've been using), would not come close to biasing in correct range. They did this as a group, their individual biases all coming up in the same range as each other, so I doubt that any one of these tubes was 'bad', but whatever the cause, I couldn't do any auditioning with them, and I removed them from the amp after five minutes warm-up failed to budge them any closer to the desired 275-300mv from the 50-80mv they were registering with the bias trim-pots brought up gradually to maxed.

The KT-90 experience was even more aborted. The tubes I'm sure are fine, but new information has rendered this attempt a non-starter. When I first acquired these amps, I spoke to Luke Manley about a few things, one of which was tube selection, and in discussing the KT-90 option, he only addressed the sonic and reliability/availability differences VTL has found between these tubes and their chosen 6550C's. He discouraged me from going the KT-90 route, to me implying that they *could* be used, but what he apparently failed to mention at that time was that retrofitting KT-90's would entail resistor changes in the amp in order to work correctly. Oops. Having now found this out, and knowing in advance that this substitution wouldn't work out (meaning that at the very least, these tubes wouldn't bias correctly either) - and not wanting to risk damage to tubes not my own (or to my amps) - I haven't even tried putting them in the amp.

So, the one easy option available to me at this point would be trying a new set of Svet KT-88's, but I've got no clues as of yet (either experientially, or from VTL owners on this thread or in the archives) whether this would be worth me making the investment when I've still got a perfectly good set of 6550C's here. FWIW, Bea Lam at VTL says the KT-88's were passed over by them due for the most part to their softer bass response, although I think she liked their sound from the mids on up, but VTL thought the 6550C's were best overall.

There does remain the distinct possibility that the tonal balance I hear in tetrode with the 6550C's installed is actually largely accurate, and simply a reflection of my speakers' possible inherent tendencies. As usual, everything affects everything else, and unless I get a set of new KT-88's or eventually some different speakers (not an unlikely event in the course of time), my choice between running tetrode or triode for now will be based mostly on the demands and tonal balance of whatever recording I'm playing at the moment, with tetrode still getting the nod probably 90% of the time, despite the slight brightness/leaness.

That is, unless my next experiment renders this a moot point: I'm going to upgrade the amps' coupling capacitors soon, and from what I've been told, I could quite possibly expect some improvements in the very areas of tonal balance I've outlined here, along with even greater transparancy and purity, so maybe it's for the best that I don't worry about tube subs anyway until this mod is completed.

Many thanks again to Russ for doing everything he could to help me in this decision, and when he returns from his vacation soon, he and I will have to figure out how I'm going to get his tubes back to him...NYC road trip? :-)

Update: Bought and installed Electro-Harmonix KT-88's, after two of twelve Svetlana 6550C's failed at around 1 year and another had become noisy. I hadn't been thinking of going the Sovtek/EH route, until Kevin Deal at Upscale Audio recommended them as being the most durable new KT-88 while sounding just as good as the competition - that they're less expensive than Svet/SEC's to boot was only icing on the cake.

(Disclosure: I didn't wind up buying from Kevin this time, as his rather costly testing regimen hadn't seemed to correlate with longevity on the 6550C's. Supposedly, the internal Sovtek testing and matching regimen is much improved, so I bought from my usual source for new tubes at the best prices, Triode Electronics, but note in fairness that they don't claim to burn-in and pretest them the way Kevin does. So far, so good, but it's early yet.)

In my old amp that used EL-34's I hadn't prefered the sound of Sovteks vs. the Svet's, but they *were* durable (so were the Svet's, but not the JJ's). However, I am told that the EH EL-34 is indeed a slightly different tube from the Sovtek-labelled version, which is cheaper. Who really knows for sure, but I decided to give the brand another try, whichever name they put on this particular tube. The so-called KT-88EH's, while cheaper than SEC KT-88's, are more expensive than SEC 6550C's (and also of course than the so-called 6550EH's).

This comparison cannot be completely fair, for although my surviving 6550C's should still be in good shape at 1 year (and do bias well), obviously the new KT-88's should be expected to be a bit stronger. The fact that I'm switching brands as well as types also introduces another uncontrolled variable to the mix. One thing I can say for sure is that IMO, the fat-bottled KT-88's look more impressive just sitting there installed in the VTL's (though I suppose Freud might have something to say about that :-) Interestingly, Australian audio designer Patrick Turner, on his manufacturer's website, features a page devoted to his test results on the EH 6550, wherein he states that EH KT-88's tested at the same time displayed identical measured properties.

On to the sound: It was my sense that most users who had done the 6550/KT-88 comparison felt the KT-88 was a 'warmer-sounding' tube, but such was not my finding. Maybe these impressions were primarily formed in response to earlier 6550 versions, I don't know. In fact, Kevin Deal had said these two tubes wouldn't sound much different at all, and as it regards my amps anyway, he is mostly though not totally correct.

To reiterate what I was hoping for, I felt that the 6550C's might have been a touch bright through the presence range, and also a little lean from the lower mids down to the upper bass. But since first beginning this thread, I've come to the conclusion that perhaps my choice of input tubes was mostly to blame for this result. At the time I started this topic, my 12AT7 inputs were 60's Sylvania Gold Brands (also bought from Upscale), but they were more recently discovered to have become quite microphonic.

Bringing in a solid-state McCormack DNA-125 as a spare amp had highlighted for me how the sound of my reference VTL mono's had subtly deteriorated in ways that I had only really internalized subconsciously, and these flaws were found to be directly related to the input tubes. Those tubes were replaced, shortly before the output tube swap, with 60's Mullard ECC81's, that basically fixed some problems with tonal balance and imaging control, restoring to the entire presentation (even with the old 6550C's) a certain calmness and fullness that had slowly gone missing since the Sylvania's began losing it.

I've burned-in the KT-88's for about three days so far, and though they might still be changing, I feel I can report some results now. My first impression upon hearing the new tubes was that they had noticably more extended bass. Performing one-speaker-mono comparisons, where I'm switching one speaker cable between the two monoblocks, each fitted with a set of the competing tube types, has confirmed this finding head-to-head.

The bass region generally is where I've heard the biggest improvements to date. I am being careful to remember that my 6550C's have a year's use on them as I say this, but I never recall hearing bass out of them like I'm hearing now, even when new. The bass is not just more extended: it's also more evenly weighted, with a welcome equality of emphasis given to walking bass fiddle lines. This is both a matter of the improved deep bass extension and a concurrent reduction in mid-bass warmth that was probably spurious. The result is a bass presentation that is distinctly weightier, but no more preponderant overall. In addition, bass notes are more redolant of tonality with the KT-88's installed, while the 6550C's come across as having a bit of undifferentiated 'flat aspect' to their LF cast. All of these observed differences are fairly minor in degree taken by themselves, but together are important to better conveying the musical message.

Beyond the bass, the differences are primarily not ones of tonal or harmonic presentation. What I notice most is that the KT-88's give a slightly more distant spatial perspective, which I happen to prefer to the 6550C's more up-front presentation, and that they also tip the balance between direct and reverberent sound to more emphasize the qualities and dimensions of the recorded soundspace. In addition, individual images are better kept free of one another, with more space and less smearing between them. The only other qualities worth commenting on either way for each tube would be that the KT-88's seem a touch faster, while the 6550C's might be a smidge more extended on top (but I'm reluctant to draw final conclusions about that last observation, since I find that HF airiness is often the last aspect of a tube's sound to fully develop during the break-in process).

As of now, I am coming down in favor of the EH KT-88's (in my VTL 185's, through my Thiel CS2.2's, and for the time being with Audience Au24 serving as the reference speaker cabling) over the SEC 6550C's. Their (in my judgement) superior bass fidelity, coupled with a more separated and laid-back spatial presentation, better press my personal sonic hot buttons.

But I have not found this tube to sound any warmer than the 6550C's - to the contrary, if anything they are a touch cooler, a little more analytical and removed-sounding. You could even say that the KT-88EH's sound a little more like solid-state, the SEC 6550C's a little 'tubier', but again any such differences are very minor everywhere above the mid-bass. Anyway, my experiments to date with input tubes convince me that this is the tube position where one should focus on the tonal and harmonic balances, and I may yet try something a touch warmer here, because the Mullards sound quite hi-fi but not warm themselves.

Besides the generally advantageous sonic qualities I've described, I'm also hoping that the KT-88EH's will prove more durable compared to my so-so experience with the 6550C's, but that only time will tell.

P.S. - My thanks again to Rcprince for making the effort in trying to assist me with auditioning some of his KT-88's before having to plunk down for 'em myself - despite the unforseen complications, I think it's all worked out in the end Russ! (I'll be sending you a tracking number tomorrow :-)