|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
68.6.131.58
In Reply to: RE: 12AU7 Reviled posted by airtime on July 03, 2021 at 06:51:09
For the ST70? EL34 / KT66 seem to have a large following.......it COMES with the EL34.
As for the ST120, where I'm 'heading'? comes wiht 6550, but lots of buzz going UP to KT88 with the Gold Lion getting some good mention. I think you can even go 'up' from there. I doubt you get more 'power' from the uprated tubes, but maybe more headroom while the sound profile of these changes makes the amp better for some......
The Driver board? Another can of worms, altogether. While the 'octal' base 6SN7 is nice, a LOT of leverage from simply swapping out the 'middle' of the 3 tubes from the 12AU7 to one of several alternatives. And while the 12AU7 is the red-headed step child of the tube world, it STILL gets lots of play. I think HOW is as important as WHAT, in this case.
Going to a SS rectifier AND time delay startup? Don't know that buys me anything TO START. some people love the GZ34 rectifier......
Amp should be purchased with all the extra / better capacitance they offer.
I'd say that while I'm NOT wealthy in the sense that I have a File CAbinet full of 'other' tubes to roll, I'd still consider trying something else....down the road some. And since my background with tubes is limited to running a tube checker when I was a kid...(Zenith.....the quality goes IN before the name goes ON!) I'm taking it at a gradient. After all, I have to TRUST that someone sellling a product will provide what is in thier judgement a good 'bang 4 buck' solution......
Too much is never enough
Follow Ups:
The 12AU7 sections comprising the long tailed pairs are probably not too critical in terms of linearity. Like cathodyne phase inverters and cathode followers, the LTP topology inherently applies a significant level of negative feedback. The first ("middle") 12AU7 is different though, and this is the one that should be replaced for best sonics.
One more issue here is driver gain. The original 7199 circuit has a gain of roughly 150. The 12AU7/6SN7 circuit comes in around 40-42. That's not sufficient to meet the somewhat standardized sensitivity rating of 1V RMS for full output. In fact, the ST-70 with this driver board and EL34s will require upwards of twice that voltage. Whether that's a problem depends on the other system components. The 9-pin driver board allows this to be easily changed, because there are many dual-triodes to pick from with various gains. The octal version offers fewer options.
--------------------------
Buy Chinese. Bury freedom.
Interesting! Could low gain cause a lackluster sound by not providing enough "push" going into the output stage?
Reduced gain could have a few negative consequences. The most significant might be increasing distortion from a preamp that's required to use a larger portion of its voltage envelope. Regarding the 12AU7 driver board, a 12AY7 or 12AV7 in the first position would improve the situation. Plate and cathode resistors would need to be changed, but it's not a difficult job.
--------------------------
Buy Chinese. Bury freedom.
It was changed, in part, because some were complaining of a hair-trigger, and the switch to 12AU7s mitigated that to a degree. I don't even know if they offer the "standard gain" (12AT7) option anymore, but it only involves changing a few resistors if anyone needed the extra gain, or felt the 12AT7 was a better tube choice.
Note that some of the plate resistors listed as 36K in the 12AU7 version above list are now shown as 30K in the schematic you posted, TK. I'm not sure if that's due to the addition of CCS, or for some other reason.
My voltages never matched up to what was shown on the several schematics (different voltages indicated on each) that I was sent. All resistors were checked on an ohm meter prior to installation, so I always wondered if the discrepancy was due to differences among variations in 12AU7s.
Maybe swap out the 12au7 for that?
I tried the 12BH7 in my VTA ST-70. It's been a number of year since the amp was in my system, but as I recall, the differences were subtle.
In a post linked below, there's a list of changes and / or build upgrades I tried. As noted there, the Russian 6P3S-E in the outputs made the most significant change for the better, but not enough to win me over, long term.
I've been told that the 12AT7 is a (slightly) more linear tube than the 12AU7, and I've thought about converting my amp to use them. I didn't, partly because I didn't need (or want) the extra gain, and I also lost interest, and moved on to other projects.
Do you have the necessary test equipment to measure output power? Do you know if the 6P3S-E tubes were able to produce full output in the Dynaco? I'm asking because I have output transformers rated 25-30W at about 3.8K plate to plate. The original tube types are NLA, so I need to substitute something sustainable. I have a lot of the 6P3S-E, but I question whether it's well suited to this load impedance. The 6L6 definitely is not, it works better at 6.6K or so. Given that the 6P3S-E has often been sold as a 5881 and that it appears to work well in 6.6K amps, I'm concerned this will be an issue.
--------------------------
Buy Chinese. Bury freedom.
...have the ability to test output power.I tried the 6P3S-E in my ST-70 based on frequently posted recommendations by Mike Samra to try them in the VTA, as well as in other amps. Along with the K40Y-9 PIO caps, Mike was a big cheerleader for some of the surplus Soviet stock. Perhaps he was a Putin pawn, eh? (Mike would see the humor in that, I'm sure)
I bought my 6P3S-E back when Jim McShane was still selling them, and as noted, they were the most pleasant for me to listen to in that circuit, when compared to the original Dynaco Mullard EL-34, and reissue Gold Lion KT-77s. Before I set off a firestorm from KT-77 lovers, the key words are "for me" and "in that circuit".
I talk to another VTA ST-70 owner, and if I remember correctly (given those particular email discussion were a couple of years ago) he liked actual 6L6 (or it could have been KT-66) better than the EL-34s in his amp too.
As I'm sure you recall from your Dynaco repair days, the original ST-70 manual said "In emergencies, ...the 5881 and KT-66 can be used in place of the EL-34's..."
I'm not suggesting that that was an optimal substitute, as your 3.8K vs 6.6K transformer difference indicates, only that Mike Samra and a few others of us thought it "sounded good".
Is the main result of mismatched tubes to transformers limited to reduced output power, or does it also increase distortion, or cause some other detrimental measurement? I consulted my 1964 Basic Electronics text book, but if the answer is in there, I missed it.
Edit: I found something for me to study regarding output transformer impedance (link below). The second sentence says, "If the wrong transformer is used, the results can be low output and loss of tone quality. ". The loss of power is obvious, but the "loss of tone quality" is more subjective, at least the way it was written. Perhaps the resulting loss of tone in the VTA ST-70 is what some of us found pleasing. Or am I drawing the wrong conclusion?
Edits: 07/09/21 07/09/21
It's interesting that you mentioned Mike Samra. I had a few discussions with him here regarding the "correct" loading for the 6P3S-E. From what I recall, he hadn't made the measurements necessary to confirm the extent to which its performance might be degraded when installed in these lower impedance output circuits.
Load impedance has a big effect on the 6L6. Here's a graph from the datasheet showing its performance over various loads. The operating condition is Class A single-ended, so the actual impedance isn't directly interchangeable with AB1. However, the curves show that tube performance peaks significantly at one particular load impedance.
There's also the AB1 chart published for this tube, which shows a major difference in output power for 3.8K VS 6.6K. Both are measured at the same maximum permissible grid-to-grid swing of 45V under these operating conditions.
My current project uses 6.6K transformers rated for 35W, and that's no problem for the 6P3S-E. However, the next amp will be a "re-creation" of the HK A500 with 3.8K outputs. I want to maintain its original output of 25W/ch, so this issue of load Z is a concern. Voltage on the anodes will be somewhat higher than the original circuit, but it's not clear to me whether that will be sufficient to reach the desired output power if the 6P3S-E is indeed analogous to the 6L6.
The output stage of my A500 is still operational (barely), so I guess this needs to be tested before the transformers are removed. The new chassis has to be built with all the necessary changes determined and planned in advance. From what I see of the market in terms of sustainable tube types, there's only one choice If the 6P3S-E isn't suitable - the EL34. Using this tube for 25W/ch seems a waste, but more important, the amp will need an additional, separate 6A filament transformer.
Three of the amplifiers I'm reworking have 3.8K OPTs, so I need to resolve this issue. It's really surprising that no one AFAIK has published any technical info for the 6P3S-E under these conditions. Hundreds of owners must have tried them in the ST-70 over the last few years. It would be very helpful to know how the 6P3S-E actually performs in those amps.
--------------------------
Buy Chinese. Bury freedom.
Mike often mentioned the 6P3S-E as a great tube, so perhaps not so unusual that his name came up. The only greater example would be a K40Y-9 PIO discussion. Mike was responsible for a LOT of guys buying and trying that coupling cap, including VTA, who used to supply them, but now seem to offer "4 hand matched Russian K42-Y PIO coupling caps" as an upgrade, I'm guessing because the supply of the K40 finally dwindled.
As for the power output, perhaps that's why I never noticed an issue, since I normally drive my speakers with a 10 WPC amp, so the loss of power while using the 6P3S-E wouldn't be as apparent to me as it might to some.
If I'm interpreting that THD graph correctly, it appears to me to be at its lowest at 3.5K Granted, that's at about 1.75 W, but, at least with high efficiency speakers, wouldn't a 3.8K transformer perform better than a 6.8K?
I'm guessing you're right, that hundreds of ST-70 owners likely did try the 6P3S-E. If they had the ability to measure the performance, it's too bad they didn't make the information available. Maybe someone reading this may offer something up.
"If I'm interpreting that THD graph correctly, it appears to me to be at its lowest at 3.5K"
Yes, but that's only one tube. It needs to be doubled (approximately) to obtain the primary plate-to-plate impedance of an equivalent push-pull transformer.
Maybe I should start a thread on this in the Tubes forum. Someone, somewhere must have done this and made measurements. I'll bet Bob Latino did... :)
--------------------------
Buy Chinese. Bury freedom.
"It needs to be doubled (approximately) to obtain the primary plate-to-plate impedance of an equivalent push-pull transformer."
Yes, for Class A
For Class B, AA/4 and what do we do for Class A/B?????
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
You're correct, the Class A charts don't indicate the exact load impedance that will produce the best performance in AB1. However, I do believe the relative coincidence between maximum output power and minimal distortion holds reasonably true across the various classes of linear operation. Based on that, I rely on the manufacturer's recommendations for the "typical condition" that produces the most efficient operation, i.e. highest output power. This approach hasn't disappointed me yet. :)
--------------------------
Buy Chinese. Bury freedom.
I would agree, and posting the query over on DIY might have a better chance of geting you a response.
Engraving the front panel now...
--------------------------
Buy Chinese. Bury freedom.
...just wow. Nice work. You take DIY to the extreme. Next you'll be glass blowing your own tubes!
You should definitely be posting this over on the Tubes DIY Asylum.
Now THAT makes sense. And I've seen a listing of the gains of the various interchangeable tubes for the driver board.
My pre is rated to 7 volts, but I'm not sure that is critical.......to the discussion.
A higher gain tube might get a couple good votes.......Better sonics. Better LIneariety.
And given that I'm leaning ST120, the EL34 is off the table. 6550 maybe off, too, but the KT88 has a lot going for it.....
Too much is never enough
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: