|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.28.239.50
I note that Bambi B, who obviously knows a lot about tubes, and others don't like this term. So...anyone know what it means and where it came from?
Follow Ups:
I first came across the term "tube rolling" in the early 90s in the mfgr comments section of S'pile...the good ole days when it was a journal sized mag and high end audio was fun. There may be earlier references but I'm not aware of them.David Manley and Roger Modjeski were going at it publicly over the relative merits of each of their amp design philosophies. One of Roger's claimed design advantages was the capability of his amps being able to accept a wide range of output tubes with minor bias and output transformer tap adjustments. David claimed that his designs were highly optimized for a certain output tube type and really no other and therefore the sound of his designs was better than that of Roger's. David derogatorily referred to the willy-nilly switching around of output tubes as tube rolling . Current usage seems to refer to all forms of tube switching at all positions in an amp and includes both switching within a given tube type and between tube types.
My interpretation was that this is sixties slang, e.g. "I scored a couple of pinched waists on ebay" or presumably "I rolled a few tubes and it was a blast". I think we've all discovered that sound engineers are fond of the creative use of plant life.
"David claimed that his designs were highly optimized for a certain output tube type and really no other and therefore the sound of his designs was better than that of Roger's."Seems a strange claim for Manley to make. As I understand it, he was notorious for misusing tubes by exceeding maximum ratings.
As you may recall, one of Manley's philosophies at the time was to run output tubes at near max rated voltage but at relatively low current so pd was usually well within ratings. Some tubes would take this OK such as US mfg 6L6GCs @ 500V or 6CA7s (beam tetrode version) @600V+. Others would not such as those Russian/German/England 6L6 like things or RFT EL34/6CA7s which were all over the place in the early 90s. Even late 60s/early 70s Mullard EL34s had problems much above 450/475V even though rated by the book at 800V...yeah right. My take is that Manley didn't misuse tubes as such but I don't think he accomodated the low quality of 90s tube production. IOW his designs were based on NOS and he didn't clearly reveal this. He did allude to it by recommending that replacement tubes come from him so I suspect he was fully aware of the issue. Weren't those heady times?
Manley wasn't/isn't the only boutique designer to opt for the high voltage low current route to high power. As some are discussing below, even RM had an amp that somehow squeezed something like 30W from a pair of EL84s, presumably using HV and near class B design.
Or is his topology the way to go?I'd LOVE to upgrade my power trannies in the LEAK St20's from a life / TBF perspective alone.
AND, I think I'd need to just to get sufficient Bplus! to get the RM wattages.
I feel this may be the more sensible mod/learning path, as against going with a CCS'd / pure A OPT stage.
WarmestTimbo in Oz
The Skyptical Mensurer and Audio ScroungerAnd gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
For starters, I don't know RM's topology or the operting points he used. However, if his designs allowed many different tubes to be used, it suggests that he didn't go for extremes of either voltage or current - commendable 'middle of the road' stuff.Not wishing to offend, I must say that I don't think the Leak Stereo 20 would be a good starting point for experimenting, unless you are prepared to do a lot of redesign, rebuilding and parts replacement. Even getting it to perform well at its rated power requires that.
The St. 20 has a power transformer that tends to run hot, poor PS smoothing with no choke, and rather 'ordinary' OP transformers. I don't think the rather cost-conscious Harold put any more into it than he had to, to get 10W per channel out. (As you probably gather, I'm not much of a Leak fan).
Midband minimum in standard form was 16 watts and around 12 @ 20hz.I agree with you about Harold, though. Some Leakies think he was a saint er sumfin'.
I know the mains tranny heat issues, as I live in inland SE Aussie where summer temps exceed 40 degrees C! It WOULD be replaced if I went with either option; CCSing the output stage, OR higher AB1 watts! are there any proprietary items?
But you MIGHT be wrong about the St20's OPT's, they weigh as much as the OPT's in the St60 which I owned for a while.
I did NOT like the 60 at all with 5881's, it was OK with big envelope 6CA7's, but still cloudy.
Didn't think much of it cf both of Radford's PP-EL34 models (on ltloan!), and after I finally got the 60 back from Allen Wright (yes!), I swapped it for two St20's! Good deal eh?
I was very norty but, I wanted to go with NoS PLUS a choke but it was just TOO expensive here in Aussie. I went with a nice quiet bridge of quiet/soft-r SS!!!!!! diodes, snubbed as well! - and, 3 or 4 BIGGG bypassed caps*.
Slam? yep!!!!
I've also owned a pr of the TL12.1, IMO a far better application of two KT66's than the QUAD II, similar power but far clearer and nicer too. Even into the ELS! * I have owned two complete sets of the QUAD amps, but never used either pre!
Of the LEAK range, only the TL12.1and the later TL12+, and St20 do it for me. IMO the TL12+ CAN have somewhat better bass when done up.
Do you know of any proprietary over-rated mains trannies, with a bias secondary?
I do have CCS's FETs on the driver tubes wher I used 6cg7's local NoS AWA/AWV ones. They sounded heaps better than the 12AX7's even before the CCS! Should I add a trim pot?
Mind you the sound with CCS'd drivers IS startlingly good, quite close to SEUL 13E1's at least! 'Glow' wise.
Would adding a second input twin-triode tube be a good idea, plus a single CCS each, and an AC balance trim pot?
Is a trim pot essential for a CCS'd opt stage, too?
Cut the OPTs NFB, too!? Not needed?
WarmestTimbo in Oz
The Skyptical Mensurer and Audio ScroungerAnd gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Sorry, Timbo, I've only just seen your response - I don't visit this forum every day. I'll try to add comments where I can."I know the mains tranny heat issues, as I live in inland SE Aussie where summer temps exceed 40 degrees C! It WOULD be replaced if I went with either option; CCSing the output stage, OR higher AB1 watts! are there any proprietary items?"
CCSing the OP stage (I assume you mean a single CCS in the tail of each pair of OP tubes) would force the amp into Class A and might sound very good but would limit the power somewhat. I suspect you might need to reduce B+ to 250v to get the optimum operating point.
To get higher AB1 power, you'd need to change to fixed bias, approx -15v.
(I don't understand what you mean by 'propietary items').
"But you MIGHT be wrong about the St20's OPT's, they weigh as much as the OPT's in the St60 which I owned for a while."
I meant the qualities of the OPTs are nothing special, not that they are too small.
"I was very norty but, I wanted to go with NoS PLUS a choke but it was just TOO expensive here in Aussie. I went with a nice quiet bridge of quiet/soft-r SS!!!!!! diodes, snubbed as well! - and, 3 or 4 BIGGG bypassed caps*.
Slam? yep!!!!"
That'd be a useful improvement, IMHO, especially if you want to go to fixed bias and more AB1 power, which will need a stiff power supply. A smoothing choke would certainly help here, and a Hammond would do the job. Just make sure that the change to SS rectification doesn't cause B+ to increase to a destructive level.
"Of the LEAK range, only the TL12.1and the later TL12+, and St20 do it for me. IMO the TL12+ CAN have somewhat better bass when done up."
Leak models are reputed to get worse as they get bigger. I don't have experience of the 50 and 60 w amps but I'm inclined to believe it.
"Do you know of any proprietary over-rated mains trannies, with a bias secondary?"
I use an Amplimo 7N697 toroidal power tranny, with 230v primary and secondaries of 340v @600mA, 6.3v @6.8A and 50v@100mA. I can certainly recommend it but it might be difficult to fit it on a Leak St 20 chassis. With 340v to play with, you could use a choke input filter to get around 300v with good regulation. An E-I model would be a better fit, though, and you might look at what Hammond has to offer.
"I do have CCS's FETs on the driver tubes wher I used 6cg7's local NoS AWA/AWV ones. They sounded heaps better than the 12AX7's even before the CCS! Should I add a trim pot?"
I'm not at all surprised they sound better that ECC83. With a CCS in the tail, you need to use equal, well-matched plate load resistors, and there is no need for a trim pot. The AC gain will be very well balanced and if you need to balance out the DC operating points you could try adding a resistor of, say, 470 ohms in each cathode above the CCS.
"Would adding a second input twin-triode tube be a good idea, plus a single CCS each, and an AC balance trim pot?"
I can't really answer that one but I would guess you probably don't need to add any more gain, despite the lower gain of your 6CG7s. Leak made the St 20 ridiculously sensitive, so less gain is probably a good thing.
"Is a trim pot essential for a CCS'd opt stage, too?"
No, but the OP tubes need to be well-matched.
"Cut the OPTs NFB, too!? Not needed?"
Leak used heavy NFB - it was 'fashionable' 50 years ago. Nowadays it's considered a sin by the 'High End' fanatics, but WTF do they know? Anyway, considering the St 20 is an ultralinear design, it doesn't need as much NFB as Leak gave it. Don't forget, though, that you have effectively already reduced the NFB quite a lot by changing to the lower gain 6CG7 tubes for the drivers, so reducing it any further might not be helpful.
If you have the test gear, you might like to check that the phase compensation in the plate circuit of the IP stage and across the feedback resistor are still doing their jobs properly, now that you have changed the driver tubes. Their values might need to be adjusted.
Cheers,
Good point and we did set them up to give 'more than enough class A' for most listening.the triode amp is almost all pure class A! being only 6 or so watts.
WarmestTimbo in Oz
The Skyptical Mensurer and Audio ScroungerAnd gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
Well, if you need power, you need power. But CCS plate loads for the input and driver stages and a Real Class A operating point and load for the output tubes gives less distortion. I like the idea of less distortion from electronics.Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
AFAIK PP CCS'd EL84M's/7189 IE A1, are gonna struggle to dissipate 20 watts in pure A - continuously - as at idle, even in pentode mode*.And, so long as I have a watt or three of A in AB1 I'll be happy.
Especially if fixed bias, were used, so I could adjust over the power tubes life.
And I might go with a CCS on each input/gain stage, even TWO single triode input tubes (6C45?), with an extra socket!
With fixed (adjustable) bias for each PP tube I think we'd be getting pdclose to SE sound, anyway!
MY spkrs are 91db/w, and an EASY 8 ohm load, I'd just like more reliability, AND a little more headroom!
I would need to upgrade the power trannie anyway - for the 'CCS'd pure diff'l opt stage' route, to be worth doing - as I don't want to LOSE power, either!
*I don't even KNOW what continuous output the OPTxers could cope with.
WarmestTimbo in Oz
The Skyptical Mensurer and Audio ScroungerAnd gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
[AFAIK PP CCS'd EL84M's/7189 IE A1, are gonna struggle to dissipate 20 watts in pure A - continuously - as at idle, even in pentode mode*.]The 7189 is rated at 12 watts anode and another 2 watts for the G2. So a total of 28 watts for a pair of tubes. A CCS in the common cathode circuit set for 20ma would therefore not be a problem if the B+ isn't too high.
[And, so long as I have a watt or three of A in AB1 I'll be happy.]I doubt you will feel that way once you go to real class A amplification. Especially when in pentode a tube is very critical as to the best load. When in class A the tube sees half the plate to plate load but when it moves to class B it suddenly sees only a 1/4 of the plate to plate load. Not good for a pentode. THD will be many times higher and you can dang sure hear the difference.
[Especially if fixed bias, were used, so I could adjust over the power tubes life.]
If anything fixed bias allows wider current variations through the tubes
[With fixed (adjustable) bias for each PP tube I think we'd be getting pdclose to SE sound, anyway!]I don't understand why you thing a pentode class AB1 stage is going to sound like a class A1 SE stage stage simply because of fixed bias.
[MY spkrs are 91db/w, and an EASY 8 ohm load, I'd just like more reliability, AND a little more headroom!]
Yes, you do have to take the speakers into account.
[I would need to upgrade the power trannie anyway - for the 'CCS'd pure diff'l opt stage' route, to be worth doing - as I don't want to LOSE power, either!]
No, there is simply no way to have the same power in class A as you can in class AB. There is no need to change the PT and nothing would be gained. If anything class A would usually be run at lower voltage and higher current. A small first cap followed by a choke can make this happen and then the PT could deliver more current.
[*I don't even KNOW what continuous output the OPTxers could cope with.]
In class A the output power will be lower and the OPT's will have less to cope with. The CCS on the output tubes (if done right) will force perfect AC balance something no other method can do. You could still use a little fixed bias to adjust for equal standing bias.
And tah!
WarmestTimbo in Oz
The Skyptical Mensurer and Audio ScroungerAnd gladly would he learn and gladly teach - Chaucer. ;-)!
'Still not saluting.'
This post is made possible by the generous support of people like you and our sponsors: