|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
4.244.186.252
In Reply to: Any benefit to lower than maximum watt output for tubes? posted by glassear on February 26, 2007 at 11:14:19:
Duhh, we covered this last year. Run em 5/8ths of Maufacturer's spec and the sound will be FAR less strained sounding, and FAR FAR FAR more relaxed and easy sounding.I experimented with this in the past, and two weekends ago See my recent report below, as the very last two paragraphs in particular apply directly to your question, and what we heard.
Jeff Medwin.
Follow Ups:
I have found that I like em run at high current and lower than max voltage. I have not found any ratio that I like, just by experimental tuning with fixed bias arrangements.
Just my $.02 :)
Bob
boo Ivan303.
Jeff, you have reported on changes to power supplies that, in your opinion, sounded better than other ways of doing things. OKYou now say that a tube running at 62.5% will sound better than one running closer to max. OK
I am learning to not trust your ears.
300B's in Class A push pull, at an output level of 15 watts will have .0486% 3rd harmonic distortion and .0033% 5th harmonic distortion when the idle current is at 62.5% of max plate dissipation.
When the idle current is turned up to 90% of max plate dissipation the 3rd harmonic distortion will be .0188% and the 5th harmonic distortion will be .0008%.
The 3rd harmonic distortion is now 38% of what is was and the 5th harmonic distortion is now 24% of what is was.
You must like the sound of distortion.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
I do not think anyone can hear any differende between 0.02% and 0.05 % distorion, it is lower than -66dB....
Academic however because at 15 watts any speaker available will have orders of magnitude greater 3rd and 5th. It's worth noting too many triode-connected pentodes respond much differently to lower static current than a true triode. The third harmonic often nulls right around the 60-70% dissipation mark.
Here are the numbers for a PP pair of 300b in Class A at 1watt..0027% 3rd .0005% 5th at 62.5% of max plate dissipation.
.0009% 3rd .0005% 5th at 90% of max plate dissipation.Here are the numbers for a pair of 6ca7 triode connected, PP Class A at 1watt.
.3932% 3rd .0072% 5th at 62.5%
.2428% 3rd .0047% 5th at 90%I don't use triode connected pentodes.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
The bane of writing is things gets left unsaid for the sake of time. I in no way agree in a Universal Constant of 62.5%, it usually takes hunting an pecking to find the 3rd null. A 6CW5/EL86 for example will drop (from memory) about 20 db under some operating conditions. Anyone who has and trusts SEAmpCad can confirm the behaviour quickly. I suspect, though haven't tried, that an 6CA7 will demonstrate a significant 3rd harmonic null at some standing current below 90% dissipation.Could you elaborate on the source of those 300b figures? That's mighty impressive performance. Were those the dominant harmonics? (Not assured for every PP design.)
Those are just numbers from the push pull tube cad. 300Bs loaded with about 6.6K. Cathode biased with 430VDC at the plates. And yes, that is assuming complete balance so that there is no even ordered harmonic distortions. Doesn't happen in real life.My only point was to show that running low current in output tubes is usually not a good thing.
Please understand when Jeff says "Run em 5/8ths of Maufacturer's spec" he means 5/8 of manufacturer's suggested operating point. Not 5/8 of max.
In this post
http://www.audioasylum.com/forums/tubediy/messages/117656.htmlJeff says "Of Maufacturer's $ugge$ted operating point$, and NOT max points at all."
That's just crazy. A 300b with only 37.5ma of current. There will be very little power and very high distortion.
I don't think Jeff really understands audio electronics. He just talks like he does. And speaks as if in decrees from above.
"you must have low DCR in your power supply or your system is Lo-Fi""All output tubes sound better if they are run at 5/8 of suggested operating point"
"The value of a grid stop resistor is critical to .01 ohm"
The thing is, people seem to listen to him. I'll never understand why.
Tre'
P.S. Those numbers for the 300b are impressive. But take some time to look at the plate curves for the 300B. It's a very linear tube. The 2a3 is better and the 45 even better. The 45 is near perfect. Loaded right you will get almost no distortion from a 45. It's the lowest distortion amplifying device ever made. I wish I has some really good horn speakers. I don't so I use 300b's to get some power going.
Hey man,YOU have NOT listened to ANY of the points I have been called on to the matt - as being no good - in your above post.
I HAVE LISTENED to ALL of the above, UNlike you!! I know that what I contribute up here comes from listening experience and it sounds superior, even though you do not comprehend it academically - and you never will. I don't care about academics, if the stuff doesn't perform really well according to ear !
How about adding to the above list "reaching critical inductance at 120 HZ" ! Ha, what a sonic crock, now disproven.
And who CARES about 300B distortion figures?
What a way to whack off, when a 45 or 2A3 will sonically creme a 300B?? The bottom line ( and summary ) on 300Bs remains - that 300B users are ONLY for those people "who don't get it" with speakers, and use not-high-enough-efficiency speakers. People just like you.
Thanks again for the extended reply. Many DHTs have beautiful curves but I don't recall ever seeing ideal PP distortion figures for the 300. The discussions usually centres on SE. It's a real eye opener. Regarding recommendations of the absolute and specific type you mentioned, let's just say they don't hold much attraction.
Thx again!
You're very welcome. If you haven't yet, check out Lynn Olson's website.I think he has some very good things to say. He's a bright guy.
You need to dig for it, but the paper entitled "Amplifier Design Philosophy" is a good overview of what I think makes for a good amplifier.
It's at the very end of the "The Amity, Aurora, and Karna" section.
http://www.nutshellhifi.com/triode4.html
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Dear Tre,I am learning to trust Jeff's ears.
As controversial as his ideas (or from Dennis) sometimes might be on first sight, closer examination every time showed me a way towards the goals I want to achieve. For me, this has been true for the low DCR argument and the layout argument, so I’m open for the lower op-points argument. Further, as Jeff posted, not only his ears came to the conclusion that the lower op-points gave better subjective results, others ears concurred.
You might think about 'distortion' as an absolute figure or percentage where smaller is better, but given the fact that 'everything' in the audio chain distorts (source, amp, speakers, room, your ears, probably your brain too), it might help to look at 'distortion' as part of the total equation. Further, lower op-points, might help in other areas in such a way that 'on balance' you're better of with running your tubes less hard than the current ‘best-practice’.
As mentioned in a recent thread, the increase or deliberate creation of certain distortions (harmonics) in certain stages can help in the cancellation of distortions (harmonics) in other stages, or in the creation of a (more) ideal harmonic structure. So instead of starting an endless fight to avoid distortions (which you will never win), one might start to think about managing (controlling) distortions in such a way that at the end one gets closer to his goals.
There is quite an analogy about this and discussions about the cancellation of resonances or speaker cone break-up. Since resonances and break-up are unavoidable (like distortion) the ‘more intelligent’ approach proved to be the management (control) of resonances and speaker cone break-up. So the whole subject is rather more 'sophisticated' than your statement about distortion percentages in a tube output stage.
So yes, I find the insights of Jeff very useful. Again, his subjective experiences (verified with others) gave me the ‘inspiration’ not to take things for granted. And again, his subjective experiences say nothing about the ‘science’ behind it, nor does Jeff make any claims in that direction. IMHO the subjective experience is the first step towards science, but it is the experience itself that is important. (For me that is, since I’m more interested in the art than the science YMMV).
I just had to say that.
Jeff has been at this hobby for a long time and has a wealth of experience. While I have made some harsh posts I never doubted his ability and I want to be sure Jeff and others know that.I was just mad at the way the info was presented and dribbled out....almost like a tease.
With some of the undefendable things Jeff has said I don't understand how you can say you ".... never doubted his ability".All I can do is throw up my hands.
Take care......Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Tre, we have exchanged posts for a year or two. We see eye to eye on many items, like the need to compare with live unamplified music, etc. So let's say we continue on for twenty years and still see eye to eye. And then one day you "discover" something that seems out in left field. Maybe you didn't personally discover it or even understand how it works but you use it and know it sounds better. At that point should I discard the past twenty years?Jeff has been at this for longer than that. He has a reputation and in particular he is considered to be a better than average listener and judge of sound. Nothing he has said nor done invalidates his past work and if he claims it sounds better I don't doubt him.
Still I am not happy with how the information was presented. But he is under no obligation to anyone here to tell or teach them anything. So consider the info a gift and do with it what you want. We have grown accustomed to thinking folks here are required to defend their statements. That we have grown accustomed does not make it so. Jeff need not explain anything, or he may explain it in anyway that he cares to. That is his right. Jeff has been honest about his inability to explain things unlike some that offer explanations that are clearly incorrect.
"he is under no obligation to anyone here "So it is your opinion that anyone can say whatever they want and there is no need to prove it in any way?
I think Jeff does have an obligation to explain the statements he has made. You may disagree with that.
Peace, Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
nt
Hi Jeff,With 5/8 of specs do you mean than 5/8 of max. voltage, max. current or max. anode dissipation?
Regards,
NC
Hi NC,Of Maufacturer's $ugge$ted operating point$, and NOT max points at all.
JM
Jeff, I'm still not understanding. Would that be 5/8 of suggested plate current at suggested plate voltage or 5/8 of suggested plate current at 5/8 suggested plate voltage?Assuming 5/8 suggested plate current at suggested plate voltage.
You would run a 300b (I know you don't like 300b's but if you were going to run one) at 37.5ma at 350 volts. Right?What would the proper plate load impedance be for a 300b running at that operating point?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
WE 300B max ratings times 5/8 (.625)...400V X .625 = 250V
100ma X .625 = 62.5ma
36W X .625 = 22.5W250V X 62.5ma = 15.625W
So which is it?
__________________________________________________
Boo!
I don't think Jeff knows the answer to that question. I think Jeff is just making this stuff up as he goes.Jeff really said 5/8 of suggested not max, so that would be
350V X .625 = 218.75V
60ma. X .625 = 37.5ma218.75V X 37.5ma = 8.2 watts But that's not an operating point for the 300b.
250V X 62.5ma = 15.625W is not an operation point for the 300b.
350v X 37.5ma = 13.125W is also not an operating point for the 300b.Either way, the load impedance would have to be very high to keep things somewhat linear that it would be impractical.
I really don't understand why anyone pays any attention to Jeff at all. He clearly does not know what he is talking about.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
is 300V @ 62ma. That looks fine to me and is likely close to what WE actually used in amps they designed back in the day.That would equal a plate dissipation of 22.2W which is very close to 5/8 max (35W X .625 =21.875W).
About a 5 watt amp after transformer losses and a WE 300B would last for a long time at that dissipation rate.
__________________________________________________
Boo!
Yes but Jeff said 5/8 of suggested. Not 5/8 of max. 5/8 of max is the WE suggested operating point (well, very close). But that's not what Jeff is saying. Jeff said "Of Maufacturer's $ugge$ted operating point$, and NOT max points at all"So Jeff must mean
350v X 37.5ma = 13.125W 5/8 of suggested current
or
250V X 62.5ma = 15.625W 5/8 of suggested voltage
or
218.75V X 37.5ma = 8.2 watts 5/8 of suggested voltage and current
not
300V @ 62ma. = 22.3W as this is 5/8 of max. voltage and current
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
This post is made possible by the generous support of people like you and our sponsors: