Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
In Reply to: RE: From what I understand the Zero's... posted by firstname.lastname@example.org on May 08, 2017 at 20:46:50
"Why do you have to put a transformer in an OTL??it's not OTL anymore!!"
As a general principle, I agree with you. To me, the term "OTL" has a very precise connotation, namely that there is no impedance-matching transformer between the tube output stage and the loudspeaker. This is a simple "yes" or "no" issue.
I find it to be a bit of a smokescreen when it is asserted that an autotransformer is allowable without invalidating the notion of OTL. As I pointed out before, if an impedance-matching autotransformer still allows one to call it OTL, then *any* tube amplifier could be easily made into an "OTL amplifier" by the mere contrivance of appropriately rewiring the output transformer so that the secondary was in series with the primary.
Of course, to use this argument would be rather absurd; I just presented it in order to demonstrate that one cannot take the autotransformer versus dual-winding transformer distinction as the determining factor in deciding whether an amplifier is configured as an OTL or not.
The only other line of argument would seem to be that if the primary-to-secondary turns ratio is "sufficiently small," then it can still be called OTL. Thus with two-to-one the term OTL would still be allowed, but for twenty-to-one it would not. This is really just a matter of sophistry, though. One would now have to declare some arbitrary ratio, lying somewhere between two-to-one and twenty-to-one, as the cutoff between OTL and not OTL. This would really make little sense.
To me, the simplest and cleanest definition is just the obvious one; OTL if there is no impedance-matching transformer between the tube output stage and the loudspeaker, and non-OTL if there is.
An entirely different issue is the question of whether a particular system sounds better with, or without, an impedance-matching transformer. But I don't believe that it is helpful for these more subtle questions to be blurred by what, to me, are unconvincing claims that a set-up is still operating in OTL mode even when there is an impedance-matching transformer between the output stage and the loudspeaker.
With the Zero's that the amp was still an OTL in the purist sense. I was just asking about the speaker impedance matching with OTL's and how critical it was after reading that article. I just made the comment that the Zero's were not a OPT like you were trying to say they were. So please keep the facts straight on that just like you are about the OTL design.
Some speakers give OTL amps a rough time due to their impedance or reactance and the result is bad sound. Then the OTL's get a bad rap for not sounding good or thin. Some of us do not have the money to burn to keeping buying amps or speakers until we get a perfect match, so the Zero's are maybe the better way to go to match OTL amps with our existing speaker systems IF there is an issue with the matching (yes, like a crutch).
As you know, just because a amp and speakers look good on paper....doesn't mean it will sound good in your system or home together. So the Zero's are the next best (cheapest) thing until you can either make or buy some speakers that match the OTL amp you have properly.
So instead of just arguing about the OTL purist design and the Zero's being an OPT (which they are not), why not bring some good advice or examples of what OTL/ Speakers you have or others that actually work well together so we can all learn.
Well, my own OTL amplifiers are home-built. Two of them use 6C33C tubes in the output stage, totem-pole style. (One pair of tubes in the output stage.) These give about 25W maximum into 8 ohms. One of the OTLs is based on a design by Hans Beijner, the other a design by Tim Mellow. I built another OTL that is a circlotron, based on a design by Alan Kimmel. This one allegedly gives up to 100W into 8 ohms, but I've never tried pushing it to measure that. Finally, I have one I really built just for fun, loosely based on the very early Dickie-Macovski design, with no iron-cored components whatsoever.
The speakers that I use are all high or fairly high efficiency. Lowther DX3 drivers in bass reflex cabinets; Klipsch Forte II; and Paradigm Monitor 7. With these, I've had no problems at all driving them from any of the OTLs. In fact the Lowther DX3 drivers have paralleled Audio Nirvana drivers too, so I suppose they are then nominally 4 ohm. But the sensitivity is so high that running out of headroom is never a problem.
I fully agree with you that if you have already an amplifier and speakers, and if the amplifier is not able to drive the speakers comfortably because of impedance and current-delivery issues, then Zero transformers may well be a reasonable solution to the problem.
Personally, I would be left feeling slightly unsatisfied if I were having to do that with an OTL amplifier, since it does, to some extent that different people can hold different views about, partially negate the whole OTL philosophy. For me, if I ran into such a situation it would be fairly straightforward for me to rebuild one of my OTL amplifiers in a beefed-up form, for example by doubling up on the output tubes. (In fact it would give me an excuse to build another one, which I have been wanting to do for a while, but currently find it hard to justify the need for a 5th one!) I can see, though, that if one is buying ready made amplifiers the financial considerations would be somewhat different.
Regarding whether Zero transformers should be called output transformers or not, I think that all that can be said on the subject has been said already. My main point in my observations on the topic is that one is going to be somewhat hard pressed to come up with a universal definition of what is, and what isn't, an output transformer if the definition is not simply going to be crafted for the express purpose of "allowing" Zero autotransformers while disallowing everything else.
Chris, Thank You for the info. Now this is what I was looking for in replies. :)
I totally agree with you about either building a new amp or speaker if there was an issue with proper matching OTL amp and speakers. I have built a tube amp from kit and have done a few DIY speakers but I am no expert in building tube amps from scratch let alone OTL amps.
What got me interested in OTL amp designs is how people who owned them said they were the best sounding amps the have ever had but speaker matching was a must to get good bass. I would really hate to add anything between the OTL amp and speaker that would take away from what the OTL amp brings in sound. So I do understand where you are coming from.
I hear those 6C33C tubes are pretty nice? Are these largely available?
I now have got some good info and direction/s on what to do which will help me out big time. I just wish I could build that many amps from scratch. There are some OTL kits I could try and hopefully with the DIY GPA driver (Altec) based 2 way with Zilch Z19 xovers should be a good match. If that doesn't work out I do have a Line Array design in the works with BG planars and Silver Flute 6.5 drivers I have already that I can arrange to get the proper matching impedance for the OTL amp/s.
I will look into some of those designs you mentioned
Thanks again for the info!
(1) The impedance conversion afforded by the Zero autoformer is much lower than that of any conventional tube output coupling transformer, because the impedance matching can be as low as 2:1 and almost never more than 4:1. (The output Z of the OTL is nearly the same as that of the speaker, usually at least a bit less.) Whereas in a conventional transformer coupled (OTC) tube amplifier, you are facilitating an interaction with power tubes that have output Z in the thousands of ohms to a typical speaker with a Z of 4-16 ohms. This makes for wider bandwidth and lower distortion for the Zero vs the conventional output transformer. The job of the output transformer is much harder.
(2) Because it's an auto former and because of the reduced workload, you would not have hysteresis distortion, or at least much less of it.
I don't say the Zero is perfect, but I do say that to my ears the combo of the Zero with an Atma OTL is still superior to every OTC tube amp I have tried on the same set of speakers. Audio is full of trade-offs. If you want to stamp your feet and walk away in protest, that is your privilege. Or find a pair of speakers that don't need the Zeros.
"I don't say the Zero is perfect, but I do say that to my ears the combo of the Zero with an Atma OTL is still superior to every OTC tube amp I have tried on the same set of speakers. Audio is full of trade-offs. If you want to stamp your feet and walk away in protest, that is your privilege. Or find a pair of speakers that don't need the Zeros."
I can absolutely believe that there are circumstances where a particular OTL amplifier may drive a particular set of speakers better if an impedance matching autotransformer is inserted between the amplifier an the speaker. And I can absolutely believe that the resulting system may sound spectacularly good.
My only gripe is if it is claimed that such a configuration is still adhering to the spirit, and the letter, of the concept of an OTL. I think OTL has a very clear and precise connotation: no impedance transformer between the output tubes and the loudspeaker.
As I said before, if you wanted to claim that with a turns ratio of 2-1 it is still an OTL system, but that with a turns ratio of 20-1 it isn't, then you would have to come up with some specific ratio somewhere between those two values, and assert that it is still OTL one side of the line, but not the other. This would be an entirely arbitrary cut-off, with no difference in qualitative behaviour between one side of the border and the other. Hence I simply go back to the (almost orignal) Dickie-Macovski notion, of the absence of any iron-cored components in the amplifier. (Actually they went one stage further and had no iron-cored components in the power supply either, but that is another story, and a side-issue for the present discussion.)
As with many things in audio, I think it is a shame when the image of a probably superb-sounding system risks being tarnished by virtue of claims that it is something that it really isn't. (And don't get me started on the subject of the ZOTL!!!)
Chris, You and a few others are the ones taking the purist stance. I have no desire to argue the point except to say what I said. Yes, the Zero is a kind of crutch to improve the match between an OTL and a typical low impedance speaker. That doesn't bother me. Most people who have employed zeros with their Atma and other OTL amplifiers report that they are happy with the pairing. I was too, until I found a way to modify my speakers to raise both their efficiency and impedance curve such that I no longer needed to use the Zeros. As you know, too, there are also ways to reduce the output Z of many OTLs, by adding NFB, that also might do the job and obviate the need for Zeros. It depends upon which is the lesser of two evils. It's all good, so far as I am concerned.
I would maintain though that a well designed autoformer, like the Zero, used at a 2:1 impedance ratio (I never went above 2:1; the Zeros permit up to 4:1, I think) is in fact likely to give wider bandwidth and lower distortions compared to a true audio output transformer as found on most OTC amplifiers, driving the same speaker, for reasons noted. If you want to say that an autoformer-coupled OTL no longer fits the strict definition of an OTL, I would not argue with that.
One other small point, the Zero is placed at the speaker end of the path from the amplifier outputs, not at the amplifier end, for whatever difference that makes. You might say it acts on the speaker. But I don't want to go there, either.
"Chris, You and a few others are the ones taking the purist stance."
Well, I am a theoretical physicist in my day job, always questioning, seeking to extract general principles, universalities, and precise definitions. I suppose I tend to apply a similar approach in the rest of life; automatically questioning, looking for inconsistencies or ambiguities, and so on. (Doesn't always go down too well, I have to admit!)
So in this case too I just like to have a clear, precise and robust definition of what OTL means. For me, it means one thing and one thing only; no inductive impedance-matching device between the output tubes and the loudspeaker. As far as I can see, any attempt to broaden this definition so as to allow some impedance-matching transformers but not others is just opening a can of worms, and it will end up with arbitrary criteria being introduced in order to try to justify something that really cannot be logically defended.
In any case I don't see any need for such a broadening of the concept of OTL; better just to stick to the simple yes/no criterion of whether there is any impedance-matching transformer between the output tubes and the speaker or not.
In a case where a Zero autotransformer is being used, it can perfectly well be described as an OTL amplifier supplemented with an impedance-matching transformer that connects to the loudspeaker.
I guess the thing that prompted me to comment in the first place was that for myself, I would feel it to be a bit of an admission of defeat if I had set out to build an OTL system and then ended up resorting to using impedance-matching autotransformers in order to do the job satisfactorily. Now, I'm sure it is true that certain speakers with especially low impedance might be quite challenging loads for some OTL amplifiers. If I had such speakers, and was determined to use them with an OTL amplifier, I think I would direct my attention to beefing up the OTL so that it could handle the load, rather than using transformers.
In the end this comes down to a matter of personal taste, I suppose. I built my OTL amplifiers because I was fascinated by the idea that tubes could drive ordinary loudspeakers directly. I wouldn't want to give up on that motivation, and if I had very low impedance speakers that provided a difficult load for the OTL amplifier, I would view that as challenge to overcome by improving the OTL amplifier itself.
I am an MD/molecular biologist-virologist. That mixed identity is confusing enough during my day job. I also have a view of audio that is analogous to something Frank Sinatra is reputed to have said about religion: "Whatever gets you through the night". He went on to say that in his case that was a bottle of Jack Daniel's.
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: