OTL Amps: The Work Of The Devil?

[Mr Kipling]

When the Audio Research SP8 came out Martin Colloms reviewed it for HFN&RR and also interviewed William Z. Johnson. During the interview MC asked him for his thoughts on output transformer-less designs. WZJ's reply was that it really wasn't a suitable way of operating valves. I think he likened it to a car's gearbox and that it was integeral to the design of an amp.

I've never heard an OTL and just wondered what others thought. I've seen a simple design using the 6C33 which I like the look of, but I' still quite happy with my inexpensive Audioromy FU29 amp.

[Thermionic Idler]

Depends which devil designed them really. The output transformer, unless it's a very good (for that, read expensive) one, can be a significant limiting factor in the performance of a valve amplifier. Replace it with a straight piece of wire and what you theoretically end up with is a very cost-effective way of getting reference-grade performance.

Or at least it should be cost effective, the prices I see being charged for a lot of OTL's would actually buy a pretty decent pair of output transformers, so it kinda begs the question, why would you bother. You're very limited in the range of output valves you can use, they need to be able to pass a decent amount of current, so that's the EL509, 6C33C, and a handful of others. Oh and you also need lots of negative feedback which is anathema to a lot of audiophiles (I used to avoid amps with feedback like the plague). However, IMO there's really nothing wrong with using feedback, but it has to be done properly if it is not to rob the sound of its musicality.

OTL's have a reputation for unreliability, massive heat and electricity consumption, and blowing up speakers. Transcendent amplifiers do none of these things. I built a pair of the T16 designs as monoblocks in 2004 and they are still working perfectly eight years later, and the more recent stereo "Son of Beast" OTL I use now has never given me any trouble aside from a slightly noisy mains transformer. They just work and if you're good with a soldering iron I think they're a massive bargain. You can see from the signature below the capabilities of the partnering equipment it's used with.

Yes they do chuck out a little more heat than a normal amplifier. The S.O.B. dissipates about 168 watts at idle --- it's a push pull design and one of the lowest waste heat OTLs on the market. The single ended and more powerful designs put out a little more. Some designs from other manufacturers put out as much as a space heater!

The circuits are also fuse protected to shield your speakers against valve fault conditions such as grid to cathode shorts.

OTL's project a cavernous soundstage like nothing else I've heard, and I find the latest S.O.B. design very difficult to fault. My previous T16's could be a little bit shrill at times, better quality source components mitigated that to a degree. One critical factor really is the speakers, present an OTL with a difficult load and it will sound crap, no question.

[DSJR]

Well, bearing in mind how awful (coloured processors rather than transparent line-buffers/amps) so many vintage ARC amps sounded and how unreliable the old ones from this company are/were here in the UK (with telephone number bills to put them right/update/upgrade as well), William Z Johnson is the LAST person I'd respect for a balanced reply regarding anything to do with audio!!!!!

OTL's are difficult and apparently not much good at current delivery (I think I've got this right way round) and some have a terrible reputation for unreliability, but get them right with properly matching speakers and they can sound so much clearer than many transformer-coupled valve amps. GOOD transformers are VERY expensive these days and even then, bass distortion and damping factor can be a serious issue with some amp/speaker combinations (Prima Luna and Harbeth come to mind :)).

Could Anthony TD chip in here please???????????

[jandl100]

I had a pair of Chinese OTL monoblocks .....

http://i262.photobucket.com/albums/ii114/jandl100/OTLvalveamp.jpg

They sounded really good! - clear, dynamic, fast. Super-dooper!
I kept them about a year, and they worked well with all of the speakers that passed thru Jerry Towers in that time. They even managed to inject a bit of life into some Spendor SP1/2 !! :eyebrows:
Really nice into Quad 57 stats, too.
Hmm :scratch: - OK, they did struggle a bit with Thiel CS2.3, a notoriously challenging load, but managed at medium to low levels if not too much was required at the bass end.

But then, for most of my listening, I'm not really a headbanger type of listener.

Sadly, a week after I sold them, one of them blew up. :doh: :(
Still, they only cost me £600 or so - less than £2 a day. Not too bad!

[anthonyTD]

HI All,
Valves [compared to transistors] are high impedance, high voltage, low current devices, they were not designed to be partnered directly with low impedance speaker drivers, therefore, to get them to do so is unnatural, and why you have to [generally speaking] use them in their multiple 's [and use lots of feedback mainly in an attempt to gain some realistic damping factor] to have even a chance at getting them to drive a nominal 8 ohm impedance!
On the other hand i agree about output transformers, their design can be the making or the downfall of even the best circuit designs.
The other main issue that most people contemplating using OTL's must be aware of is the safety aspect of their speakers, it is of the utmost importance that any OTL amp use the very best [fastest] decoupling circuit in their output stage, this is the only chance you have of saving your precious speaker drive units when one or more of the output valves decide they have had enough and go short circuit!
I don't disagree that when done right these circuits can perform very well, and can in many cases see off a lot of traditionally designed valve amps using mediocre or bad output transformers.
There are many more comparisons to be made between the different types of OTL's and traditional valve circuit designs, therefore, there will be more comments made on this subject, of that i am sure. :)
Hope this helps.
Anthony,TD...

[Zoidburg]

When the Audio Research SP8 came out Martin Colloms reviewed it for HFN&RR and also interviewed William Z. Johnson. During the interview MC asked him for his thoughts on output transformer-less designs. WZJ's reply was that it really wasn't a suitable way of operating valves. I think he likened it to a car's gearbox and that it was integeral to the design of an amp.

I've never heard an OTL and just wondered what others thought. I've seen a simple design using the 6C33 which I like the look of, but I' still quite happy with my inexpensive Audioromy FU29 amp.

I cant claim to know much about valve amps but its nice to hear some one else with an Audioromy FU29. Ive been using mine for quite a while now and im very pleased with it. :)

[Marco]

Valves [compared to transistors] are high impedance, high voltage, low current devices, they were not designed to be partnered directly with low impedance speaker drivers, therefore, to get them to do so is unnatural, and why you have to [generally speaking] use them in their multiple 's [and use lots of feedback mainly in an attempt to gain some realistic damping factor] to have even a chance at getting them to drive a nominal 8 ohm impedance!


Indeed, Anthony. You may be interested in reading my comments here on the subject, from post #12 onwards:

http://theartofsound.net/forum/showthread.php?t=22810&page=2

I'm sure that some OTL amps are fab, but I think it's important to address the hugely misguided notion that output transformers-less valve amps are automatically superior to ANY tranformer-coupled designs. They most certainly aren't. One must consider other important factors, notably whether a transformer-coupled design may be of overall higher quality than its OTL counterpart, and also be sure that, transformers (or lack thereof) aside, one is comparing like with like! :exactly:

As an aside, you may be interested to know that Nick (lurcher) was round the other day at Ian Walker's, letting him hear his new preamp (which apparently was superb), where he had the opportunity to listen again to the copper amp in Ian's system, now that it's performing optimally.

To say he was impressed was an understatement, even though those pesky sonic abominations, known as transformers, were in the way and 'ruining' the sound!!! ;)

Marco.

[DSJR]

"Our" Glenn went over to Mos-FET#s for very good reason - and IMO the results speak for themselves, since the little Series 7 even in basic form is one heck of a good amp and gives nowt away to full tranny-coupled valve amps - unless your looking at several thousand quid's worth of TD or EAR and even then, the EAR is double the price it should be in the UK now sadly.....

[Reffc]

I've only heard a Berning OTL and haven't really liked what it did. Big sound stage but coarse sounding. Croft MOSFET design seems a more reliable, safer and more economical solution not to mention better sounding! However, nothing has yet beaten the sound of a valve amp using top class transformers to my ears. My current Mastersound is amongst the best I've heard for the sort of music I like but good transformers come at a high cost.

[Ammonite Audio]

There's an interesting little article on Enjoy The Music this month by Andy Grove and Peter Qvortrup which touches on OTL designs. See http://www.enjoythemusic.com/magazine/manufacture/0105/

[Marco]

I've only heard a Berning OTL and haven't really liked what it did. Big sound stage but coarse sounding.


Ditto my experience of them, so far, Paul: rough as a badger's boz, and with no finesse.


However, nothing has yet beaten the sound of a valve amp using top class transformers to my ears. ...but good transformers come at a high cost.

+1. I'd be all for using an OTL design - *providing* that it was safe and comprehensively outperformed my Copper amp, by showing exactly what its 'nasty 'ol transformers' were doing wrong! ;)

Marco.

[Mr Kipling]

Thanks for the replies.

I always remember Martin C. being quite taken with the SP8 and the interview with WZJ stuck in my mind. I think the man himself died last year.

Thanks for the photo Jerry. What was the output valve in the amps?

Thanks for the link . What was said about OTL designs is pretty much what William Johnson said to Martin C.

Searching the internet I haven't found anything negative regarding the Audioromy FU29 amp. One buyer did have his arrive with an output valve smashed with them being fitted. Another with friends who had McIntosh and some high-end German amps both bought an FU29 after hearing it. For £199 (which included carriage at £46) I really can't complain.

[jandl100]

Thanks for the photo Jerry. What was the output valve in the amps?

They used the 6AS7 (6N5P) output valve, quite a common one in OTL amps, I understand.

When I first bought them from eBay, they hummed badly. The seller sent them to a repair guy to get them sorted.
Quite an interesting story! -- http://www.chambonino.com/work/hifi/hifi10.html

[Mr Kipling]

Thanks Jerry. Quite an interesting read that.

[Ali Tait]

6as7 is equivalent to a 6080, a commonly used valve in otl amps.

[montesquieu]

Tried out a Croft III OTL with the Tannoys. Lightweight beastie using quite a few 509s. Had been recently serviced and revalved (otherwise I wouldn't have let the thing near them).

Sounded rough, with lots of detail missing. I had been interested in OTLs because of all that wonderful theory about directness blah blah blah.

On the plus side it did sound big and bold and cheerful, and reasonably musical. But not for me.

The Tannoys are a relatively easy load and should have been an ideal partner. Perhaps unfairly I haven't bothered with OTLs since.