The worlds least efficient amplifier

[Jeb-D.]

Quite possibly that is.

Thought I'd share the schematic since I'm finally; pretty much done, tweaking with it. I'm sure some of the sand masters will laugh at my screen regulator, but it seems to be doing the job quite well.

Puts out 14-16Wrms (varies between output tubes). Uses 3dB of NFB. Its -3dB between 70-80kHz. It relies on harmonic distortion cancellation. So the distortion characteristic varies depending on how you set the input pentodes Vg2.

Overload characteristic is quite good, especially for a cathode biased amp. There are a few things going on which may be hard to notice at first glance. These "tricks" significantly reduce blocking distortion when the amp is over driven (even with the larger sized caps).

I can post some measurement data if there is any interest.

Best Regards and Happy Holidays. Booze it up responsibly

[kevinkr]

Pretty cool Jeb! Novel somewhat RH like feedback arrangement.. (CF inside the feedback loop)
Wondering about LED bias on the pentode, wonder how that would that affect your distortion canceling?

Here's the amplifier that I once nominated as the world's most inefficient amplifier.. Six 6C33-C-B per channel, and I never succeeded in getting more than 25Wrms per channel out of them. The power target was 120W into 4 ohms, and about 200W into 8 ohms.. They were quite effective voice coil fryers.. Sadly they never reached fruition because I this was about the time I started to get interested in SE amps, and ultimately after sitting in storage for years I gave them away - they were subsequently scrapped for the custom power transformers.. Here: http://www.kta-hifi.net/projects/amp...3amp/6c33.html

I suspect I would like your's a lot better..

Have a great Holiday!

[Miles Prower]

Originally Posted by Jeb-D.

Quite possibly that is.

No not really. For pure inefficiency, you can't beat a resistor-loaded parafeed. You'd be lucky to hit 10% with that.

I'm sure some of the sand masters will laugh at my screen regulator, but it seems to be doing the job quite well.

It isn't a regulator. It's an active decoupler (a.k.a. capacitance multiplier). Serves to isolate the load from any AC riding on the DC rail. Used the same idea with very high gain SS amps to prevent instability. It's also a good way to replace heavy and expensive ripple chokes. Done that too.

[Jeb-D.]

Originally Posted by kevinkr

Pretty cool Jeb! Novel somewhat RH like feedback arrangement.. (CF inside the feedback loop)
Wondering about LED bias on the pentode, wonder how that would that affect your distortion canceling?

Here's the amplifier that I once nominated as the world's most inefficient amplifier.. Six 6C33-C-B per channel, and I never succeeded in getting more than 25Wrms per channel out of them. Sadly they never reached fruition because I got interested in SE amps, and gave them away - they were subsequently scrapped for the custom power transformers.. Here: 6c33

I suspect I would like your's a lot better..

Have a great Holiday!

I went the same route; OTL to SET. My OTL failed as well. It was my own experimental topology. It did put out 20W with 1 pair of 6C33C's and it had good steady state/ sine wave performance. However it had some really strange transient behavior.

The LED would probably change the distortion characteristics quite a bit (as would bypassing the resistor). I had to try several different loads on the pentode before I got the desired behavior. The LED could probably be done with good results, but you would most likely have to re-tune the plate load on the pentode.

[Jeb-D.]

Originally Posted by Miles Prower

It isn't a regulator. It's an active decoupler (a.k.a. capacitance multiplier). Serves to isolate the load from any AC riding on the DC rail. Used the same idea with very high gain SS amps to prevent instability. It's also a good way to replace heavy and expensive ripple chokes. Done that too.

Technicalities The DC output impedance is also low enough to prevent Vg2 DC voltage from moving around.

[Fuling]

Re. low efficiency, my SE OTL probably beats almost anything.
It employs 4x PL504/channel in a choke loaded CF topology and consumes over 200W constantly to produce 2x 0,75W @ 8R. Not very eco in any sense of the word, sounds quite good though.

[kevinkr]

Originally Posted by Fuling

Re. low efficiency, my SE OTL probably beats almost anything.
It employs 4x PL504/channel in a choke loaded CF topology and consumes over 200W constantly to produce 2x 0,75W @ 8R. Not very eco in any sense of the word, sounds quite good though.

Just wondering if that 0.75W is a typo or is it really just 750mW? I guess that would have to win the award for world's most inefficient amplifier, or the world's most musical space heater..

Merry Christmas!

[Jeb-D.]

Originally Posted by Fuling

Re. low efficiency, my SE OTL probably beats almost anything.
It employs 4x PL504/channel in a choke loaded CF topology and consumes over 200W constantly to produce 2x 0,75W @ 8R. Not very eco in any sense of the word, sounds quite good though.

Wow! you win.
And I thought 16W out/ 200W consumption was bad.

[Fuling]

750mW, it´s not a typo

Merry x-mas!

[woody]

Steve Bench had an amp on his web page that may realy take this prize. I believe he
used 16 6as7's in inverted mode for ~ 16w single ended otl power and it only burned about 540w! per mono block! My memory isn't all that good I may be wrong about that power output it might have been more like 8w. I am sad to say it looks
like that web page no longer exist.