Tube OTL and Zotl amps

[Mr. Moscode]

Starting this thread for Tube OTL and Zotl amps. I asked for A and D power amp forums. Until we get one this will have to do.

 

[Lampie519]

Nice ! There are many OTL amps out there but not many Zotl amps. This need to change as not only the Zotl technology determines the quality of the amps but also the circuit itself like in any amp. So by just shouting to have acomplished to have built a certain topology it doesn’t mean it is any good until you have auditioned it.
Not all OTL amps are alike just because they are OTL. I have heard very dull sounding OTL amps and some really great ones. Same for Zotl i guess and for direct drive ESL amps.
Plus you only get out what you put in, so your source is most important. But we all know that ( i hope).

 

[Mr. Moscode]

Did you build the 6 tube (looks like 6LF6s) amp? If not who makes it?
You are right about every amp sounds different. Futterman uses a current starved 6EJ7 input tube. F rolls off like an opamp, so NFB is limited in the HF territory.

 

[Lampie519]

The amps were build by a friend of mine. I designed the electronic circuit and the pcb for these. PL519 tubes are used here as these were available. A cascode input stage and PC900 inverter tube (triode with shield between grid and anode to reduce capacitance). These amps were built more then 20 years ago and now needed 1 new tube as a coupling capacitor failed. The tubes are still strong as they have been replaced 2 years ago already.

Later i switched to full range ESL's and built new HV OTL amps for these as
well...

 

[Mr. Moscode]

Very cool. Kudos to you.
Did you bootstrap the phase inverter?
Have ever used Julius's circuit or front end?
What country are you in? I see 220v outlets.
Thanks,
George

 

[Lampie519]

Yes, the phase inverter is bootstrapt like in the original Futterman amps.
I live in the Netherlands therefore we have no 6LF6 tubes but PL519 (hence the name Lampie519) or 509 tubes instead. I tried the original front end but somehow i didn’t like it (30 years ago). I never looked back to the original as the cascode worked fine for me, also i changed te ESL’s so the 8 Ohm OTL was of no use for me anyway.

The next step is HV Zotl ( i have experimented with that as well but because the existing amps are doing so well i post pone this project as so many other projects need to be finished first.

Are you the one who developed the NCP 1 preamp?

 

[Mr. Moscode]

I did the NCP II (2) which was adding switching to the NCP 1. Jon Syder did the NCP 1. I then proceeded to redesign it as a mod. Putting the volume control on top of a cap to ground to keep a cap out of the signal path was causing the OTLs to blow fuses by creating a feedback loop through the power line. 100 db gain at 10Hz with the volume all the way down. That had to go.

OTL front end:I changed the plate R to 900k to open up the top end. Put in darlington followers to drive the output tubes, allowed me to change the 82k output grid r to 2 megs which lowered the pole by a factor of 20. Also allowed me to take NFB directly off the output cap instead of the separate nfb 800mfd 'lytic. More stable, tight bass, cleaner top end. Depth of image much deeper. Take out the 6LU8s and use the sockets to go 8 tube triode.

There you go.

 

[Lampie519]

Cool !

I still have all the copies of the "Tube Bible" that shows all the changes (we had no internet at the time).

I used a clone of the NCP II for many years ( late '80's early nineties ) until a CD player needed to be added. Then i changed to passive for some time but that had not enough drive. Later i tried Allen Wright's preamp ideas but these needed to be changed to my taste (too harsh on violins and compressed). I am allergic to compressed in an audio set.

I found an old picture of my place (1987 LOL)

 

[Mr. Moscode]

"Later i tried Allen Wright's preamp ideas but these needed to be changed to my taste (too harsh on violins and compressed). I am allergic to compressed in an audio set."
Me,too. Too much gain with too much NFB will do that to you. Sucks the wind right out of you.

 

[Mr. Moscode]

Tube Bible - That's the RCA tube manual. I have Harvey's understanding Tube Electronics. I went to work there in 1982, I think. I invented & developed the Moscode hybrid amp.
G

 

[Lampie519]

I never had the chance to audition a Moscode amp but i was very interested at the time as it combined tubes with Mosfets. I still want to try something like that.

 

[thor.zmt]

Hi,

For "OTL Tube Amp" I would probably just do a Futterman style Amp using a sensible output tube that can drive 600 Ohm (maybe a few 6N30 in Parallel) and add P-Channel Mosfets in a compound feedback pair to drive low impedance loads.

Yes, it's not strictly "OTL Tube" BUT is will give really high power and sound very much like a tube OTL Amp.

ZOTL? Why not just make a Class D Amp with a Tube in the first stage?


Thor

 

[Lampie519]

ZOTL? Why not just make a Class D Amp with a Tube in the first stage?

Zotl is not Class D (only if you want it to be). Please read the patent.
https://davidberning.com/technology/patent5612646
The HV ac power supply is polarized but not filtered (using diode bridges). In this case the tubes do get almost 100% pos voltage (with a small gap because of the transition. This can be overcome by using 2 square waves overlapping each other (Quadrature). In this case the tubes get 100% DC on their anode if well executed.

With Class D amps you will need a filter on the output of the amp, with Quadrature Zotl not. This because of the push pull nature of it will nullify the transitions on the output node (what is left over from it as there shouldn't be any in the first place) .

 

[thor.zmt]

Zotl is not Class D (only if you want it to be). Please read the patent.
https://davidberning.com/technology/patent5612646

I am familiar with it.

A switching DC-DC Converter (Boost type) creates a High Voltage (originally with a P-P converter, but SE is possible) and the modulated supply current for this switching converter becomes the audio output.

In effect a Class D Amplifier is a switching DC-DC converter with the output modulated by a signal current.

In both cases the signal current may be created by tubes or transistors.

Can you explain to how on basic level a switching DC-DC convertor with the output to the speaker modulated by a signal current is different from a switching DC-DC convertor with the output to the speaker modulated by a signal current?

The HV ac power supply is polarized but not filtered (using diode bridges). In this case the tubes do get almost 100% pos voltage (with a small gap because of the transition. This can be overcome by using 2 square waves overlapping each other (Quadrature). In this case the tubes get 100% DC on their anode if well executed.

With Class D amps you will need a filter on the output of the amp, with Quadrature Zotl not.

Incorrect, you usually still need a filter to pass EMC Regulation with anything ZOTL and Class D can operate filterless, even with EMC Regulation compliance.

This because of the push pull nature of it will nullify the transitions on the output node (what is left over from it as there shouldn't be any in the first place) .

You can do the same in class D. Just saying. The excessively hyped Class D amplifiers are NOT the main or only way to do class D.

And no, on the output there should and must be RF, that is the nature of switching converters.

Realistically, both ZOTL and Class D are switching DC-DC converters with a modulation signal applied somewhere and a (usually) lowpass filtered output driving a speaker driver.

Placing a tube frontend with a Class D amplifier can be used to give the same In-Out transfer curve as a full tube Amp with a "ZOTL" Circuit replacing PSU and output transformer.

If so, how are the two different terms of operation and results. Again, just saying.

Thor

 

[Lampie519]

It is all very interesting but i have build several Zotl amps and can see what is actually happening ( not just saying but doing).

Best regards,
Frank

 

[Mr. Moscode]

I can tell you from my years of development and refining the Moscode which is entirely analogue that using an output filter coil will F-k up the sound. All versions are rock stable with no filter because I use only 10db of negative feedback.

 

[Lampie519]

I use adjustable feedback so you can set your sweetspot depending on loudspeaker used. So even no feedback is possible if you like ( some purists insist of having none).

Funny how most think they understand how things suppposed to work but never had hands-on experience.
I am not talking about this topic here just in general .

Like telling someone how to drive a car without having driven themselves ever.

Best regards,
Frank

 

[Mr. Moscode]

Hi,

For "OTL Tube Amp" I would probably just do a Futterman style Amp using a sensible output tube that can drive 600 Ohm (maybe a few 6N30 in Parallel) and add P-Channel Mosfets in a compound feedback pair to drive low impedance loads.

Yes, it's not strictly "OTL Tube" BUT is will give really high power and sound very much like a tube OTL Amp.

ZOTL? Why not just make a Class D Amp with a Tube in the first stage?

View attachment 107244
Thor

Because it's not as good. There are some of those around. It works but it won't sound like a Futterman OTL especially in the top 3 octaves. I have yet to hear an amp that sounds as good as a FOTL. Not even my Moscode tweaked to the max.

 

[Mr. Moscode]

I use adjustable feedback so you can set your sweetspot depending on loudspeaker used. So even no feedback is possible if you like ( some purists insist of having none).

Funny how most think they understand how things suppposed to work but never had hands-on experience.
I am not talking about this topic here just in general .

Like telling someone how to drive a car without having driven themselves ever.

Best regards,
Frank

That's true for every forum I've been on. Motorcycles, cars, audio, etc. But at least here there are some pros despite the fact there is a real distinction between the pro audio and the Hi End consumer audio audience. Peavy, Fender and every other guitar amp is not high end audio by any means. 95% of Pro audio just has to work good enough, it doesn't have to let you see God. There are audiophile record labels that do care but it isn't a big market. Maybe it's time for a new thread about this.

Frank, are you willing to share your Zotl schematic? I downloaded the patent and I get it, a bridge circuit with the "DC transformer" and with tube loading. The transformer is really a keen invention for reflecting resistance down to DC. Does it go to 0 ohms? I give it up to David, who I've met many times at the HIFI wars. It's a genuine genius invention. That it accurately works and sounds as good as the reviews say is amazing. I've not heard one and the $75k amp only does 60 watts. That's a $kilobuck/watt and the bass isn't great. But someone will buy it.

My quads definitely like 200+ wpc.

If you have a LRC network on your Zotl amp and use less than 10db NFB can you get away remove/ bypass it? A snubber would probably have to stay.

 

[thor.zmt]

Because it's not as good. There are some of those around. It works but it won't sound like a Futterman OTL especially in the top 3 octaves.

I have not come across this, outside the the work of JC Morrison. His Silbatone Amp's using this, last time I heard them on antique WE speakers at Munich in 2019BC (before COVID) sounded fine.

I would be interested in other examples on existence.

Thor