• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

New build: 6F12P - 6N8S - 6P36S triode push pull.

It's nice to see the formula broken down and calculated. I knew 1/2piFC but I just use ElectroDroid because I don't "enjoy" math per se and it's quicker for me. As you see, I'm also too lazy to use the coding to write that formula as you did it, and I say it "one over two pie eff sea" out loud. Then again, I've always been more of a mechanic than an engineer. :D

My knowledge has been acquired from old books like "Basic Electronics" from 1955 that do a great job of giving you formula but also simplifying and illustrating what is being taught (https://mirror.thelifeofkenneth.com...sic Electronics, Volumes 1-5, (1955)_text.pdf) and experimentation. "What happens if we do x? This amp build is an example of this.

Not many people use both choke and capacitor inputs for two different voltages at least that I've seen. Not many people would hook up a 6 ohm (average) speaker load to the 16 ohm tap of a transformer to drop the Z from 4300R to 1600R to get better use of the peak current capability of a sweep tube. It was an experiment of trying to repurpose parts on hand really. So far I like the results, but IMHO I'd rather have a lower voltage transformer and ditch the choke altogether. The nice thing about this experiment is that I can simply replace the magnoval sockets with octals, rewire the power supply, reconfigure the OPT wiring, and use KT88 instead. I know it would make for a wonderful 60W (limited by OPT) amp.


ElectroDroid Pro – Apps on Google Play

I love how you know so many formulas and calculations and are able to "run the numbers" so easily. It's really quite impressive IMHO.

Koda
 
Last edited:
No, I haven't. Grid 2 is tied to plate. A quick SIM in PPcalc says the difference is about 10W of output power (30W vs 39W), higher 4rd harmonics but lower 5th harmonics...

I might try it, as the output transformer is rated at 318mA and the circuit as built will peak almost 600mA.
 
So now that I've finally had a chance to beat the crap out of it, I'm impressed. I mean, I could make a similar amp with Triad toroids for far less money, but they would be a tight fit on a 10x17 chassis unless I stacked them.

Now I get to wait for Pavel to repair my bias board for me. Then I will install it and have a "set it and forget" system that makes tube wear and changes seamless...
 
Member
Joined 2004
Paid Member
Agreed. I've been running them at 340V/90mA for months. This build is scrapping vacuum rectifiers for SS diodes, so the voltage will be higher.

Very impressive for a tube with a spec of about 16-18 watts dissipation for plate and screen combined. I have some 6P36S lying around; dug one out to take a look and visually compared it to a 6P3S-E. The plate structure on the 6P36S looks bigger than the 6P3S-E plates. I have used 6P3S-E for many years with 450 V / 40 mA (UL-PP, 6k6, 35 W) without any problems, so the datasheet plate dissipation on the 6P36S seems very conservative. Does anyone have similar long term experience with 6P36S and their power and endurance limits?
 
Last edited:
Member
Joined 2004
Paid Member
So now that I've finally had a chance to beat the crap out of it, I'm impressed. I mean, I could make a similar amp with Triad toroids for far less money, but they would be a tight fit on a 10x17 chassis unless I stacked them.....

Interesting project that I might want to build. I have a few questions please.

1. Could you comment on how this amp sounds. I have read many of your threads on various amps that must have been good-sounding too. How does this one compare. How much power output?

2. Would you recommend the Triad toroids in stead of the Hammond 1650N based on performance or cost? Which of the VPT-18 transformers would be the right one?

3. Why did you include the driver behind the phase splitter. I thought the 6GB5 does not need a driver in pentode, UL mode, so I assume you needed it for driving the 6P36S in triode.

Thanks
 
It's nice to see the formula broken down and calculated. I knew ½πFC but I just use ElectroDroid because I don't "enjoy" math per se and it's quicker for me. As you see, I'm also too lazy to use the coding to write that formula as you did it, and I say it "one over two pie eff sea" out loud. Then again, I've always been more of a mechanic than an engineer. :D

My knowledge has been acquired from old books like "Basic Electronics" from 1955 that do a great job of giving you formula but also simplifying and illustrating what is being taught (https://mirror.thelifeofkenneth.com...sic Electronics, Volumes 1-5, (1955)_text.pdf) and experimentation. "What happens if we do x? This amp build is an example of this.

I love how you know so many formulas and calculations and are able to "run the numbers" so easily. It's really quite impressive IMHO.

Koda

Somewhere in my apparently misspent youth, I too had a bunch of 1950s and even some early 1960s high-end electronics-for-engineers books which I liberated from the nickel-a-book bin at the local library yearly sale. With deep irony, not having enough nickels … to buy all the books I found interesting was a real sobering prospect. As it was, I had to make 8 trips on my little red bike to the library and home; I emptied both my and my unknowing brother's ceramic piggy banks of all their coin. But books I bought, and read, and read, and read, and read, until their covers fell off.

No one (pessimistically) told me “little boys can't read college-level electronics engineering books, lad. Go outside and hit a ball or scare a cat or fall out of a tree or something!”

By the time High School came around, and I learned algebra, “the lights went on” so to say.

All those darn books with their hundreds of formulæ, which I “remembered, dâhmn it” by rote, suddenly took on newer, harder, more satisfying meaning. I re-read the books again … this time able to understand ALL the equations, and with my trusty slide-rule, I could calc out the examples and come up with the results that were otherwise so innocently spoon-fed by the authors to the reader. Heaven!

Somewhere along the path of life, I lost all those books, but never their contents. That is why I always dig back to “the basics” in formulæ. I really don't remember the frankly-easier-but-more-numerous 'cheat sheet' formulæ. Just the basics. ALL of the basics. Forever etched in my melon at age 10, from the nickel-a-book sale of the public library.

I come to DIYaudio emphatically not to “show off”, but to impart some small bit of this precious “muscle-memory” electronics mathematronics to those who might be interested. Surprisingly many of us are, yet depressingly fewer each decade. I think the great “new world” will be dominated by SPICE sim-math, and with almost soul-searching irony, I can say … its really a good thing. We don't really need to remember 2πFL and ½πFC I guess. Not when a time in a few CPU seconds can compute response curves and plot them from 1 Hz to 1 MHz … for arbitrarily complex circuits having surprisingly well-modeled mathematical components. It is the future.

Lastly, you're welcome, Koda. I'm glad to know you.

GoatGuy ✓
 
Interesting project that I might want to build. I have a few questions please.

1. Could you comment on how this amp sounds. I have read many of your threads on various amps that must have been good-sounding too. How does this one compare. How much power output?

2. Would you recommend the Triad toroids in stead of the Hammond 1650N based on performance or cost? Which of the VPT-18 transformers would be the right one?

3. Why did you include the driver behind the phase splitter. I thought the 6GB5 does not need a driver in pentode, UL mode, so I assume you needed it for driving the 6P36S in triode.

Thanks

It sounds clean and transparent, like any good amplifier should.

I usually go with the Triads based on cost. They also handle far more current, and have a lower DCR increasing output power. VPT18-5560 should be fine. Use two per channel wired series interleave. See my 6P45S monobloc thread for how to wire them.

The VA/PI/Driver is a universal design I use based on the drive requirements of sweep tubes in triode connection needing 60V-80V RMS drive to clip. If wired in pentode mode, the 6N8S section could be omitted. Use a screen voltage of ~150V.
 
Somewhere in my apparently misspent youth, I too had a bunch of 1950s and even some early 1960s high-end electronics-for-engineers books which I liberated from the nickel-a-book bin at the local library yearly sale. With deep irony, not having enough nickels … to buy all the books I found interesting was a real sobering prospect. As it was, I had to make 8 trips on my little red bike to the library and home; I emptied both my and my unknowing brother's ceramic piggy banks of all their coin. But books I bought, and read, and read, and read, and read, until their covers fell off.

No one (pessimistically) told me “little boys can't read college-level electronics engineering books, lad. Go outside and hit a ball or scare a cat or fall out of a tree or something!”

By the time High School came around, and I learned algebra, “the lights went on” so to say.

All those darn books with their hundreds of formulæ, which I “remembered, dâhmn it” by rote, suddenly took on newer, harder, more satisfying meaning. I re-read the books again … this time able to understand ALL the equations, and with my trusty slide-rule, I could calc out the examples and come up with the results that were otherwise so innocently spoon-fed by the authors to the reader. Heaven!

Somewhere along the path of life, I lost all those books, but never their contents. That is why I always dig back to “the basics” in formulæ. I really don't remember the frankly-easier-but-more-numerous 'cheat sheet' formulæ. Just the basics. ALL of the basics. Forever etched in my melon at age 10, from the nickel-a-book sale of the public library.

I come to DIYaudio emphatically not to “show off”, but to impart some small bit of this precious “muscle-memory” electronics mathematronics to those who might be interested. Surprisingly many of us are, yet depressingly fewer each decade. I think the great “new world” will be dominated by SPICE sim-math, and with almost soul-searching irony, I can say … its really a good thing. We don't really need to remember 2πFL and ½πFC I guess. Not when a time in a few CPU seconds can compute response curves and plot them from 1 Hz to 1 MHz … for arbitrarily complex circuits having surprisingly well-modeled mathematical components. It is the future.

Lastly, you're welcome, Koda. I'm glad to know you.

GoatGuy ✓

Quite amazing. When I was that age, I was taking apart tape recorders and the like to "see how they worked". I had a book called "Wires and Watts by Irwin Math" (still in the reference library here in Toronto) which taught me the basics of electricity, series and parallel circuits. Even how to make a fuse from tin foil. Then I started building computers. I figured out that I could use a 1/4 HP motor to turn the motor from an IBM floppy drive using a rubber band as the belt, and use it as a generator that would power my Nintendo. Then I got older and started repairing my car. Even yesterday I did the brakes on my car. Everything I know about tubes was learned from old books and people like you on this forum. I wish they still taught electronics courses based on vacuum state but no more.

20 years ago, my computer would take almost an hour to encode an MP3. now it takes a few seconds at most. The future is now, apparently :D

I'm glad to know you, too.

Koda
 
Last edited:
Ok. One more update to the design.

I added an RC filter to the main power supply. This reduces ripple to ~8mV.

I also cleaned up and corrected the main schematic. The grid stoppers are in the right place now, and the values are correct.

I rewired the entire rack today, and chased some hum away in the process. Results!

Koda
 

Attachments

  • 6F12P-12SN7-6P36S-3.svg.jpg
    6F12P-12SN7-6P36S-3.svg.jpg
    799.3 KB · Views: 317
  • 6F12P-6N8S-6P36S-Power-Supply.svg.jpg
    6F12P-6N8S-6P36S-Power-Supply.svg.jpg
    530.3 KB · Views: 300
Question for those among us with "mathing" skills.

If Ra-a=4300 and screen taps are at 40%, does that mean Rs-s = ~2000? I'm looking for a way to get lower Ra-a from these but still be able to switch 4R or 8R secondary instead of using 16R for 6R to reflect the lower Ra-a.
 
Ya. In the two years since this thread was last touched, the MA-1 was created, and all outputs are now using dual triad PT's as OPT driven by triode connected horizontal or vertical sweep tubes, and those 4k3 Hammond coils have been collecting dust under the bed.