Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

New build: 6F12P - 6N8S - 6P36S triode push pull.
New build: 6F12P - 6N8S - 6P36S triode push pull.
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 9th September 2019, 10:46 AM   #11
baudouin0 is offline baudouin0
diyAudio Member
 
Join Date: Mar 2019
Many thanks did not know there was a PCB, I think I will get a couple of the boards in for the 6550 output stages. One thing then is I don't have to use such a low value (47k) of grid leak resistor which may make the driver easier.
  Reply With Quote
Old 9th September 2019, 06:02 PM   #12
smoking-amp is online now smoking-amp  United States
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
21HB5A has almost identical plate curves to the 6GB5 tube on the curve tracer here. Similar ratings mostly, 18 Watt Pdiss, no plate cap, 12 pin. It's a TV Horiz. sweep tube also.
  Reply With Quote
Old 9th September 2019, 06:14 PM   #13
piano3 is offline piano3  United Kingdom
diyAudio Member
 
Join Date: Aug 2010
Location: London,UK
Yes, 6P36S is often described as a 6GB5 equivalent, but actually is nearly twice the size. Thus Koda has been running at nearly 30 watts dissipation without problems.
  Reply With Quote
Old 9th September 2019, 06:32 PM   #14
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
So… KodaBMX, let's see if I get the circuit.

U1A 6F12P pentode: voltage amplifier, inverts signal. Has GNFB injection to cathode. Cool.

U1B 6F12P triode: phase splitter, no gain. Cool.

U2A 6N8S triode: slightly cathode-lifted (wider dynamic range … but enough?) VAS

U3A 12SN triode: Same thing as U2A, but different valve? Hmmm…

U4,5 6P36S triode: Capacitively coupled push-pull output power valves. Cool.

Either the U3A as a 12SN7 is a typo, or it begs why. I'm taking a bet on “typo”.

I was kind of curious as to why the phase splitting outputs are then voltage amplified in the next pair of valve sections. On the one hand “I see it”: whatever nonlinearities of amplifcation of U1A will be modest if the signal is low. Even amplified, a low signal won't have the ability to drive the 6P36S valves on the east side of the 0 dB gain phase reflector.

So, added gain needed. I guess the “feature” is that then using 2 triodes to amplifier each symmetric phase introduces more triode nonlinearity into the signal stream, but now it is complimentary, not just more-to-postive-over-negative signal excursions. If that was the thinking, then I like it.

Finally, one must hope that the output valves' grids never venture in to the positive charge space by much. The prior pair of VAS valves' anode load resistors (hence overall impedance sort of) is fairly high, and actual positive-going signal conduction would add quite a bit of nonlinearity to the output.

Anyway, that's my take.

What transformer doth thou recommend for the PP output? Yet another of the near-magical toroidal power transformers commandeered to the job?

Yours,
GoatGuy ✓
__________________
Burnished unicorn horn circuit design only go so far. When the unicorn's magic fails, its best to rely on maths. JustSaying™
  Reply With Quote
Old 9th September 2019, 07:08 PM   #15
kodabmx is offline kodabmx  Canada
Spreading triode love.
diyAudio Member
 
Join Date: Oct 2011
Location: Toronto
New build: 6F12P - 6N8S - 6P36S triode push pull.
A couple of updates. The amp is now running at 375V B+ for the output stage with each tube fixed biased to 60mA. I blew up a board (make a couple of errors while building including connecting the screens of one tube to the plate of the other. This blew the 220R screen resistors (which I spec as 0.5W film because they blow like a fuse and save the tube) and the cathode resistors on the module.

The VA/PI runs on 530V, the driver 560V.

A curiosity is with the gNFB network. To get 10 kHz square wave to look good, one channel needs 2000pf || 1k, the other only needs 1000pf || 1k.

GoatGuy. I was in the middle of things when I wrote the schematic... 6N8S = 12SN(7) in this case, I forgot to change both of the labels.

I thought a split load phase splitter had a gain of 2? Doesn't matter anyway.

The point of the extra driver is for voltage swing. This VA/PI/Driver should make over 100VRMS from 0.2V input. This allows me to use this circuit for 6P45S triode connrecion (-80V bias roughly)
The output tubes have about -60V bias so the driver needs to be able to swing it for full output which happens before grid is at 0V (The 6P36S will try and move over 1A of triode plate current with the grid still under 10V negative).

It also allows for more gNFB.

I'm using Hammond 1650N for the output, but I'd recommend the dual toroid method. Triad VPT18 with the 4 primaries interleaved in series (to cancel DC core flux) works very well. Not only is there less distortion, there is more power, and less likeliness to "motorboat" with too much gNFB. Also much more robust. An arc won't take out the coil as easily since they are designed for much more current.

I bought all the iron before I discovered toroids and triode connected sweet tubes, and these parts don't really suit anymore so I might sell the lot. Great to make a KT88 amp or the like

Koda

Here's the current schematics...

And a photo of it running...
Attached Images
File Type: png 5F12P-6N8S-6P36S-Power-Supply.svg.png (321.9 KB, 120 views)
File Type: png 6F12P-12SN7-6P36S-2.svg.png (469.1 KB, 118 views)
File Type: jpg 2019-09-09 16-45-29.jpg (616.3 KB, 107 views)
  Reply With Quote
Old 9th September 2019, 09:17 PM   #16
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
Quote:
Originally Posted by kodabmx View Post
Here's the current schematics…
Thought… maybe R15 and R16 might be better around 4.7 kΩ. A bit more linearity over the whole dynamic range, and less likelihood of having an “internal stage clipping” situation. Won't change the overall gain too too much.

Just saying,
GoatGuy ✓

PS: I see the 12SN7 mislabeling has “gone away!” Bravo, bravo.
__________________
Burnished unicorn horn circuit design only go so far. When the unicorn's magic fails, its best to rely on maths. JustSaying™
  Reply With Quote
Old 10th September 2019, 03:35 AM   #17
Junm is offline Junm  Philippines
diyAudio Member
 
Junm's Avatar
 
Join Date: Dec 2008
Location: Dau
Send a message via Yahoo to Junm
just a question, i think the supply (575v) for the driver section is by a voltage doubler but on the schema is just FW rectification, Is that correct.
  Reply With Quote
Old 10th September 2019, 03:41 AM   #18
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
The 375V supply is *choke input*.
  Reply With Quote
Old 10th September 2019, 05:50 AM   #19
baudouin0 is offline baudouin0
diyAudio Member
 
Join Date: Mar 2019
You can solve the motor-boating issue by making sure just one of the coupling stages is the dominant pole at LF. In this case reducing the value of C8 and C13 - and making C1, C3 larger - you have add a zero C4 in the NFB which will help with the correct choice of value. Having two poles C8/R40, C3/R11 in the same place is a bad idea. Maybe C8/R40 10Hz, C3/R11 2Hz. The transformer pole is about 3Hz.

For the HF stability it may be worth adding a dominant pole say 12KHz with a zero at the first pole of the transformer about 80KHz. This will bring you loop gain down to a much lower value before you add the zero. You could try series RC across R5 - however if the plate resistance changes with voltage then a non-linearity is created. Don't know - others may give you better advise - I would try the first stage on ltspice.

Sorry C4 is a pole too in the nfb for the LF stability. So you have 4 poles. Nothing you can do about the transformer - so make two poles low in frequency say 2Hz or less and one pole (dominant) say 15Hz to the output stage - this also helps with blocking. Then the dominant pole will bring the fb gain downto 0dB at which point the transformer pole will kick in.

No point having an amp near LF or HF instability.

Hope this helps.
  Reply With Quote
Old 10th September 2019, 11:39 AM   #20
kodabmx is offline kodabmx  Canada
Spreading triode love.
diyAudio Member
 
Join Date: Oct 2011
Location: Toronto
New build: 6F12P - 6N8S - 6P36S triode push pull.
Thanks for the suggestions. The amp is stable as it's built, but if more gNFB was needed, these suggestions would increase stability for sure.

Your suggestions made me realise I made an error in the schematic. R38/R40 are 330k, not 510k, so that pole is different than the first two. Not 15Hz of course, but higher than 1.6Hz...
  Reply With Quote

Reply


New build: 6F12P - 6N8S - 6P36S triode push pull.Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
6F12P and 6P43P push pull triode amp kodabmx Tubes / Valves 6 5th August 2019 11:54 AM
12AV5GA push pull triode amplifier. kodabmx Tubes / Valves 22 12th November 2018 11:22 PM
6L6 Triode AB Push/Pull Pout famousmockingbird Tubes / Valves 18 10th March 2015 08:00 PM
Push Pull DHT triode James D. Tubes / Valves 14 8th April 2004 03:43 PM
DHT triode push pull amplifier design James D. Tubes / Valves 2 5th December 2002 08:44 AM


New To Site? Need Help?

All times are GMT. The time now is 04:31 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki