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 7591A Amp Design, 35W x2
New 7591A Amp Design, 35W x2
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 13th February 2020, 04:30 PM   #1
TavishDad is offline TavishDad  United States
diyAudio Member
 
Join Date: Aug 2014
Location: New York
Default New 7591A Amp Design, 35W x2

I was inspired by a thread I read here on diyaudio back in November about current production 7591A tubes by New Sensor and JJ. Id known about the 7591A, but I hadnt played with it before, and I hadnt known that it had been reintroduced by two different manufacturers. I designed a circuit board for a very simple stereo amplifier and built a prototype. It turned out better than I expected, so I thought Id share it.

I really like this little tube!! It seems that Westinghouse was telling the truth about its linearity.

I spent a few days with a THD analyzer optimizing component values, and especially exploring the operating point for the 7591A push-pull output stage. Im getting about 35W per channel with both channels driven, and quite respectable THD readings. The Tung Sol Reissue 7591A works well, and I have a set of JJ 7591S that I intend to try as well. The JJ 7591S in particular seems reasonably priced, at least in the US.

I had originally intended to use ultra-linear (UL) operation, but I found that the 7591A seems to work much better as a pentode (see the THD readings: "proto_meas.pdf"). And when I added a 390v screen regulator, I was very pleased to see the improvement in THD. THD at 1kHz is below 0.1% up to 25W, and only 0.15% at 35W. Note that the THD data was taken with a 5-ohm load on the 4-ohm tap of a Triode A-470 output transformer, which yields an effective 5400-ohm plate-to-plate load. Power output and distortion are almost the same with a 6500-ohm plate-to-plate load.

The amp sounds quite nice when driving my Maggies, even with its modest output power. The next step for me is to revise the circuit board design, to include the screen regulator on the PCB. And then build it into a box. The PCB is only 10 x 6, so the final amp will be quite compact, limited by the size of the transformers.

But first, I was curious if anyone in the diyaudio community has insight into the 7591As strong preference for operating in pentode mode, especially with the regulated screen supply?? Looking through the schematics of classic equipment from the 1960s, I dont find any using the 7591A in UL mode. I did see a post from a couple years back using the 7591 with cathode feedback, but without any measured data. If theres anything I can do to improve the linearity even further, its best to do it now, before I redesign the PCB.

If there is interest, Id be happy to post the revised schematics when the new PCB is done. New amp designs using these reintroduced 7591 tubes seem relatively rare, but I think they deserve more attention.
Attached Images
File Type: png 7591amp.png (24.8 KB, 217 views)
File Type: png 7591powersupply.png (10.9 KB, 217 views)
File Type: jpg allegro_prototype 001_comp.jpg (181.4 KB, 215 views)
File Type: jpg allegro_prototype 003_comp.jpg (193.6 KB, 212 views)
Attached Files
File Type: pdf proto_meas.pdf (25.9 KB, 29 views)
  Reply With Quote
Old 13th February 2020, 04:41 PM   #2
petertub is offline petertub  Sweden
diyAudio Member
 
Join Date: Jul 2015
Location: gothenburg,sweden
New 7591A Amp Design, 35W x2
Keep on, hard data is welcome so is listeing impressions
__________________
My home is at www.tubular-well.se
  Reply With Quote
Old 13th February 2020, 05:12 PM   #3
Eli Duttman is offline Eli Duttman  United States
diyAudio Member
 
Join Date: Apr 2004
Location: Monroe Township, NJ
I have some "tidbits" for you to mull over.

The 7591 is easy to drive, a fact that Fisher, Scott, Sherwood ... took full advantage of. Small signal circuitry that works with "12" W. tubes, like the 6BQ5, also works with the 7591. Your Mullard style setup may have too much small signal gain.

Current production 7591s are intolerant of liberties being taken with the grid to ground resistance limit specification. FWIW, my method for "finessing" that issue is employing combination bias. "Stand" a PP pair on a shared 100 Ω/47 μF. bias network. A single bias trim pot. suffices for the pair. Notice the convenient bias set test point. Matching requirements are reasonably relaxed. As is always the case, gm numbers must be close, but some variance in cathode current numbers is OK.

Search the archives here and over on AA for "El Cheapo Grande".

Your findings about UL vs. full pentode are interesting. Perhaps employing (costly ) separate screen grid winding O/P "iron" would yield yet better results. UL combined with regulated g2 B+ has been previously shown to be advantageous. Another, less expensive, possibility is "iron" whose UL taps are at approx. 20%. Edcor has at least 1 23% model. Maybe Edcor would wind 6.6 Kohm primaries with the UL taps at approx. 20%.
__________________
Eli D.
  Reply With Quote
Old 13th February 2020, 05:27 PM   #4
audiowize is offline audiowize  United States
diyAudio Member
 
Join Date: Aug 2005
Location: Seattle
Send a message via AIM to audiowize
I believe the driving force for the 7591A was to create a tube that would fit in the receiver form factor and allow for a "100W" output power rating. To get there, pentode operation sure helps and the Westinghouse datasheet provides some UL operating points that support this.

I still need to bench test my 1000A to verify the 100W claim. I suspect it's close, but not close enough!

Let us know what you think of UL, you may be one of the only people alive to evaluate 7591As in UL.
  Reply With Quote
Old 13th February 2020, 06:37 PM   #5
ontariomaximus is offline ontariomaximus  Canada
diyAudio Member
 
Join Date: Jan 2011
Looks like a nice amp! Your topology is exactly what I have been thinking about for the past few weeks, though I was thinking 6L6's. I just bought a pair of 8k 40W output transformers for a stereo amp project. Can you talk about your CCS and where is VCC going?

I have a pair of IXCP 10M90S - can those be used?
  Reply With Quote
Old 13th February 2020, 06:40 PM   #6
tikiroo is offline tikiroo  New Zealand
diyAudio Member
 
Join Date: Nov 2007
Location: New Plymouth
I have an amp with 7591's. Was initially in triode mode but power was too limited so I changed to pentode with screen regulation. I don't have ultralinear taps on the output transformer so can't try that. I initially had JJ's installed but one went bad after about 4 years and I replaced all four with EH's. Power output increased, I think the "knee" on the JJ's is at a higher voltage so I was getting reduced maximum voltage swing.

For your design (which looks great) you might want to rearrange the bias circuitry as with the current arrangement if the pot wiper loses contact you lose bias voltage. Other arrangements can drive the grids to full negative voltage if the wiper loses contact.
  Reply With Quote
Old 14th February 2020, 04:32 PM   #7
TavishDad is offline TavishDad  United States
diyAudio Member
 
Join Date: Aug 2014
Location: New York
Quote:
Originally Posted by ontariomaximus View Post
Looks like a nice amp! Your topology is exactly what I have been thinking about for the past few weeks, though I was thinking 6L6's. I just bought a pair of 8k 40W output transformers for a stereo amp project. Can you talk about your CCS and where is VCC going?

I have a pair of IXCP 10M90S - can those be used?

OntarioMaximus,
Vcc is a 6.3v DC regulated supply for the 6SL7GT heater. I posted the schematic below. Your IXCP 10M90S should make a fine current source, but you'll have to use a different circuit than I did. My circuit is just the classic bipolar transistor constant current source, which works well here.



Everyone,
Thanks for your comments! I re-posted the THD measurements as a png file, so they're easier to look at. So far, I've drawn two conclusions:

1. UL operation doesn't seem to be a good choice for the 7591A. If you look at distortion at moderate levels around 12W, THD in UL mode is almost 8 times higher than it is with a regulated screen supply.

2. The 7591A is an exceptionally linear tube. You'd have to work pretty hard with any other common beam power tube to get distortion as low as the 7591A gives with just this simple circuit. I think the Westinghouse engineers did a good job back in 1960. And I think the Tung Sol Reissue 7591A is a very good copy of the original. I have the Westinghouse 1963 data sheet, if anyone would like me to post it.

I plan to measure THD with the JJ 7591S this weekend, and I'll post the results.

Scott
Attached Images
File Type: png 7591heatersupply.png (9.3 KB, 77 views)
File Type: png THD_vs_operating_mode.png (72.2 KB, 45 views)
File Type: gif 7591-1960.gif (74.9 KB, 40 views)
  Reply With Quote

Reply


New 7591A Amp Design, 35W x2Hide 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
PP amp with Toriody Transformers and 7591A gregt Tubes / Valves 35 24th October 2018 12:48 AM
7868 to 7591a conversion help vrusso123 Tubes / Valves 2 13th July 2014 04:29 PM
Acrosound 7591A amp mr2racer Tubes / Valves 4 3rd July 2011 04:11 AM
What the difference between 7591a & 7868 hoyhoysum Tubes / Valves 16 19th December 2009 04:47 AM
7591a Alternatives wrenchone Tubes / Valves 16 4th June 2009 06:34 AM


New To Site? Need Help?

All times are GMT. The time now is 02:35 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki