• 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.

SE SV-811-10 power supply

I've been wanting to build this amp for 20+ years since the Eric Barbour article in Vacuum Tube Valley (I think) came out. One representative schematic is is here: SV811/SV572 SE1.
I have a collection of components gathered together for the power supply section, but I've never designed a power supply and I'd love to have some more eyes taking a look. Using Duncan's Power Supply Designer, it seems like I will get 630V and 430V B+ and B1+, respectively, which should be OK. And fairly low ripple. But I don't know if I have correctly modeled the loads. I've attached the power supply schematic from PSUD. Anybody willing to take a look and critique?
Thanks!
Lewis
 

Attachments

  • sv811-10 PS.pdf
    31.5 KB · Views: 585
Last edited by a moderator:
Hi. Currently i am building this. If you are not in a harry i will soon post the results of the project.
 

Attachments

  • 811akairo.gif
    811akairo.gif
    10.6 KB · Views: 1,309
Karsten, thanks for the input. If I understand correctly, you are talking about the final section that is labelled "RC Filter Section." In the Power Supply Designer simulation, that section drops the voltage down to 430V and also reduces a 7 mV ripple to less than one mV. Is your suggestion to replace that RC filter with just a resistor or maybe a voltage divider?
 

PRR

Member
Joined 2003
Paid Member
....it will not work with SV811-10. ...

I agree it looks very dubious. The power tube is nearly zero bias with large positive swings. And with C-R coupling to the grid. At first glance it "must" de-bias itself on every positive peak. The 4uFd cap "sees" 47K on the down-swing and <1K on the up swing. It is a rectifier which will ratchet-down the grid voltage with a 0.2sec recovery time.

I found the original. Vacuum Tube Valley, Issue 5, Fall 1996. Here's a dim scan.

The link that Lewis1 gave is a faithful copy of Barbour's plan. The text gives Barbour's thinking. I trust Barbour >99%... it worked for him. The last page has reviews by other ears, who praise the sound.

It is worth noting that the performance report shows peak power higher than continuous. As expected for a de-biasing amp. However 7 Watts continuous is not a small amp, and speech/music is far from continuous.

All I can think is that the grid impedance of this tube is not that low, and in real listening rarely driven hard.
 

Attachments

  • Barbour-SV811-10.pdf
    477.7 KB · Views: 389
I too am gathering materials to build the Barbour 811-10 amp and I have a pretty specific power supply in mind. You can skip the barrier diodes, or, if you like, I can go into further detail as to why they are there in an aside (PM).

The other file is from Glass Audio containing a suggested rework of the front end. Please excuse all the margin notes unrelated...

I am following in the footsteps of SET12 from AudioKarma who's 811-10 build I heard some years back and it's been an inspiration ever since.

Ill be looking for your further posts regarding the 811-10. I'm taking my time with this build, savoring it. I'll post here when it is all done...maybe fall of this year.

Best of luck to you!

:)
 

Attachments

  • 500VDC B+ copy.jpg
    500VDC B+ copy.jpg
    454.8 KB · Views: 1,008
  • Dan's 6em7 driver copy.jpg
    Dan's 6em7 driver copy.jpg
    924.3 KB · Views: 909
jdg,
Interesting stuff. I flipped through all the VTV issues this afternoon looking for a follow-up, because I remembered that there was one. Just didn't realize it was in a different mag! I'll have to pull out my stack of GAs to see if I have that issue.
Pardon my ignorance, but what are the 6DEAs in your PS schematic?
 
.....
It is worth noting that the performance report shows peak power higher than continuous. As expected for a de-biasing amp. However 7 Watts continuous is not a small amp, and speech/music is far from continuous. ....

My notes show that on a modified TubeLab SE board, with SV 811-10 I was getting 9.5 watts continuous at 1 KHz. Idle power was only thirty (30 ) watts. OPT was Hammond 125ESE at 5K, 445 volts on the plate, 67 mils. Lots of power left on the table.

Tubelab's Powerdrive will drive the ever loving snot out of these things. I wouldn't even fool with a cathode follower. My amp is torn down for overhaul and a bigger power supply, unfortunately other projects have moved ahead of it.

Win W5JAG
 

Attachments

  • TSE_811-10_1kHz_sine_yee_ha.jpg
    TSE_811-10_1kHz_sine_yee_ha.jpg
    436.5 KB · Views: 524
Last edited:
.... I found the original. Vacuum Tube Valley, Issue 5, Fall 1996. Here's a dim scan.

There are also at least three ( 3 ) technical bulletins worth taking a look at: Svetlana U. S. nos. 8, 9, and 46, describing various implementations and technical details of the SV 811-X / 572-X tubes.

I have them, unfortunately they are in HTML format. I note that when I click on a picture link, it takes me to the Internet Wayback Machine, so they can probably be accessed by going there, and pulling up the old Svetlana website .....

Win W5JAG
 
The full write-up starts in Glass Audio Volume 8, Number 3 (1996) and then Eric Barbour has a follow-up with some specific circuit changes in Volume 9, Number 5 (1997). Seems like the VTV article was sort of a side thing. Much more detailed information in Glass Audio articles. In fact, the GA articles read much like a construction guide, along with the discussion of the how and why of the circuits. I have those two issues and I'm now re-experiencing the excitement I felt when I first started planning this project. Definitely moving forward this time! Let me know if you want any info from the two GA articles or the VTV article.
 
Some more SveltanaSV811 docs

Here's a link to a directory of archived Svetlana tech bulletins:
Svetlana Technical Bulletins
And several of them related to the SV-811 attached as pdf docs. Thanks to w5jag for the idea!
 

Attachments

  • sv811-10 PS.pdf
    31.5 KB · Views: 216
  • SV tech bulletin 8.pdf
    204.4 KB · Views: 272
  • SV tech bulletin 9.pdf
    176.8 KB · Views: 249
  • SV tech bulletin 36.pdf
    165.6 KB · Views: 183
  • SV tech bulletin 46.pdf
    184.2 KB · Views: 223
I have built countless different implementations of the SV-811-10 amplifier, and I can say that ALL of them sounded fantastic. The earliest Barbour driver circuit used capacitive coupling to the 811 and provided about 6 clean watts. Sounded amazing into high efficiency speakers.

Barbour later added a cathode follower to the driver, it raised output to 8 - 9 watts and also sounded wonderful.

THE VERY BEST implementation for running A2 on a zero-bias triode is either a cathode follower that returns to a NEGATIVE SUPPLY, or a MOSFET equivalent such as the TubeLab driver board. I got 15 CLEAN watts continuous out of my SV-811-10 using a 6L6 follower that was running balls out at 50 mA idle! It is the best-sounding amp I've ever owned, and many of my audiophile friends agree.

As I was developing the CF, I started with smaller power tubes such as 6V6, and kept increasing idle current, then stepped up to larger and larger tubes with more current each time. When I finally got to the 6L6 I stopped. In my mind, I had arrived at what I was seeking.

The SV-811-10 zero bias tube is an unsung hero, the sound quality if driven properly is astounding.

One other note, the longevity of those SV-811-10 tubes is incredible. I ran my first pair at least 8 hours per day for over two years and they still sounded and tested fine on my TV-7 D/U transconductance tester!
 
Last edited:
Lewis, I recently moved to another state, so my notebook is still packed away. The implemention is pretty simple. The cathode of the 6L6 follower is connected to a power resistor as usual. Instead of the resistor bottom leg connecting to ground, it is connected to a well-filtered, negative HV supply...I think in my case it was around -300 VDC. The objective is to set idle current through the tube so that the cathode is sitting at ZERO volts, which then connects directly to the grid of the SV-811 power tube...direct coupled, no cap! Remember, the SV-811 wants it's grid at approximately zero volts for proper operation at ~ 100 mA idle. You need to add a simple bias adjust voltage to the 6L6 grid, which in turn will set the cathode at 0V +/- whatever you want.

The beauty of this arrangement is that as the 6L6 conducts, the cathode can swing both positive AND negative in a symmetrical manner. It extracts maximum swing from the driver, and yielded 15 W RMS output in my setup. Clipping was also symmetrical. It sounded great.

With -300 VDC on the resistor, and 0 VDC at the cathode, at 50 mA this works out to a 6k ohm resistor for the cathode follower. It will dissipate 15 watts idle, so use a 25 W resistor! Imagine throwing away 15 watts in the driver just to get 15 W out of the power tube? It's inefficient alright...but sounds fantastic.

You can do the same thing with a power MOSFET, as done by TubeLab in his circuit. Don't know if it sounds much different. I just like the look of all tubes.

Don't be afraid to experiment on a breadboard.
 
Last edited: