• 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.

Need PS schematic for Musical Machine

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've decided to build Poins' Musical Machine amp, but a slightly modified version with 6sn7 driver and Pimm CCS. I also intend to build the amp as a pair of monoblocks and am looking for a very good, preferrably tube rectified, LC power supply design that would be sufficient for one channel. I will use film caps (no electrolytics) and want something that is robust, clean, and very quiet. I'd also more than likely have custom PTs made for the project so as to simplify things for the 4 different voltages needed.

If anybody could help, it would be much appreciated.

I'm blind so can't work with schematics very effectively, nor tools such as PSUD II, and would like to find something a little simpler than the design posted with the schematic on Poins' web site.

This amp will be a labor of love, so I am looking for quality and performance as opposed to on the cheap.

Thanks,

Jim
 
Jim,

6SN7s in the LTP will give you a stage gain of approx. 5. That's WAY too little :( Poinz switched from the 5965 to the 6GK5, in part, because a mu in the 40s proved marginal.

As for your vacuum rectified B+ supply, the 6BY5 dual damper feeding an ARRL approved LCLC filter should prove highly satisfactory. IMO, you should use a high quality 'lytic in the 1st LC section. That cap. is not in the signal path and there is no good reason for failing to realize the space and cost savings a 'lytic provides.
 
AA10,

Thanks, but aside from the trouble of dealing with schematics, and as long as I have a good written description of the circuit, it's generally not too difficult. I had 20+ years of experience working on electronics before I lost my sight 19 years ago, so I'm sure that helps a lot. I'm learning to dislike PC boards though :).

Eli, well, I see no mention of changing from the 5965 so I suspect that the web site is way, way out of date. I'm also not exactly sure what an ARRL approved supply is. I know what ARRL is, but that's about it.

I'm kind of cautious whenit comes to damper diodes when it comes to full duty cycle power supplies. I know people rave about them and some of their advantages seem appealing, but they were never designed for continuous duty use, but rather intermittent, pulsed power supplies, and it makes me wonder how long the will actually hold up in a power amp.

Hopefully Poins will chime in here. I sent him a couple of emails this week but have not heard back yet.

Thanks,

Jim
 
Administrator
Joined 2004
Paid Member
6BY5 has proved quite reliable in applications in excess of 100mA continuous in a number of friends 45 based stereo amplifiers, and in my headphone amplifier over a good number of years. IMO They seem to outlast current production 5AR4 in similar applications and the Sylvania and GE types I have used have given me no trouble at all, not something I can say for anything other than a Mullard or Tele 5AR4...
 
Poindexter's 6GK5 version is shown in posts 6&7 of this thread, one with a mono tube rectified B+ and one with SS rectified stereo B+

http://www.diyaudio.com/forums/showthread.php?s=&threadid=136562&highlight=

I am in the final stages of breadboarding this design; presently installing a startup B+ delay due to SS rectification. Poinz has been very helpful answering my newbie questions although it usually takes him a few days.
 
Got it.

The 6GK5 version of Poindexter's Music machine is in post #2 of this thread, and the (SS) power supply is in post #27.


http://www.diyaudio.com/forums/showthread.php?s=&threadid=124853&highlight=

It appears that it would be fairly straightforward to downsize Poindexter's tube rectified PS for the mono 6V6 version. I have quite a number of PSUD models of both the SS and 5AR4 supplies modeled, so I'll give it a crack and post the results if you have your transformer voltage and secondary impedance handy.
 
Jim,

Let me provide some of the details. The LTP employs a 10M45S as the tail load, to force symmetry. The CCS is set for 9 mA. and the B- rail is 66 V. 43 KOhm/4 W. 6GK5 plate load resistors are called for. 300 V. is applied to the 6GK5 loads. The B+ rail voltage is 320 V. The C- voltage is 22 V.

Reply to the EMail I just sent and you'll get Poinz's gif, as an attachment.

As you are planning on monoblocks, 6SN7 LTPs would be OK, provided you use a line stage with gain. I suspect that buffering DC coupled MOSFET source followers are necessary. That would not be bad, as the 6V6 grid leak resistor value could come down.
 
As far as your bias transformer goes, the Amveco 62035 (7 VA for a stereo bias supply) is also available as a 3.2 VA transformer for a mono application (62015). It's also available as a PCB mount potted version (that's what I used) as the following P/N's 70035 (7 VA) and 70015 (3.2 VA); both available from Digi-key. They actually give 44VAC with the secondaries wired in series.

If you use a primary transformer with a 50V bias tap (such as a hammond 300 series), you can skip the amveco xfrmrs above entirely and just use the bias tap, although if you want to energize the bias supply before the B+ comes up (start-up delay) a separate bias transformer gives a few more options.

You could also (I'm pretty sure) provide 320V B+ to both the 6GK5's and the 6V6's to simplify the PS design slightly, although the additional cap and dropping R are easy enough to implement. A Hammond 300 series (370DX or so) can do all the B+ voltages you need for a mono application.

What are the largest film cap values that you are planning to use in the PS?

I have just modified one of my PSUD PS schematics for this app with a 5AR4 and a 275V secondary transformer as a CLCLC supply but it is difficult to post the results here and still have them be legible. I can email a JPEG if you are interested.

To clarify, I believe that you can get the -66V and the -22V off of the bias tap (or amveco xfrmer) pretty easily. That's how Poinz did the EL34 version. I said "bias" above but it's actually the 10M45 voltage that comes off of the amveco in Poinz' schematic.
 
You could also consider the 200 series hammond power transformers and use the Amveco for the -22V and -66V. The 200 series are less $$ and do not have the bias tap or universal input, so the cash saved can be applied to the Amveco (or other small VA transformer) and allow additional flexibility for start-up delay schemes, since you can apply the bias voltage and feed the CCS instantly upon startup and delay the primary side of the B+ transformer if req'd instead of having to deal with a start-up delay switching the secondary side.

The issue I'm presently having with the EL34 breadboard is alot of B+ overshoot on start-up due to SS rectification and a 385V transformer when the tubes are cold. I have have doubled-up all of the electrolytics in the PS to handle the voltage (close to 550V on start-up). It appears that there would still be some B+ overshoot even with tube rectification, although I haven't got that far.

With the 6V6 design voltages, 630V film caps, and tube rectification this will probably be less of an issue.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.