DCAudioDIY.com

If you would like to see the various voltages at various points in my dc-coupled 300B Siegfried amplifier they are indicated on the schematic in the owners manual. This can be found on my web site http://www.davidberning.com/products/siegfried
Note that the cathode for the 300B sits at 386 volts and the plate sits at 800 volts with respect to ground.

Of course, all of these are controlled by servo from the speaker output. This uses the ZOTL, which is an output transformer emulation circuit. The MOSFETs are not an amplification stage and there is no power gain associated with this. Just pretend that the string of MOSFETs is an output transformer that can pass dc.

I feel that the voltages represented here are whats needed to get the dynamic voltage swing required to drive the 300B, and that's why I am passing this on.

David Berning

That's very interesting David. I've never seen the schematic before. I'm pretty wide of the mark in terms of headroom for the driver, I may have to rethink the implementation and do a separate power supply for that. The current version has to fit on a pair of old Heath W5 chassis, so there's a limit to what I can do. That said, it sounds pretty good. It's not a power issue (12 watts would be sufficient) but I'd like to think it will be clean and sweet. Right now, I've got a 6J5 with 47K on the plate and 470 on the cathode, direct-coupled to a long-tailed pair 6SN7 with 33K on the plates and 12K on the cathode. The B+ is 430 volts through a 6BY5. The 300Bs are running at around 410VDC with a combination of fixed and cathode bias, because my bias tap is smallish and it actually sounds pretty good. 280 ohms on the cathode and -60 on the grids for about 70 mA each.

I tried a classic Williamson and my wife, Tanya, said, "Ecccchhh, what happened to the other circuit?"

One question I have is that it seems like you are running the second 6SN7 pair at very low currents. I assume the plates are sitting at around 300 volts, with 275 volts across the 160K plate resistors. Is that the case? I think I could squeeze a bit more swing out of my driver stage with higher plate resistors, a longer tail and lower current. I'd be interested in your thoughts.

In case anyone is wondering, these are for my friend Jim (some of you know him and have even graciously repaired some of his equipment). The idea is a budget PP300B amp to get a taste of what's possible. I've been working on my own for a while and I like what they do.

One thing about Jim is that I know what he'll do: After three months he'll say, "What about V-Caps? Would they sound good?" So I'm doing something I don't think I've ever seen done before: The coupling caps to the 300Bs will be swappable via a set of banana plugs on the chassis. So rather than schlep the amps back and forth across the country, he'll be able to buy some V-Caps, attach banana plugs and insulations sleeves to the leads, and plug them in. I have to see how it sounds--it may be awful, but then again it may be a cool idea. Of course, he can buy 6SN7s to his heart's content, and for the 6J5 there are a number of options, not the least of which are the gorgeous Mullard L63s you can still get on eBay--for astounding sums. (Never underestimate Jim's pocketbook. He just sent me *SIX* V-Cap oils from some mod board for a CD player.) So the idea is that, within its limits, he can play with tuning in a wide variety of ways. He currently has a pair of ProAc Response 1's, so I have a fair idea of how they'll sound.

BTW. my current coupling cap of choice is the Mundorf Supreme (just Supreme, no silver, etc.) About $15 a pop, and they are clean and very musical. A good deal, IMO.

Grover,

I'm using a simple pair of MOSFETs as a constant-current source as plate loads in my phono amp (that somehow hasn't been finished in a couple years.)

You can see the circuit in the "Project Discussions" section under "Guy's Phono Amplifier." The top MOSFET handles all the current, and I can adjust the current through each tube to get the plate voltage I want for a given tube. I even have some PC boards to assemble four of these CCSs, and cut the board in half to make two pair of CCSs with different supply voltages. If you have access to Facebook, go to the group DCAudioDIY, and look under Photos -> Albums for "Quad DMOS CCS PCB." It shows the boards, schematics, and a brief discussion of how to use them.

It's a simple circuit that works well given it's simplicity. It would be easy to build on perf board if you want to try it out.

_________________
- Guy

You'd have some turn-on transients because the 0A3 needs up to 105V to ignite the gas before it starts regulating. This happens very fast but it's something to consider. I think Charlie Phelps has proven that VR tubes fed by CCSs are very quiet. Obviously Roscoe has experience as well.

_________________
- Guy

Grover,

A suggestion about swapping caps, it may be more secure to use a screw type barrier terminals with crimped or soldered "spade" connectors on the capacitor leads. You can get a good airtight/high pressure connection that way. I would not trust banana plugs.

David

That's a good point, David, thanks.

It looks like I have about 310 volts across the 160 k resistors, and the plate to cathode voltage is around 200, as the actual plate is at 265 V and the cathode is around 65 V for this second stage. This is 2 mA through each 6SN7 plate and is a nice sweet spot for these tubes.

David

Grover,

Another thing I should mention here is that I do like to keep the currents as low as practical as it does allow more dynamic swing. In my case with just about everything I build dc coupled, at least in part, this is important so that the build up of dc voltage with each stage does not get too crazy. I do find that to get the desired bandwidth on the output triodes that I do need the follower. One advantage of the lower plate resistors (33 k) that you are planning to use is that you can probably get away without the follower. But you may want to bump up your voltage supply for this stage to get the dynamic swing up a bit so that the distortion in the driver stage remains low.

Good luck,

David

David, thanks a lot, that's big help. I think I've found a way to get more B+ to the input stage.