audio-talk

Not much progress, still in the thinking stage. Had a play around with the 6EM7 curves and then went back to the 6DN7 curves.

The previous plan was to use Mr I's big tx for the GM70 and a 750V CT Hammond tx for the driver/pre. Then I had a play with using the big tranny for the whole amp, direct coupling the two stages of the 6DN7, and ended up with the circuit attached choke loading both sections of the 6DN7.

Dunno if it's right though . Any thoughts/comments/criticisms, particularly the direct coupling? I wonder if both 6DN7 sections are biased a little close to 0 grid volts...

Power supply will probably be CLCLC for the GM70s, and LCLC with a tuning cap for the 6DN7s. Voltage regs for the 6DN7s and probably passive DC for the GM70s.

Welllll I can see what you are doing, one way or another - Aren't you going a bit the extra mile with the PSU config. though??

Why not build a 6DN7 amp and a GM70 power stage.

That way you get two amps for (almost) the price of one. Bit like when I used a 300b amp to drive my 845 power stage.

Biasing looks alright to me, dunno what much else you can do with them. Gaps between grid lines look even.

Don't need the 1M resistor though on V2 grid.

 

andrew Ivimey wrote:Aren't you going a bit the extra mile with the PSU config. though??

Because of the two mains txs, or the two chokes? Yeah possibly on both counts, but I have the iron . The (C)LCLC reduces the HT ripple to very small, and the separate tx for the 6DN7 means I don't have top drop a load of voltage, but that's perhaps a bit spurious...

I guess using the single tx would let me move the OP away from 0 grid volts.

pre65 wrote:Why not build a 6DN7 amp and a GM70 power stage.

Unless I'm misunderstanding you Phil isn't that pretty much what this is?

Mike H wrote:Biasing looks alright to me, dunno what much else you can do with them. Gaps between grid lines look even.

Don't need the 1M resistor though on V2 grid.

Thanks Mike.

Ah, so if direct coupling I don't need a grid leak?

simon wrote:

pre65 wrote:Why not build a 6DN7 amp and a GM70 power stage.

Unless I'm misunderstanding you Phil isn't that pretty much what this is?

Spose so !

Oh, should give 15W .

simon wrote:Ah, so if direct coupling I don't need a grid leak?

Yes indeedy.

It am do get its grid DC bias orft first anode innit

Think what Phil meant was with that much current on V2 it can be an output stage as well.

Which is kind of where we came in

 

Mike H wrote:Think what Phil meant was with that much current on V2 it can be an output stage as well.

Yep .

Three and a half years after starting this thread I'm spurred on to actually get something built after Andrew's amp at Owston. The power supply tweaks for Madness will have to wait. So after giving myself a good talking to I'm going to get something built with a 12GN7 driver a la Nick Chua and our Andrew to "get me going".

Looking at the biasing of the GM70 I can't get it straight in my head. For arguments sake let's say I'm biasing it at 525V, 95mA and -38 grid volts. With an indirectly heated valve this gives a cathode resistor of 38/0.095 = 400R.

As I'm going to use DC filaments there'll be 20V potential between the two ends of the filament. This makes quite a difference, so am I correct in saying that if I attach Rk to the + filament say the voltage on that pin should be the bias plus half the heater voltage so (38+10)/0.095 = 505R?

I used 560R when I did my version.

Your calc looks right Simon.
Good to see a thread up on this subject.
I have been quietly buying a few bits for the filament supplies already , as well as the valves ( which have left Uzbekistan but not entered Parcelforce.. ) .
So what value of transformer primary are you going to be using ? ( 7k ? ) .

Thanks Mark, sometimes I get mental block with this electrickery.

I have a pair of AE amorphous OPTs just waiting to be put to good use again. They're 5k (100mA 25W) which is a bit low really, but I have them so they're the right price! They're a decent compromise for GK71 too and the plan is to have a dual setup using the same HT and filament supplies, same OPTs, and same driver. Maybe a slightly flawed comparison but should be interesting nonetheless.

Just need to gather the filament raw DC supply and I'm nearly ready. Then I just need to find the time amongst everything else