Differential input stage for T8 OTL-OCL from newbie

[aparatusonitus]

Hi all,

Below you'll find a somewhat diferent(ial) input stages for Transcendent Sound T8 OTL-OCL amplifier. It is not perfect I know, particularly anode voltage at V2A, but I am about to fix that.
I would like to know what do you think about this shematic and to point me in right direction.
Be gentle, this is mine first circut, and everything I know about tube deseign is from Bruce Roseblit's books.

Regards, aparatusonitus

[aparatusonitus]

Sorry.

[SY]

Ah, computers, making life easier!

[fdegrove]

Hi,

Hopefully this works:

DIFF STAGE.

Cheers,

P.S. The trouble you're experiencing is typical for Geocities.

[runeight]

aparatus, i managed to find my way to your diagrams!!!

i haven't fully digested your plan yet, but it looks something like this:

1. provide tube CCSs for each stage
2. use a diff amp (common cathode) for the input and send the NFB back into the grid of the outupt section of the CK stage

correct?

so, maybe a few preliminary thoughts.

the CFs that drive the EL509s see a fairly low load. not only do the 509s exhibit high interelectrode capacitance, but the 30K resistors strapping the grids to the negative bias supplies (for stability and fault tolerance) cause the CFs to have to deliver fairly high current through the coupling caps. thus, while the CCS on the cathodes of the CFs will keep current constant through itself, it won't keep current constant through the tube because the load current has to come from somewhere. so, the effect of the CCS will be diminished considerably, possibly to the point where it isn't relevant. at least this seems so to me.

long phase splitters evidently do benefit from CCSs in the tail, in which case the load resistors should be the same value. in monoblock mode, this might actually be the best use for the extra 12ax7 section that is not used.

i'm not quite sure what to make of the design for the first stage. i can see maybe adding a CCS load in the original design that uses a grounded cathode stage, but the plate resistors are high enough that it may not have much effect.

now the topology of the CK you're showing is kind of a mixed arrangement. you have a bypass cap that shouldn't be there if it is truly a CK amp. to get max gain you need a plate bypass cap on the first stage of the pair. and you're using a CCS to provide the B+ for both tubes, which doesn't reallly help much that I can see. if you really want to use a CK amp for the first stage, then put the CCS in the cathode leg, bypass the first plate with a capacitor to ground, and run the NFB into the grid of the output section. Your gains will go down about 20%.

if you're simply running the triodes in parallel, then the NFB out of phase from what it needs to be to be fed back into the grid.

perhaps further looking will reveal things i'm not seeing. please help me see what you're trying to do.

[aparatusonitus]

Let me try ones more...[IMG][/IMG]

[aparatusonitus]

And output stage...There should be 8 output tubes, but program used allow me to put 50 components only.[IMG][/IMG]

[ackcheng]

May I ask which program do you use to draw this circuit?

[aparatusonitus]

Ackcheng,

I am using Protel CircutMaker Student V6.2c

Regards, aparatusonitus

[aparatusonitus]

Maybe this is better for the input stage? Any comment?[IMG][/IMG]