The Sum of it All

???
Since this type of summing amp flips the signal, it is normally fed into another inverting stage to bring things back into the correct polarity. A transformer is used to eliminate the other active stage needed to rotate the signal and a ratio of 1:1 is selected for sonic transparency, balanced/ unbalanced  buss insert drive if needed.

Looks interesting - kind of summing to a discrete opamp with dc offset nulling ..

Nice to see another thread  from you Analag

Cheers!

is there a channel for cow bell?

we need more cowbell.

give us some more cowbell please.

long tailed diff pair with reverse vector current mirror?

gain = 1/(1-A)^2 +3X, correct?

go to see ya!

Da cowbells




The gain makeup amp.  Once again transformer ratio is 1:1.

Oh...those are pan controls, all volume manipulation is handled in DAW as it should.

Very interesting.

The opamp is working as a servo here?

-Matt

Theoretical nitpicking like it used to be here a couple of years ago (but that's why you posted the schem, wasn't it):

have you thought on (a) bypassing the Q6 E-C with some sort of a f.e. 100uF cap - or (b) replacing Q6 with some sort of LED with a parallel cap and adjusting the 470R value - to allegedly improve the CCS noise performance?

Post removed.

Interesting. I didn't know about that.
(I usually throw a multilayer ceramic in parallel to a LED, out of a habit..)

Now, the example you posted got me thinking and here are some of my thoughts (purely theorizing, so these are likely to be completely wrong):

1) your example implies almost 6,5x lower current biasing of the CCS transistor, 160uA vs 1mA, which could mean that in general, this wouldn't work in a CCS set-up for higher currents (thus needing larger base current)

2) because LEDs want a higher "optimal" bias for low-noise, so they may be more appropriate for CCS's that work at higher currents.

Yes or no?

Post removed.

Post removed.

I see there was a slight misunderstanding: with 160uA I meant precisely what you explained - the "driving" current for the _base_ of the pass transistor in a CC source/sink.

IOW, I was writing from perspective of the CCS _base_ current, not the "sinking/sourcing" current. Clearly from that perspective, a 160uA is too low for a f.e. 50mA CCS, where a single red led, driven hard enough to "drive the pass transistor base", would be a cheaper and perhaps more efficient solution.

Of course, "experience" is IMHO the only way to get to the optimal solution in such cases.

Can you come back and review my post based on what I said here? I think we are on the same page, but there was some misunderstanding.

Post removed.

Political rhetoric:
"In a quest for a simple, yet satisfying solution, a DIYer must always search for the impossible!"

How about a hybrid? The "hard" bias should (at least in theory) take care of some LED self-noise and also significantly lower the impedance as seen with Q1-Q2 "ring". PSRR should (in theory) also be good.

The "instant ON" aspect may be also desirable in an opamp circuit.

Btw thanks for the links, I think I had them on my comp., but it never hurts!

Post removed.

Give the thread back? But wasn't it about pimpin' the DOA in the first place?

EDIT: and both CCS's in context of the DOA. Note: not tested.

(the ground symbol is at the negative rail because of the sim, I forgot to delete it)

Post removed.

What's all this post removed stuff all over my thread  .

tv said:

Theoretical nitpicking like it used to be here a couple of years ago (but that's why you posted the schem, wasn't it):

have you thought on (a) bypassing the Q6 E-C with some sort of a f.e. 100uF cap - or (b) replacing Q6 with some sort of LED with a parallel cap and adjusting the 470R value - to allegedly improve the CCS noise performance?

Click to expand...

The benefits far outweigh the noise which I can't seen to hear. Having built them in all those other configurations, I settled on this design.

Posted by: analag
« on: Today at 04:19:41 am » Insert Quote
What's all this post removed stuff all over my thread  .



Seems that members' posts have been removed from a number of threads... !?