mic pre #7

this is about a tube mic pre with a triode/pentode switchable ef86 input and a srpp output stage using the ecc82.

uploaded the pictures from the prototype again



inside the prototype:


schematics:
the gain switch "includes" the -20 pad, using a 4x3 pole switch.


psu:
bear in mind that the lm317 has plenty of work burning about 7 volts. thats ok for one channel, for a dual preamp i would suggest to adapt the heaters for 12 volt ( 2x ef86 in series, ecc82 work with both 6,3 or 12 volts)

Interesting how does it sound?, iIno thats hard to discribe..anything to compare it to perhaps ? thanks

Gary O.

Oh! I like this one alot. Simplistic!

You do realize that plate-to-cathode feedback is positive feedback, right?

[quote author="NewYorkDave"]You do realize that plate-to-cathode feedback is positive feedback, right?[/quote]

no, thanks for pointing this out. i didnt even think about that.

That circuit looks just like one in the Mullard Tube Circuits book, except for the triode/pentode switch.

Prototype looks very cool! :thumb: Lunch box?

[quote author="skipwave"]Prototype looks very cool! :thumb: Lunch box?[/quote]

its a cash box, similar to the one i used for my gssl.
gary o, i´ll have to make some soundsamples, but it´ll take some time.the unit isnt finished yet.

nyd(and others): what are the drawbacks when using positive feedback? i was testing a lot today, but couldnt find a real sonic problem about it. it sounds different when i take the feedback path from the output stage, but i wouldnt say that it sounds worse.

edit...see first post

Hey, is the transformer secondary really unloaded, like in your schematic? Or did you use a resistor shunting in the end?

Would be great to hear more about the positive feedback also...

hey,

no, there´s a cap and a 1meg before the grid for a switching di circuit like in the g9.
also, the 16µ 380 volt in the upper left is drawn wrong. shit, and also the 250k aside is wrong... and one 10 k is missing ...

sorry, i´ll redraw the schematic soon.

update first post

My limited understanding of feedback (negative) is that it reduces some of the voltage swing in a gain stage and makes things more predictible and stable. I think that positive feedback does the opposite.

There was a very good little paper posted maybe a year ago about understanding opamps. Negative feedback was right at the begining. I see that I didn't save it and I can't find it in the search...

Anyone elso know what I'm talking about? It is about opamps, but it really helped me understand feedback.

Joel

Umm. I'm having a hard time seeing positive feedback.
The plate and cathode are OUT of phase with each other.
Looks like negetive to me. Feedback can be applied from the output
of odd stages 1, 3, 5, ect. I'm I missing something? Must be, It's been one of them days.


RonL

Man, I would love for someone to explain this one to me... From my limited understanding of things, I see positive feedback going on, which means that this thing shouldn't be working... But then when I start to analyze it, it seems like it DOES work! :?

AFAIK, you need cathode feedback to be of the same polarity as the input for it to be negative - that's an even number of inverting stages...

Paul? Brad? Someone?

Peace,
Al. <-confused as hell

Back to Tubes 101, all of yuz

The cathode is a noninverting input. What happens when you connect an amplifier output back to its noninverting input?

Do you doubt that it's a noninverting input? Consider this: when the voltage at the cathode rises, relative to the grid, what happens? The tube biases toward "off" and the plate voltage rises. So a positive excursion at the cathode equals a positive excursion at the plate.

The only reason this preamp doesn't explode in oscillation is because the feedback ratio is so small. But then again, the gain of a single pentode stage is large so it could be sitting right on the ragged edge. Square wave response would be interesting to see...

In this circuit, in the most benign case, the feedback is doing little or nothing.

Always use positive side of sine wave to analyze, one step at a time.
Let's freeze the input signal at it's positive crest and hold it there.

the input signal drives the grid positive with respect to cathode
this lowers the resistance of tube compared to the plate resistor.
this means a voltage divider has changed
the voltage divider lower resistor (tube resistance) has shrunk.
therefore the positive signal on the grid, turns into a negative signal coming off the plate coupling cap.
this negative signal is applied to the cathode via the feedback resistors.
this makes the cathode more negative with respect to the positive grid signal. (this should be enough right there, for most)

So a positive grid signal turns into a negative plate signal, which is applied to the cathode. Since the cathode and grid are, in effect, moving further apart, the voltage between them is increasing.

This means the input signal has become bigger because of the feedback.

What kind of feedback is this?
Positive or negative?
.
.
.
Next question: Why does this circuit not oscillate?

NY Dave
Your right!. Like I said, It's been one of them days. It will remain stable
as long as the feedback signal doesn't exceed the signal at the cathode.

RonL

Would there be any benefits in a push-pull amp to have the cathode from one feedback to the plate of the other and vice-versa (like two el84's, say)? Does that then work as negative feedback? I imagine it probably wouldn't be that linear though...

Joel

Off topic, but... I was wrong about the Mullard circuit. It used an EF86 input stage coupled to a ECC83, with one triode as a gain stage and the second as a cathode follower. I just mistook it for a WCF. It was designed for mixing multiple inputs. I think Bogen used this circuit in some of there mixers.

EDIT: Things only make sense to me after re-reading, and re-reading.

to me, it sounds like the tube is working harder. i bet i have more resonant peaks with feedback applied than without -coloured- it doesn´t oscillate; though..
[quote author="NewYorkDave"]it could be sitting right on the ragged edge.[/quote]

i didnt test square wave- i´ll do that.