hi, how about this

[jeffrey_burr]

Hello all, first I'd like to say what a cool resource this site is. I'm not really an engineer, just another control-freak musician. I've learned a great deal here, and now I'm putting together a cozy little DAW studio setup primarily for recording my jazz music. I haven't done alot of electronics DIY aside from modifying existing stuff, but I've spent a couple years reading learning etc and it's time to get my feet wet now.

I found this link a long time ago, and I was thinking about building a channel or two, provided I can afford the transformers.
http://www7.taosnet.com/f10/micpre.html

Very simple....too simple? There's a link to an interesting article by Alan Kimmel, which you all have probably seen, especially those of you interested in hybrid designs. Well, anyone have any thoughts about this one? Sorry if this one has been discussed-to-death already, a search for it came up empty.

As an aside, I'm also considering Rod Elliot's p66 with the much-lauded Bloak opamp.

Thanks everyone for your input, and the general atmosphere of sharing here!

 

[capnspoony]

I just finished up an eq using that same input transformer. Indeed a bit pricey but it sounds great. Its very accurate.

-richie

 

[Guest]

It is Ok if after it you have high impedance stage, paralleled triodes don't help much. Also, I feel like IRF7xx may be too slow in such application. Though, I like how "anti-pop" switch on is implemented.

 

[jeffrey_burr]

Wavebourn, another Bay Arean, I see! gosh that's got a funny ring to it...well I'm over in Oakland anyway.

I was wondering about the parallel thing too, it would definitely be more fun to use one envelope for two channels. It's actually been a while since I took a close look at this, but I recall there was a problem with the capacitative load at the gate of the MOSFET, and the paralleled triodes were addressing this.

I'm not as good at solid-state as I am with tubes, I'm not sure what you mean by "too slow". Is there another device that seems like a better candidate for the MOSFET, or what characteristics should I be looking for? I'll have another look at the article and an IRF712 datasheet....

 

[Guest]

I am not sure it is too slow, I just feel like it may be too slow because of very high capacitances.
I'd rather use PNP transistor for a current source and an output stage instead of this FET arrangement. But you'd better try, may be I am wrong.

(A lot of inventions happened because nobody told that it will not work)

 

[peterc]

I built a mu stage mic pre, but without the input transformer & just used 2 channels of mu stage, one to amplify the + side of the signal, one to amplify the - side of the signal.

I know, my design skills are quite crappy....

It was noisy, but I have to say that this was one of the nicest mic pres I have heard. Clear midrange, extended top end, very sweet. I have used the mu-stage preamp before in a line level application & also liked the sound there as well.

Meant to rebuild this pre with an input transformer but it is still on the backburner, about no17 in line.

My 2c

Peter

 

[pstamler]

A major weakness of this design is one that's common to every design with its level control between the transformer and the active device. At a bit less than 6dB attenuation, the source impedance feeding the tube grid rises to about 66k (ideally it should be about 13.6k with this transformer -- 15k loaded by 150k. This will add to the circuit's Johnson noise, boosting it by about 6.9dB.

Peace,
Paul

 

[shabtek]

I have no experience with this mu-stage, for a starter the 1-bottle is simple enough and people seem to like it....One of my first projects was an Elliot simple mic pre follwed by the Elliot zen preamp with 1:1 xfmrs in and out. It sounds pretty good all the way around and took a weekend to build

 

[jeffrey_burr]

Well, here's another thing. I'm not sure why I've never noticed it, but down near the bottom of the mu-stage article (http://www7.taosnet.com/f10/mustage.html) there is a differential in/out version illustrated (figure 7). Seems like this should be of interest to folks around here! Especially those of us who don't have to brush aside great heaps of input transformers in order to find a pencil.

The major mystery for me about that drawing is, why the VKK supply? -100V??

Okay, after a little thought, I feel I've got a handle on it, but still it seems somewhat excessive. Large enough cathode resistors will bring the bias to a reasonable point, the noise is canceled by the differential balance (?), and maybe there's some headroom-oriented advantage. Actually since the current is (almost) constant the voltage drop across Rk is constant as well, no?

Yah, I don't get it. What would be wrong with using the ground, like everybody else? Or if it's really necessary to have the action so far from the bumpers, why not apply a fixed bias to the grid, like the pentode CF on top?