Melcor Stereo Micpre

[CJ]

I scored a couple of Melcor cards off of evilbay some time ago and finally got them racked.
They were originally summing amps from a Surburban Sound board.
This presented some problems that I did not detect until I put it together.
The summing amp uses an inverted scheme, that is, the plus side of the opamp is grounded, not the minus side. So I had to reverse the phase on the output x-formers, no big deal. Otherwise, I would have had to either re-work the original board, or else make a brand new board.
I also had to put the gain control between the upper leg of the transformer and the minus input, as the plus side was already grounded via the pc board.
The summing amp cards have a 130 k feedback resistor instead of 20k or 47k, like the standard line and mic pre schematics floating around. More gain? You bet!
I find that there is a sweet spot as far as gain is concearned with these modules, and almost went as far as to fix the gain with a resistor, instead of using a pot.
You can really taylor the highend response with the cap on the feedback resistor. Some use 47 pf, others 220pf, etc.
Maybe a tone switch could be used here with different value caps?
I think I am gonna run it wide open with no cap, and limit the high end somewhere else, ie. Pultec eq. Better to have it and not need it, than need it and not have it. There is no oscillation, so what the heck.
Anyway, hewre are a few pics:

x
x
Thats a James Greenlee (remember him!) output transformer that I put in a mu can on the left.
Thats a relammed 80 percent Nickel on the right, which has one third the dcr but three times the inductance. I can hear no difference, but I have yet to see this project on the scope.
Kind of a sloppy layout (what else is new, cj!) , could have been made better with a custom PC Board, but I just wanted to get it done!

x
I am using an external pwr supply for low noise. The Jones strip is where the pwr wires are brought in. I roughed in the stock Zobel network, but noticed no difference with this particular set of input iron.

x
I took some windings off the stock bobbin to get the inductance down where it should be. This one clocked out at about 900 mH compared to abnout 350 mH for the stock x-former. Brass hardware, no varnish on the lams or coil, just some Kapton tape.

x
I kept the original lytics just to see how they sound. At this point, I see no need to change them. This circuit uses a capacitor output to the transformer. Some API circuits use direct coupled, some use a 47 ohm resistor across the cap, for what reason I do not know. Maybe a little dc in the output x-former for the funk?

x
Not a lot of bells and whistles, just an XLR in, 1/4 inch out, a phantom switch, and of course, no amplifier sounds good without the chicken heads!
x

x
Racked above a DIY Pultec and headphone amp. I figure why not wait til I get three or four projects done, then burn the Lazertran paper with as many decals as I can squeeze on there and label everything at once.

So how does it sound? Have not had time to do extesive listening tests, but my first impression is clean and quiet! A little too sterile for vocals, I would opt for my V76 for this purpose, but I bet it really kicks butt on instrument micing. I want to experiment with a pot that would go across the output cap instead of the 47 ohm resistor. This might give me a "funk" control, as it would vary the amount of dc going into the output x-former. What do you think?

Valley People micpre getting racked tomorrow!
Happy DIY! :grin:

 

[Flatpicker]

Too cool, CJ! Thanks for sharing. Can't wait to learn how the extensive listening test turn out.

 

[daArry]

nice1 CJ :thumb: I was thinking of using these melcors on the end of a summin box - u rek this inverting config would be a go'er? I like the idea of a tone and funk control

 

[CJ]

Sounds good, you could do a new board to run it anyway you want also.
I do not know if there is a sound difference between inverting and non-inverting. I highly doubt it.
Those are some custom Reichenbach input x-formers by the way. Do not know much about them but the ratio-1:7, or 1:14. Running them 1: 7 right now.

 

[CJ]

OK, i said this box didn't cut it for vocals, but I finally got around to phasing the output transformer correctly and now it rocks for vocals. Almost (gulp) as good as my V76, and about 1/10 the hassle. But get this, the V76 is quitier than the Melcor!
You can hear the white noise from the opamps with the gain turned way up (using 130k feedback resistor, so more gain than stock), whereas the EF804's are way down there on the noise floor. This is probably why they trim the high end on the opamps with those caps across the feedback resistors.
I would say, however, that the V76 can make a crappy mic like a beta 57 sound better than the Melcor, but the Melcor sounds fantastic with the condenser mic.
Now if I could just get this Valley People thing to quit oscillating.
:shock:

 

[kent]

Cool, CJ!

I've been waiting for this project to be tried since I've got some of those API OT's and the Melcor opamp layout is floating around. When the snow flies I'll be back in the DIY with this one!

BTW - damn right about the chickenheads!

cheers,
kent

 

[Flatpicker]

[quote author="cjenrick"]...I finally got around to phasing the output transformer correctly...[/quote]What was wrong with it originally?

 

[CJ]

the boards were used as a summing amp and the opamps inverted the signal, so i inverted the x-formers to get it back to normal.
the out of phase sound lacked bottom end.
cj

 

[Flatpicker]

Oh. I thought you were speaking of yet another problem.

Wonder if my diy Melcors will sound that good? I can only hope!

 

[fum]

Kent, ya mean something like this

http://www.shinybox.com/diy/325.jpg

just finished these, still have to test them :grin:

ju

 

[owel]

Let us know... I have a bunch of Melcor PCBs I haven't populated yet.

 

[kent]

Cool, fum!

 

[Bauman]

Cool CJ!
I got some lams from Magmet to try on the API output also. Superperm 49 and 80. So do you take off some of the winding... uhmmm...

 

[CJ]

No, you don't have too. But the varnish added diameter to my winding, so when I went to rewind, I could not get the same amout of turns on there.
I think the regular stock steel sounds fine. Try the 49 first. 80 is cool, but can distort m more than M6.
Personally, I would not waste the lams on that transformer. Use them for something else, like a 600:600 or something. The Litz winding makes the API outpout pretty sterile and transparent, so changing lams won't effect it as much as an interwoven pri and sec.
cj

 

[Dan Kennedy]

I'm a little surprised by the noise difference.

You might find that dropping the feedback resistors absolute values by a factor of 10 might help. 130K is awful high. But keep the ratios...

In a mic preamp the noise of the feedback resistors adds a real amount, in a line level stage it's pretty much inconsequential.

 

[CJ]

Thanks Dan! I will give it a try.
cj :grin: