a little XLR microphone mixer idea

[Mario44]

Finally I got some time to start building a simple mixer for my band. I made a block diagram and wondering if this could work.
The mic is shure super55 and the general purpose is to split vocals into power amp and recording desk while aux will be conected only to power amp(for e.g to listen something we record).
Tell me what ya think 'bout it  
Schematics coming soon.

[AdamM]

I think you'd be able to buy a little Behringer mixer that would do that (and more) for less than you'd spend on buying the parts to build one.

Sorry - probably not the comment you wanted

[Mario44]

Yeah, maybe but I want to build it. It's fun and I can learn something too. Every reason for making effects and the whole rest is a good reason 

[PRR]

Agree that a Beeringer would do the job, fast, and do other things beside the specific setup shown.

Unless this power-amp is on the far side of the stadium, you can probably drive it un-balanced. Put hot on one pin, ground the other.

Your so-called Conversion to Balanced is really a mixer, no? It mixes the mike-mix with the CD stereo.

The microphones are not just "XLR to unbalanced", they need considerable gain, and adjustable. These are Mike Preamps. (And again, 97% of the time a hot dynamic can be run unbalanced across the typical stage.)

I would advise you to steal study the design of some low-price commercial mixer. Understand the bits and use the parts you need in the connection you need.

[Mario44]

Now I see that I made the diagram too complicated and conversion are not really needed. Well, I have never had to do with XLR's and they seem to be tricky for me. 
I remade the diagram and now I guess it's better and even simplier than before. Just wondering if there should be conversion on the output or just a simple line out will be enough. The manufacturer says the input of the active coulmn will take balanced mic or line-level signal from the mixing console or other signal source(huh?). What do you think about the diagram now?

[PRR]

Do you really want just one microphone going to the recorder? (I assume that will be *your* mike? <g>)

You have two "mixes".

All mikes to recorder
All mikes plus CD to PA

Mikes need preamps, optional EQ.



Line to PA *may* need complication to deliver large clean signal over long line; or maybe not.

Mixers can generally be passive resistor-piles, but since you may need make-up gain or line-drive you should budget an opamp apiece.

[Mario44]

Well, it would be enough for me to have 1 mike on recording out, but Your option looks better and it's more versatile.
Thanks for Your input!  

[Seljer]

I'll second the Behringer/Wharfedale mixer comments. I too was contemplating building roughly the same thing as you.

When I'd tallied up how much I'd pay for all the jacks and pots and power supply and components to make preamps I realized that for 50€ I could get a ready made mixer that did waaaaaay more. So I've now got 4 balanced inputs with preamps and 48V phantom power, two stereo inputs, 3 band eq on each and even an AUX send on each channel and an FX send.

I'd only reconsider building if this was some other very specific purpose (eg: battery powered or needs to be very small).

[Transmogrifox]

I have long been a fan of Rane's products.  One nice thing about Rane is they publish schematics of their products, so you can learn a lot from how the pros do it:
http://www.rane.com/pdf/mlm42ssch.pdf

Above is a schematic from their 4-channel mixer.  Take a look and you might get some ideas.  If you're really ambitious you could build the whole dang thing and have a really nice mic preamp/mixer product -- and DIY

In either case there will be useful circuit tidbits to consider for the purpose of learning how to handle XLR balanced and unbalanced signals.

The main thing that will help you a lot is the INA163 IC that they use for low noise amplification of the differential inputs.  From the Rane schematic you get a phantom power application example to consider also, if in the future you get a mic needing phantom power.  Note "Z1" (Z2, Z9, Z11...) are likely some kind of EMI suppression package.  At audio frequencies it looks like a short circuit through the circle, and a capacitor to ground.  You could omit this or get a differential EMI suppression capacitor like this:
http://www.johansondielectrics.com/x2y-products/x2y-for-emi-filtering.html

I agree about the cost if that were a major factor in choosing a DIY project. 

If you're building for education as well as a sense of accomplishment (and sometimes just so you can fully understand the equipment you're using), then there is a lot of value in creating something with great effort which could otherwise be obtained more cheaply and with less effort, so I see where you're coming from.
 
A case for the economics of DIY is if you build something high-end like the MLM42S mixer, then you might save on building your own over buying high-end new.   If you really know what you're doing, then you could make something a grade above most commercial products because you can trim and fine-tune.  None of the design decisions need to be made based on part tolerances and manufacturing variability -- you simply hand-select and manually trim all component tolerances to perfection.

If all a person wanted was a mixer to do the job, you just can't beat the cheap "good enough for rock 'n' roll" equipment prices even with parts cost -- they buy parts in quantities of thousands.

[Mario44]

Well, after a long break I came back to my problem. Here is a balanced mic preamp. I build it on my breadboard and it's working. The problem is it's my first schematic and just wondering if that's all correct. Here is picture.

[PRR]

R2 really should be equal to R6 for best results.

Gain of 100 is likely too much gain for stage work.

If so, you want a gain control at some point before where the signal overloads. This topology has no easy way to do that.

It is also hissier than it needs to be, though I have used such a plan in concert and nobody threw us out.

[Mario44]

Thanks for correction. Soon I'll make some tests.

[Mario44]

I tested this preamp with 2 different dynamic mics and result was the same. Gain of 100 or even more was good amount of gain. Lower gain setting were unusable. Preamp was connected into a interface, then PC and finally signal showed up in cubase program. However it sounded just well and preamp did it's task. Do you think something is wrong that I have to put so much gain on it?

[mth5044]

What kind of gain? If we are talking dB gain, then yes, that seems excessive. Voltage gain? Not really. A mic preamp based on the INA217 from TI can go up to 2000V/V gain (66dB). 100V/V gain would be somewhere right in the middle and, I would think, normal.

[PRR]

A close-talked dynamic gives 5mV or less.

Line level is maybe 500mV.

Yes, gain of 1:100 (40dB) seems utterly reasonable.

And that is what your schematic shows. (Actually 200 if you changed R2, because of the push-pull output.)

[Mario44]

Ok. Now I understand the gain problem(actually no problem with it) but still wondering about input impedances and filters. I want to make this preamp as transparent as possible. It cuts some low end. I'm thinking of raising value of C1 and C5 up to 100u and R1, R4 to 10k. Will those values create good input impedances for microphones?
Btw opamps are TL072(can those allso affect the transparency more or less than 4558 or 5532?)

[PRR]

> I want to make this preamp as transparent as possible.

Well, while this preamp path "works", it isn't the path to "real transparent".

First: R1 and R4 are not needed in this topology. But....

> mic is shure super55

Some people *prefer* the sound of Shures with heavy loading.

The primary fault with this topology, IMHO, is the high hiss level. The microphone alone hisses like a 300 Ohm resistor. In most rooms, this self-hiss is less than the ambient room noise (they design it that way), so is acceptable. However your diff-input adds 1K+1K+ 2K of additional resistance before the amplifier. This is an 8dB noise-factor, which is rather high.

Next: the TL072 (and 4558) input devices have rather high hiss-voltage, easily 20dB higher than a dynamic microphone. (5532 is somewhat better, though still not an ideal mike amp.)

Specifically-- I once had to set up a band PA which used this exact input affair. When I had levels set, I found the system hiss obnoxious, much higher than many other low-cost PA systems. True, once we let the mob into the room and the music started, the hiss was hardly audible, nobody complained, and when I mentioned the hiss to the manager he said "is OK".

There ARE better paths. Google "$10 mike preamp" for an EXCELLENT plan based on a dedicated chip with special transistors, 3 or 4 specialized opamps, and a resistor network inside. "Greene preamp" or "transformerless mike preamp" may find descriptions of the basic topology. And conversely "transformer mike preamp" finds an older less-cheap path, 80 years old and still useful.

[PRR]

Oh: while you have this a-live.....

For the Shure stage-mikes:

Scrap R1 R4. Change R3 R5 to 330 ohms. Yes, C1 C5 could be a lot bigger, 33uFd to 47uFd. Use '5532. That may be very fine for many situations. Frequency response will be 20Hz-15KHz. Hiss not low enuff for far-micing harpsichord to 24-bit digital master, but probably "low" for stage vocal work.

[Mario44]

Thats weird. You say that this preamp should produce some noise, but I didn't notice even the smallest noise in the setup that I wrote before. I recorded a few tracks of snare, tom and drum kick. Everything sounded well, but the drum kick was lacking low end, so basicly useless. The interface is tascam us16x08.
Tomorrow I'll go hunt some 5532's. Also I'll test second input topology that You suggested and compare. Oh and btw the mic used for recording samples was sm57 and Jts nx2.

[Hatredman]

Did you test it on a PA System or just using you computer speakers?