Poor Man's Tube Mixer

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
16,152
Location
Norfolk - UK
It has long been an ambition of mine to design an affordable tube mixer. I have been working on this recently and I think I have come up with something that people might find interesting. I have tried as far as possible to cut out everything unnecessary and you use standard parts wherever possible. Even so you can pare down the cost even more by omitting some transformers or even whole functional areas. It uses 14 tubes, fits in an off the shelf box and the total parts cost should be about $2200 for the full set of options. For this you get:

4 mic pres with transformer balanced inputs, HPF, echo and foldback sends, pan and rotary fader plus unbalanced insert or direct out
16 input passive mixer (2 x 25way D ) with tube gain make up and rotary fader
2 transformer balanced direct inputs one each to the left and right buses (for expansion)
Echo send master, foldback send master both transformer balanced
Bus master with two channels of REDD PLUS passive EQ, master fader and transformer balanced outputs
2 transformer balanced FX returns
PFL on all mic channels (mono), passive mixer (stereo PFL) and FX returns (mono)
Monitor section with two VU meters, transformer balanced monitor out, monitor level, PFL over ride and 2TK playback input
External power supply

I would make all the PCBs available at cost in the Black Market

What do you think?

Cheers

Ian
 
I think that sounds tremendously exciting! Also seems a good fit with the current trend towards high end gear for smaller / project studios, as with the new SSL mini-mixer.

Can’t wait to learn more!
 
john12ax7 said:
What would the metal work be like?

I am planning on using this off the shelf enclosure by Metcase with customise front and rear panels. It has built in carry handles, plenty of ventilation and even optional PCB guides. And it appears to be available on both side of the pond.

Cheers

Ian

 

Attachments

  • M6436325.pdf
    400.3 KB · Views: 142
Thanks for the encouragement. I have ordered an enclosure to tinker with. In the meantime here is my first shot at a block diagram. I have had to make a number of simplifications to keep costs down. For example, all the gain blocks are mu follower stages just like the ones in the Classic Solo mic pre. This needs one tube per gain block rather then one and a half as in the twin line amp design so this reduces tube count by one third. Mic gain is fixed at about 48dB which should be plenty for most applications. There's a 20dB pad for higher level signals and as this stage can output at least +30dB before serious distortion the mic input should be able to handle up to +2dBu inputs with the 20dB pad in. There's a bigger pad for line inputs, probably 40dB which will give a nominal +8dBu output for 0dBu line inputs.

There are 16 transformers in the design but only six of them are essential if you are happy to have unbalanced outputs and unbalanced FX returns and direct inputs. To keep cost down I plan to use OEP types. They make an excellent 1:10 mic input transformer (A187A15C) for only £40 so the minimum transformer cost is £240 if you have all four  mic inputs and the 16 input stereo passive mixer. For the 10K:10K I plan to use an excellent small OEP transformer (Z21807C) that costs no more than £15 from Canford.

I have also tried to make it easy to leave sections out so perhaps the simplest thing you could build would be four tube mic pres in a box. Just four tubes and four transformers required. For a couple more tubes you could add a stereo bus and for a couple more and a couple of transformers you could add the 16 input passive mixer (no REDD EQ , no FX sends/returns, no PFL and no monitor).

Cheers

Ian
 

Attachments

  • MiniMixer.pdf
    70.1 KB · Views: 174
Hello

Look like a great idea !!!

Few things coming in mind
-can't the passive summing section directly go to the main L/R bus ? if you need a makeup I understand, it's mostly "the fader" that look redundant to me, acting like group fader which seem (to me) unnecessary for this concept.
-more monitor input (at least one) should not cost that much (just connector and more way on the switch) and always useful...
-maybe PF insert on master ?

Is monitor section tube or IC ?

Best
Zam
 
zamproject said:
Hello

Look like a great idea !!!

Few things coming in mind
-can't the passive summing section directly go to the main L/R bus ? if you need a makeup I understand, it's mostly "the fader" that look redundant to me, acting like group fader which seem (to me) unnecessary for this concept.
There is no reason the pot cannot be replaced by a couple of presets just to set the overall make up gain right. API's The Box has a fader per input -I have no room for that so I thought one overall fader would do
-more monitor input (at least one) should not cost that much (just connector and more way on the switch) and always useful...
Do you mean you want a couple of external inputs to the monitor - as you say not too hard/expensive
-maybe PF insert on master ?
[/quote
It would have to be unbalanced. I am not 100% sure what the level will be here but I would probably need to include at trimpot at the bus amp input to set this. But otherwise quite possible.
Is monitor section tube or IC ?
At the moment everything is tube, even the PFL bus amps. The rest of the monitor is passive. I guess I could replace the PFL bus amps with ICs. They would have to run off the 12V heater supply because it does not make sense to add +/- 15V rails just for a couple of chips.


Cheers

Ian
 
Hi Ian

Yes, fader per input make more sens but if you don't have room... and then it's the extending function race start... why no aux , why no EQ etc... KISS...I'm not even sure PFL is needed at passive summing as it send 16ch at a time (I don't see right now practical use of this)

yes I'm talking about external input for monitor section, but second output can be nice too (just a switch/relay after pot) so one can flip to headphone or second monitor or headphone splitter/preamp, I imagine this kind of concept mainly used in a small one room studio, so monitor to headphone will be used when tracking.

I personally don't have issue with unbalanced insert, and preferable than no insert.

I just ask about IC regarding cost, I mean I guess no one will complain if there is IC in pfl/monitoring section of a tube preamp/mixer in the 2K range you evaluate, but now if it change other practical things in design (like power rails) maybe there is no cost gain, you know better.

Cheers
Zam
 
I'd say your undercutting yourself with the 'poor mans' reference ,  I dont think your doing it justice at all  with a title like that.
I worked in hifi sales for a while , they had it drilled into us never to refer to our stock as cheap  in front of customers.
 
zamproject said:
Hi Ian

Yes, fader per input make more sens but if you don't have room... and then it's the extending function race start... why no aux , why no EQ etc... KISS...I'm not even sure PFL is needed at passive summing as it send 16ch at a time (I don't see right now practical use of this)
Exactly because as soon as you have a fader you need pan so that is 32 knobs for starters - the DAW already has these controls
yes I'm talking about external input for monitor section, but second output can be nice too (just a switch/relay after pot) so one can flip to headphone or second monitor or headphone splitter/preamp, I imagine this kind of concept mainly used in a small one room studio, so monitor to headphone will be used when tracking.
These are simple to provide but I am not sure if there will be space on the rear panel for all the connectors required. For example I thought of adding a third DB25 to give 24 inputs to the passive mixer. I guess there will need to be some trade off here.
I personally don't have issue with unbalanced insert, and preferable than no insert.
OK
I just ask about IC regarding cost, I mean I guess no one will complain if there is IC in pfl/monitoring section of a tube preamp/mixer in the 2K range you evaluate, but now if it change other practical things in design (like power rails) maybe there is no cost gain, you know better.
Yes, it is all the consequences, extra power rails, extra power supplies, extra cable, more pins on power connector.

Cheers

ian
 
Tubetec said:
I'd say your undercutting yourself with the 'poor mans' reference ,  I dont think your doing it justice at all  with a title like that.
I worked in hifi sales for a while , they had it drilled into us never to refer to our stock as cheap  in front of customers.

This is not something I plan to sell - it is a DIY project and as such very cost sensitive.

If I did decide to build one to sell I would definitely call it something quite different and sell it for a lot more  ;D

Cheers

Ian
 
This looks like a fantastic project.  I think keeping it simple is a great idea and if folks want to modify it for their needs they can.  If you try to please everyone, this might not ever see the light of day.  I've seen several promising mixer designs never get off the ground on this site because of that.
 
I could see folks possibly wanting different features so, by generously providing the nuts and bolts of the bits to adapt, they can tailor it.
If it were for me,  I'd opt for your REDD EQ on the channels with direct outs and gladly give up eq on main outs, pfl, extra 16 channel inputs etc if I had to.

Each to each and all the better for it  :)
 
Winston O'Boogie said:
I could see folks possibly wanting different features so, by generously providing the nuts and bolts of the bits to adapt, they can tailor it.
If it were for me,  I'd opt for your REDD EQ on the channels with direct outs and gladly give up eq on main outs, pfl, extra 16 channel inputs etc if I had to.

Each to each and all the better for it  :)
Funny you should mention that because it occurred to me that all the gain blocks are identical apart from some might need a trimpot at the input. So they could all be identical on the PCB each with an input and output Molex socket and trimpot which you could short out if you don't need it. That way you could wire them up any way you liked. All the customisation would be on the front panel PCB.

Cheers

Ian
 
Hi Ian,

Something like this has been in the back of my mind for years.
Having available PCBs would make it much more practicable to achieve.

I would choose to mount it in a Rack Case though.

 
ruffrecords said:
Funny you should mention that because it occurred to me that all the gain blocks are identical apart from some might need a trimpot at the input...

Well that makes it a whole lot easier for you and anyone doing their own thing with it.
So does each input comprise of two independent but identical stages?  I'm assuming the trimpot is a level pre-set that can also determine the 2K5 source impedance for your eq but?
Edit:  scratch that last bit, you have this on your eq boards already

 
Back
Top