8 in, 4 out unbalanced tube line mixer

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

johnnytoobad

Active member
Joined
Mar 24, 2010
Messages
29
Location
Marseille
Hi!
I am tube amp builder and repairman, sound technician and musician.
I usually build and repair guitar amp and few HIFi equipment.
I am now designing the mixer I need.
I need a ten unbalanced line inputs and 4 outputs, for playing live with drum machine, sampler and synths but also for mixing what gets out of my daw.
So I went with this design...
I don’t remember from where I found the mixing/summing design but I took it somewhere and modified it a little to make it sound how I like, but never used it for more than 4 inputs.
Here it is for your advices and criticism.
Thank you
Edit-1: tube mixer is a 12ax7 and not au7, and cap badly added before preamp cathode follower...
 

Attachments

  • 01B806A2-C551-49F1-9B3F-4D99213A621F.jpeg
    01B806A2-C551-49F1-9B3F-4D99213A621F.jpeg
    577.6 KB · Views: 69
Are these diodes really necessary? I understand the ones in the inputs can do no harm, but I'm concerned about the ones in the summing amp.
Actually, the 100k resistors that feed the faders are not necessary.
I would think the optimum value for the faders would be 470k-1 Meg.
 
hope it will be a smashing tool for your music!

wouldnt the output faders need to be followed by a buffer? the output impedance would vari up to about 25Kohm...
 
It is a virtual ground summer amplifier so it shouldn’t need buffers, I am not even sure that the cathode follower is needed before the faders... i’ll see while prototyping, if it works as intended or if I need some more buffers or  less or more gain....
 
Ruffrecords, i don’t know the math to calculate the faders (if you or anyone else knows, please tell me!), so i’ll have to try to find the right value
 
johnnytoobad said:
Ruffrecords, i don’t know the math to calculate the faders (if you or anyone else knows, please tell me!), so i’ll have to try to find the right value
If it is unbalanced line level inputs then why not go straight into the pots. You do not really need any gain. The two triodes on each input will only add noise and distortion. So how about going straight into 10K log pots.

Cheers

Ian
 
johnnytoobad said:
i don’t know the math to calculate the faders (if you or anyone else knows, please tell me!), so i’ll have to try to find the right value
  • The faders resistance must be  larger than the cathode resistor of the cath-follower in the input stage, for max headroom. Since there are 4 faders on each input stage, the value must be at least 4 times 100k.
  • The half-travel impedance at the wiper (about 1/4th of nominal value) must be lower than the mix resistors (470k). That puts a max limit of 1.9Megohm.
Hence the limited choice of 470k or 1Meg.
It could work with 220k or 2.2Meg, but with degraded performance.
Changing the 100k cathode res in the input stage to 68k or even 47k would help with headroom, and maybe allow using 220 (250)k pots. You would need to check the wattage of this resistor.
 
abbey road d enfer said:
  • The faders resistance must be  larger than the cathode resistor of the cath-follower in the input stage, for max headroom. Since there are 4 faders on each input stage, the value must be at least 4 times 100k.
  • The half-travel impedance at the wiper (about 1/4th of nominal value) must be lower than the mix resistors (470k). That puts a max limit of 1.9Megohm.
Hence the limited choice of 470k or 1Meg.
It could work with 220k or 2.2Meg, but with degraded performance.

I was going to try for 500k. So I was guessing not so bad, thank you for the explanation.
 
ruffrecords said:
If it is unbalanced line level inputs then why not go straight into the pots. You do not really need any gain. The two triodes on each input will only add noise and distortion. So how about going straight into 10K log pots.
I will probably lower the gain of the first triode (100k plate and 1,5k cathode), but I need to be able to add distortion for "in the red" dirty mixes if needed to add harmonics, compression and thickness (or sickness...).
 
abbey road d enfer said:
Changing the 100k cathode res in the input stage to 68k or even 47k would help with headroom, and maybe allow using 220 (250)k pots. You would need to check the wattage of this resistor.

I would fear that with a 200V supply the 47k cathode resistor  would  make the 12ax7 over dissipate...
 
johnnytoobad said:
It is a virtual ground summer amplifier so it shouldn’t need buffers, I am not even sure that the cathode follower is needed before the faders... i’ll see while prototyping, if it works as intended or if I need some more buffers or  less or more gain....

Hi.

i was refering to the fader at the output.
 
5v333 said:
Hi.

i was refering to the fader at the output.

The summing amplifier being  "low Z” output, I am not really sure it needs a buffer...
But I don’t really know. I’ll when building it.
Do you think that the summing amp won’t be able to drive a line input?
For sure I know that it can drive a HiZ guitar input, I already used it for that purpose.
 
in the schematic there is a fader after the summing amp.

if that fader is set to anything else than max, the output will not present the low impedance of the summing amp.

if the fader is linear for ex, at 50%, the output impedance will be around 25K. this is similar to a guitar signal, and will have large effects with cable capacitance etc.
 
5v333 said:
in the schematic there is a fader after the summing amp.

if that fader is set to anything else than max, the output will not present the low impedance of the summing amp.

if the fader is linear for ex, at 50%, the output impedance will be around 25K. this is similar to a guitar signal, and will have large effects with cable capacitance etc.
You are right, i didn’t think about it.
Anyway i’ll Probably add another stage with a parametric mid, followed by baxandall, then a cathode follower.
I probably won’t go with a fader but à pad for constant impedance at the output.
Thank you for pointing it!
 
johnnytoobad said:
Anyway i’ll Probably add another stage with a parametric mid, followed by baxandall, then a cathode follower.
I probably won’t go with a fader but à pad for constant impedance at the output.
If you add this after the summing stage, you should insert the pot somewhere in between. Constant-Z pads are difficult to set-up; that's the reason why most modern equipment is bridging (output Z<<<load).
 
I modified the input stage for less gain, and smaller Z out, using a 6CG7.
Let me know what you think about it...
 

Attachments

  • 069A622D-EC00-4FBD-96FD-A3CAC46B60C1.jpeg
    069A622D-EC00-4FBD-96FD-A3CAC46B60C1.jpeg
    1.8 MB · Views: 21
abbey road d enfer said:
You'll have extra headroom, but not much. Do you really want those 100k pots?
I can send small signal, and I got a gain of 20 at maximum setting.
I need the send faders to choose how much I want to send...
How would you do?
 
Back
Top