tube mixer?

[caress]

is this a possibility?
a tube mixer based around the first stage of the subcaster using a 6111 submini...  just a rough idea right now but i thought it might be best to stick with simplicity...

[caress]

oh i thought i should add that in this configuration, it would not be used for boosting gain, simply summing the signals.  sort of just using the 6111 instead of a dual opamp...

[frequencycentral]

Yeah, no problem. Google search Valvewizard, I thinks there's something there. if you can't find anything I'll sort it out when i get home - limited web access at work!

[mth5044]

Oh you people and your tubes. Looks like a cool idea, but what does it have over using an op amp? Just because, or is there a better reason?

[DougH]

why??

[frequencycentral]

Oh you people and your tubes. Looks like a cool idea, but what does it have over using an op amp? Just because, or is there a better reason?

An opamp is a dozen or more transistors. For me, the less parts the sound has to go though the better. And also, just because.

Caress, I can't access your schematic here at work, but i have a few ideas which i will post later today.

[DougH]

It'll make a nice heater anyway, I s'pose.

[drewl]

Oh you people and your tubes. Looks like a cool idea, but what does it have over using an op amp? Just because, or is there a better reason?

Better sound quality?

[DougH]

Not likely. But try it and let us know how it turns out.

[frequencycentral]

I'm having trouble working out your schematic - probably due to me being dumb.

I think this might work for what you want. I haven't tried it, so you may need to experiment.

Each triode acts as an inverting buffer, so the inputs vs output is non inverting. Using the resistor values given should give unity gain.



From experimenting I have found that low voltage tube circuits 'round off the tops' of waveforms - if you get the plate voltage up to 30 volts you'll have a result. Here's some test results using a trianglular wave input at various plate voltages:



I used a MAX1044 to get the plate voltage higher, a bit naughty as I was running 12 volts into it (its maximum is 10.5 volts). The LT1054S would be a better bet.

[caress]

why??

i'm just interested in trying something out really.  i plan on using it in a digital circuit, so i thought it might add some "warmth"... maybe it will just heat up the box a bit...   it's worth a try.



rick --
the way i set up the tube symbol is like an opamp...sort of.  reading counterclockwise from the bottom left pin are pins 1-8.
i wanted to use each half of the tube to buffer each signal seperately, instead of summing them together directly.  not sure if there is any benefit to this, but the one side will be digital and may have some extra digital noise and i thought this might help to keep the signals seperate and bleedthrough-free.

comparing my schem to yours, i'm using 220k resistors instead of 100k from V+ and pins 4 &5 are connected directly to ground.
i was going off the subcaster and huffer scehmatics as reference points.

[frequencycentral]

why??

i'm just interested in trying something out really.  i plan on using it in a digital circuit, so i thought it might add some "warmth"... maybe it will just heat up the box a bit...   it's worth a try.



rick --
the way i set up the tube symbol is like an opamp...sort of.  reading counterclockwise from the bottom left pin are pins 1-8.
i wanted to use each half of the tube to buffer each signal seperately, instead of summing them together directly.  not sure if there is any benefit to this, but the one side will be digital and may have some extra digital noise and i thought this might help to keep the signals seperate and bleedthrough-free.

comparing my schem to yours, i'm using 220k resistors instead of 100k from V+ and pins 4 &5 are connected directly to ground.
i was going off the subcaster and huffer scehmatics as reference points.

I think your schematic will have gain as opposed to unity with those 220K and the direct to ground cathodes.

Whats the 'huffer'?

[caress]

http://www.diystompboxes.com/smfforum/index.php?topic=55453.0

Also try lowering that 220k plate resisitor for a cleaner sound. 100k is a more normal value.

hmmm.  i'll probably go with 100k to start with... and add trimpots (50k?) between pins 4/5 and ground to adjust gain.  yes, no?

[DougH]

why??

i'm just interested in trying something out really. 

Well, that's certainly a good enough reason.

Let us know how it turns out!

[frequencycentral]

i wanted to use each half of the tube to buffer each signal seperately, instead of summing them together directly.

All the mixers I've seen have a buffer after the pots. What you have at the moment seems to be a passive mixer with two buffers at the front end. Maybe you should consider tacking my circuit onto the end of your circuit - then you would have an active mixer with buffers.

http://www.diystompboxes.com/smfforum/index.php?topic=55453.0

Also try lowering that 220k plate resisitor for a cleaner sound. 100k is a more normal value.

hmmm.  i'll probably go with 100k to start with... and add trimpots (50k?) between pins 4/5 and ground to adjust gain.  yes, no?

I think 100K resistors at the plates and cathodes is the way to go, it centres the signal, but try a few things on breadboard an see.

[Dragonfly]

Better sound quality?

DIFFERENT sound quality....   

Some might like it, some might not.

[caress]

All the mixers I've seen have a buffer after the pots. What you have at the moment seems to be a passive mixer with two buffers at the front end. Maybe you should consider tacking my circuit onto the end of your circuit - then you would have an active mixer with buffers.

ahhh yes that makes sense.  now i can see it...

[frequencycentral]

All the mixers I've seen have a buffer after the pots. What you have at the moment seems to be a passive mixer with two buffers at the front end. Maybe you should consider tacking my circuit onto the end of your circuit - then you would have an active mixer with buffers.

ahhh yes that makes sense.  now i can see it...

...but, it occured to me that if you wish the output to be non inverting you could just use half of my circuit, or take the output from the cathode of the second triode instead of the plate.

If you took the first option and just used half, you could use the spare triode for and extra input buffer, but I'm not sue if you really need the input buffers ()