Blend or mix vs channel volumes for parallel tubes in preamp?

[walkman]

Hi,

I've made a little amp that has both a pentode and a triode in parallel the first half of the preamp and a triode in the second half. With the pentode and the volume control maxed i get a nice overdrive. With the triode I get a nice clean sound. At the moment i simply disconnect one of the channels as i am testing, but would like to have a choice between both channels or blend between them.

On the chassis I'm using there is only really space for two pots, I could go for two volumes and a mixing resistors , but then i'd be losing some gain off the overdrive channel, and have no room for a master volume.

Is it possible to set up a blend control between the channels , so at full rotation one channel will be off or so low it doesn't matter and the other will be full volume?

Cheers

Walkman

[loudthud]

The only solution I can think of that doesn't involve the 6dB loss in the mix resistors is to use a dual linear pot connected much like a 5E3 without the tone control. The pot would be wired so that as you are turnning up one channel, you are turning the other channel down. The problem with that circuit is often complained about by 5E3 builders. The pot will be very hard to set near either end of the rotation. A 1Meg pot will probably not be useable. All or nothing at either end and a large area of blend in the middle. You could use a 500K or 250K pot but then you start to lose overdrive from the pentode because of loading. Finding a dual pot with audio taper on one section and reverse audio on the other would work if you could find the part.

[walkman]

I wired a dual audio taper 1M pot as that's the only kind of dual pot I have right now. It didn't really work. At full rotation to either side there was no audio. As the pot is turned, there is a brief amount of turn where a clean signal can be obtained then it is over laid with whistling sounds and crackles etc. Pulling the Pentode out i get no signal from the triode section. So i added a a 1M grid leak befor the pot on the triode, now there is a signal but the same problems.

Could it be that there are interactions with the next stage as the following triode is grid leak biased with 10M ( there is a 22n cap before this grid leak)?

Anyway it may be easier to use a switch to pull one of the channels to ground either after individual volume pots or before a single volume pot for both, As either channel worked fine with a 500k pot on their own.

edit . I think i may have wired the pot up the wrong way , now that ive taken a look at the blend pot wiring on the stewmac web site. I give it another go and see if it makes a difference.

[loudthud]

Check the 5E3 schematic. The volume pots are connected backwards like in a Les Paul.

[madkatb]

If you can find a dual 500k or 1M linear pot you could try this (as per loudthud's suggestion)
The Secret Life of Pots
Use a 150k with the 1M or a 75k with the 500k.

[hasserl]

Why not use a 500k linear pot with each end terminal connected to each of the coupling caps from each tube, and the wiper conected to the grid of the next stage. With the pot set to the middle you would have a 50/50 blend of the two tubes with the same effect as 250k mixing resistors. As you rotate the pot either direction the output from one of the tubes will see less resistance and the other will see more until the pot reaches full travel where one tube output will see no resistance and the other will see 500k resistance. Turning the pot the other way will reverse this situation. Seems like a nice way to blend the channels to achieve what you're looking for. Use the other hole to mount a Master Volume.

[walkman]

Still having troubles . This is the way the circuit is wired up. When the triode is pulled the gain increases on the pentode channel and is overdriven as expected. After rewired the triode to a single gang 500k pot and disconnected the pentode and the dual gang 500k pot, it also sounds fine. ( note that the preamp B+ is around 100v , hence the grounded cathodes on the triodes , not 12ax7's)



Think that I'll have to go with a switch and two separate volumes. Is there away that i can make this switchable via a foot switch without using relays?

[walkman]

Ok , wired it up again but this time with only a pair of 470k . Eventually tracked the problem down to the 10M grid leak for the second triode . It had gone way high above 20M. Replaced it with a new 8M2 and everything's working as it should. Also the output has also increased which is a good thing.

So I will keep it as is for a little while and maybe tweak the mix resistors to taste before trying a linked blend as described by Loudthud