Need help with FX Loop master and switchable master volumes

[Gaz]

I have built a 2-channel amp with switchable master volumes and FX loop (first schematic). I had the master volumes after the loop, but the I ran into the common problem of the delay repeats becoming really loud when switching back over to the clean channel (because the master volume is typically set higher than the OD channel).

I found that the solution is to move the masters before the loop. The simplest way to do it is to make the master potentiometers the bottom half of the old voltage divider before the loop. So 1M/10K divider becomes 1M/10KA pot. (second drawing down in schematic).

My problem is that I have a thick chassis and faceplate, and there are no 10KA pots that will fit, so I'm trying to get creative without screwing up the pot taper and especially the load on the tonestack. Is this possible?

The third drawing is an idea I have that seems simple enough, but I'm afraid it will vary the load on the tonestack when rotated, or the taper will be strange on the masters.

I would really appreciate any thoughts on how I can do this. I have 250KA, 500KA, and 1MA pots available with long enough shafts.

Thanks!

[M Fowler]

My problem is that I have a thick chassis and faceplate, and there are no 10KA pots that will fit

http://www.mojotone.com/amp-parts/poten ... wv6WuNdWSo

These CTS for mojotone fit amps made with faceplates they are 3/8" thread x 3/8" in length x shaft is 3/8" long. Also, Alpha for mojotone are the same and I have found that Bourns amp pots that Apexjr sells are the same size.

Mark

[Gaz]

OK, I have the Mojotone 10KA pots and they are still too short by a few threads. I have the Apex Jr. Pots, but there is no 10KA value. I don't know about the CTS, but I am interested in finding an alternate solution rather than ordering more stuff if I don't have to. Thanks though.

[Jana]

If you did the third example, I would use the 250K pots you have.

What I think would be better, however, is to lower the gain of the effects driver so you don't have to cut down so much of your signal with the 1M and 10K divider.

Let me try to scribble up a schematic instead of trying to describe it. Go grab a beer.

[Jana]

Maybe start with R1 = 470K and use the 1 meg pots for R2a and R2b. That will give you the 1 meg load on the tone stack. If it's too hot, increase R1 and/or decrease the pots to 250K.

[Gaz]

Thanks a lot for drawing that out, Jana. I was thinking that since I needed the 1M/10K combo originally to get proper voltage into and out of the loop, that I would need to copy those values verbatim. Is my thinking wrong there?

Are you also saying that I should maybe take out some of the voltage gain at the cathode of the driver like the 100K/2.2K in the SLO as well? There's actually an error in my drawing because I am taking the signal directly from the cathode's driver at the moment to get the full drive and lowest impedance (what I read in Merlin's book).

My thinking was also that I should knock down the voltage before the loop to prevent overdriving the loop's driver. Do you think I should even be worried about that?

Thanks again.

[Jana]

Ignore my ground symbol on the treble pot. It's habit to draw that symbol on pots.

Also, I wouldn't worry too much about the load on the tone stack (within reason, of course).

My thinking is that why are you dropping so much of the signal with the 1M / 10K combo to just boost it back up again with the driver? If it was me, I would try to reduce the gain of the driver as much as possible so that you can use as much of the previous gain as possible (without clipping in the effect driver tube). The driver I drew up should give a hair less than unity gain.

You have a scope, right?

I would hook up the driver like I drew it.

Then hook up a simple 1 Meg master volume (without all the switching).

Then feed a signal and play with the pre amp gains to get the max usable signal that you would use and is available.

Then, turn up the master volume and scope the signal on the input and output of the effects driver and see how high you can go with the master before you start to introduce clipping/distortion in the driver tube. Back the master off a tad and don't touch the master volume.

Carefully unsolder the wires from the master volume and measure the resistances. You now have the voltage divider ratio you need to create the switching circuit with the master volumes.

[Gaz]

Jana, thanks for walking me through the design process - I will do exactly that. What do you think is a reasonable signal level to feed into the amp? 500mV? 1V?

I pretty much copied that driver from Merlin's book (similar to the Dumbleator as well), but I noticed in his new book he changed over to a simpler driver like in your drawing. I wonder why?

[Jana]

If you are asking about the guitar input jack, feed 100 mv. into it. That is about the level of a regular guitar pickup. If you have really hot pickups, maybe 150 to 200 mv. No more than that. I always test with a 100 mv. input signal level.

[Jana]

Maybe because it's simpler?

Simple is good. I like simple. If it gets too complicated, then I get confused and start soldering my fingers.

[Jana]

I'm fading fast and will soon retire for the night. Post back results and measurements if you get stuck--I will check back tomorrow morning.

[Gaz]

Jana,

Thanks again. I set it up like you said, and this what I did:

Fed the amp with a 250mV test signal, and scope'd the input of the new pre-loop master volume for max clean signal. Then I scope'd the output of the driver and turned up the master until the output of the driver started to clip.

(Did I do that right??)

The pot I used was 1.08M, and I measured about 900k/180K at the level just before the driver started to clip. Unfortunately, it looks like I would need a smaller pot than 250K for that setup to work with a 1M resistor, but please correct me If I'm wrong here.

With the 900K/180K divider, I measured a max voltage of 2.3Vrms at the output of the loop, meaing with preamp gain and send level control all the way up. Again, this is with the 250mV input signal. I figured this most voltage could ever come out of the loop, but don't know what voltage range I should be looking for with everything maxed out. 1.23vrms? I understand that is '+4db'.

Do I have any other options besides trying to find some smaller pots? Would my 3rd drawing potentially work with the divider in front of the master? Or would that screw with the load too much?

Thanks!

[d95err]

Instead of a fixed voltage divider before the master volumes, why not just split the 82k cathode resistor and use it as a voltage divider to set the loop level?

In my opinion though, having the MV before the loop is terrible. Any change to the MV will change the loop level. Turn the amp up and you'll distort the effects in the loop, turn the amp down and you get terrible signal-to-noise ratio...

[Gaz]

Thanks for your advice about the master volume position. I did consider the negative aspects, but since it's my personal amp, and I don't really change my settings much on the fly, I figured the benefit of fixing the loud delay trails when switching channels would be worth it.

I think I may have to move at least some of the attenuation to a divider on the cathode (like the SLO), but was concerned about clipping the driver. Should that even be a concern in the first place?

[Jana]

Gaz,
D95err does bring up a valid point about the changing levels. If you are okay with that and want to continue the present line of thinking, use a 1M for R1 and the 250K pots (the two pots will give you 125K total).