Minimal blender (seanm FET version) help

[thesteve]

I've been working on a 1590A enclosed FET-based Minimal Blender working off of the perf layout found here: http://www.diystompboxes.com/smfforum/index.php?topic=57749

and the schematic here: http://seanm.ca/stomp/minblend.html

A friend of mine wants to be able to blend a fuzz signal with a dry guitar signal. I am using a switchless design so the signal is always going through the entire circuit. I've been working on this off and on for awhile and have reached a point where I feel like I need to start over. Here's a description of what I've done and the results:

As of right now the circuit is put together as described in the perf layout, noting that the jacks in the layout are wired wrong, I have corrected for this and also connected all sleeves to the input ground. The only variation from the circuit I've made is putting a 0.1uf cap on the input instead of a 0.01uf. On the seanm page, it doesn't seem like this should be an issue. The JFET is the 2N5457. The pedal is powered by 9V adapter (Godlyke Power-All).

Right now when I plug into the circuit and have no pedal in the loop, the circuit passes signal just fine. As soon as I put a pedal into the loop, things get sketchy. My understanding is that this blender should pass the dry signal at 100% regardless of the setting of the knob. Instead what happens is that when I have a pedal in the loop with the pedal volume at 0 and the blender at 0 and the pedal on, I get a very weak signal through the output. With the pedal off, things go back to normal. I've tried this blender with a hand-built pedal a friend of mine made, and thinking that it could be a weird interaction with the pedal, I also tried it with a DS-1 and got basically the same result.

Has anyone else built this and had similar issues? I think at this point I'm ready to scrap the perf board in there right now and start all over, but I want to make sure that what I think the blender should do is actually what it will do. Any help I could get working through this would be really great. Thanks.

[thesteve]

I contacted seanm directly. Turns out my understanding of this design is wrong.

[thesteve]

So everything works fine...

Except I've already run into the phase issue. My friend intended to use this blender with a Guyatone TZ-2 Fuzz. With the blender set to 12 o'clock (50/50 mix) and the fuzz on, everything sounds great. With the fuzz off (bypass), the 50/50 volume is very quiet (more quiet than 100% dry or 100% wet).

I've tried to do a lot of reading on phase inverting circuits, but most of the blender circuits I'm seeing with phase inversion in mind already have the polarity switch built into them. In this case, I need to add a polarity switch an existing circuit. I've housed this in a 1590A, so I don't have much room to work with. I suppose I could rehouse it in a 1590B easily enough.

I know it's possible to fit a blender circuit with a polarity switch into a 1590A (Nine-Volt's Mosquite Blender achieves this), but I also understand that I don't have much room to work with nor do I have the expertise they have.

[PRR]

> I don't have much room to work with

Yes, sometimes we are too clever for our own good.

This is the FET buffer bodged to deliver inversion: