need help with circuit design (with reward)

[m-Kresol]

Hello,

I finally found some motivation and time to go back to my take on the EQD palisades, which basically is a TS with bells and whistles with subsequent boost. The original unit has it wired with the TS bypass as master bypass, meaning that the boost can only be used when the TS is engaged.
I added optical bypass to the system and while I was working that out, I figured it would be cool to have the boost available seperately. So i added this option via internal DIP5 switching to go between the master bypass and stand-alone configurations.
This is when it turned bad. I did not consider that when I used the stand-alone switching and optical bypass, the circuit would always be in the chain. or at least the part pictured below. BO1 is the TS out/boost in. IC3B is the optocoupler LDR for the optical bypass.



Like this, I always have a path to GND formed by R17||R15+R16 (approx. 9k9). This way I loose a lot of bypass volume as well as volume going into the boost, rendering it useless. With the TS on, everything works fine. I'm not quite sure why I loose that much signal. 10k is quite a lot, especially considering that in bypass IN and OUT are directly connected (checked the resistance). However, I couldn't find any other explanation.

I started tinkering around with R15 (emitter resistor of the collector-circuitry). The transistor is only used as a buffer, the gain is around 0.9 so no drastic changes in voltage, however current is boosted.
I tried going up to 100k and 1M (also adjusted R14 without any changes). While DC voltages are not really affected and change only a bit, I got gated distortion when R15=1M also during bypass.
I started modelling in LTspice and found the same (clipping of emitter voltage and output currents). I also tried varying R16, but had no luck and always lost quite a bit of volume in bypass. Also, increasing R16 would lead to volume loss of the TS, if activated.

I tought about how to actively remove a connection via a switch, but it would have to be added if the TS is activated, but that would require a 3pdt switch, which I wanted to eliminate with optical switching (so it's a nogo for me)! I'm at the point where I think this was not meant to be. I really would have liked the feature, the pcb is basically ready to go (but needed revising, I screwed up the rotaries big time). Do any of you have any input for me? I'm not sure what else I could vary (the transistor does not change a lot since it's a buffer) or how to get this to work out. I still suspect something obvious that I missed...
I'd appreciate any advice or feedback. To make it worth your while, the person getting me on the right track will get a pcb for free once it's done. thanks for looking.

well, at least I learned some LTspice in the progress.

[m-Kresol]

EDIT: just desoldered C15 and the volume drop is gone and the boost works fine. Obviously the TS is no longer functional

[reddesert]

To clarify the switching circuitry (I think):
The original unit has an overall bypass and a switchable boost, so the options are: bypass, TS, TS + boost.
Your version has: optical bypass, and DIP switches to allow configuring TS, or boost, or TS+boost? (I'm not really clear on whether there are two optical bypasses to allow overall bypass and boost on or not)?

Anyway, suppose that you have a high impedance voltage source like a guitar pickup + 250K volume pot to ground. When the TS is bypassed so that the voltage source is attached to BO1, the audio signal sees the 10K path of R15+R16 to ground, so it heavily loads the high impedance source, pulling down its voltage.  IOW, old-fashioned tone suck, the kind caused by old effects that had only an SPST for bypass and left the effect input connected.

I'm not sure if you can resolve this with the switching (esp since I don't quite understand how many switches you have available). But you might be able to redesign the TS output buffer so it does not provide a low-impedance path to ground.  You might try a common emitter amplifier - output comes from the collector, choose emitter and collector resistors to get gain ~ 1; or use an op-amp output buffer.

[Aentons]

I have one of the first run "18v" Palisades and the boost not being independent pretty much ruins it for me. It's the first and likely only EQD pedal I will ever have because of it. I had thought about building my own for a little while,  but I ended up with a BYOC Overdrive 2 which is like the Dunes but with the boost, which is  independent... close enough for me.

I don't mean to distract from your topic, but I think EQD recently switched the Palisades to relay bypass.

Anyways, I applaud your endeavor.

[m-Kresol]

To clarify the switching circuitry (I think):
The original unit has an overall bypass and a switchable boost, so the options are: bypass, TS, TS + boost.
Your version has: optical bypass, and DIP switches to allow configuring TS, or boost, or TS+boost? (I'm not really clear on whether there are two optical bypasses to allow overall bypass and boost on or not)?

Anyway, suppose that you have a high impedance voltage source like a guitar pickup + 250K volume pot to ground. When the TS is bypassed so that the voltage source is attached to BO1, the audio signal sees the 10K path of R15+R16 to ground, so it heavily loads the high impedance source, pulling down its voltage.  IOW, old-fashioned tone suck, the kind caused by old effects that had only an SPST for bypass and left the effect input connected.

I'm not sure if you can resolve this with the switching (esp since I don't quite understand how many switches you have available). But you might be able to redesign the TS output buffer so it does not provide a low-impedance path to ground.  You might try a common emitter amplifier - output comes from the collector, choose emitter and collector resistors to get gain ~ 1; or use an op-amp output buffer.

You got the switching correct. I have 2 optical bypass systems.
The impedance thing makes a lot of sense! I don't understand impedance good enough, but I will try to put  a buffer in front of it, just to test the theory!
thanks

I have one of the first run "18v" Palisades and the boost not being independent pretty much ruins it for me. It's the first and likely only EQD pedal I will ever have because of it. I had thought about building my own for a little while,  but I ended up with a BYOC Overdrive 2 which is like the Dunes but with the boost, which is  independent... close enough for me.

I don't mean to distract from your topic, but I think EQD recently switched the Palisades to relay bypass.

Anyways, I applaud your endeavor.

thanks for the feedback. This is one of the reasons I wanted to implement the switching option. The other is that I wanted it to be a bit more different from the original unit.

[reddesert]

I believe this is how you want it to work.  I haven't used an optical bypass (though basically it should be similar to any DPDT effect switching), so I looked at how the Rullywow optical bypass board works. I didn't draw the LED and LED-side of the photo-FET, etc - these would be on pin 3 of the switches.



The difficulty here is that you want to be able to move SW1_B to where SW3_B is, using internal DIP or maybe double-pole switches. That is, to go from nested switching of the two effects (stock), to series switching (modified).  I think this is possible, it just may require make/breaking a lot of connections.

[m-Kresol]

thanks, this is correct. but this is not the problem. I got that part figured out. It took me a few hours of mumbling to myself, quite a bit of beer and swearing, but it works.
Problem is that the end of the TS circuit is connected to the front end of the boost. as you show in your schematic this is connected in version A (stock) and disconnected in B, so the switch would need to be moved.

I need to think about this a bit more and see if I can do this with the DIP (1 more row wouldn't be that bad). Will take a few more hours of mumbling.
in the meantime, I'll also test another output buffer setup as alternative

thanks!

[reddesert]

Yeah, I was thinking about that more, and essentially there are several places that effectively want to be wired with an SPDT switch. The schematic helps me see where those are. For example:

TS.OUT goes to: either FET2+SW2_B_pin1, or to SW1_B_pin3. 
OUT goes to either SW1_B_pin2, or SW2_B_pin2.
SW1_B_pin2 goes to either OUT, or FET2+SW2_B_pin1.
SW2_B_pin2 goes to either SW1_B_pin3, or OUT.

I think that's it, and a couple of these connections are named twice, but it needs to be checked.

DIP switches are only SPST, but you can make them do a SPDT by using 2 DIPs. I think you may be able to do this with a DIP6, but would need to draw this out on paper to be sure.

[m-Kresol]

thanks a lot. I will do that. I was thinking about this on my way to work and I think it should be doable. The way I have it now, is not to move the switches, but some connections between them. It's similar, but not quite, I think.

I will report back, likely on the weekend. Thanks for all your input!

[m-Kresol]

reddesert, you good sir earned yourself a pcb! I figured it out with a DIP6 (instead of the existing DIP5) switch. The key was to separate TS and boost. I didn't think it would be that easy to do, but it worked out quite well

As soon as I get my switches in and verify the pinout, I'll order pcbs. Likely early next week.

Thanks again for your help. All it needed was a push into the right direction and some pen and paper!

[reddesert]

Cool. Nice work!

[m-Kresol]

Just as an fyi. This one is verified and the switching works! The mixture of 9mm, 16mm pots with toggle and rotary switches makes this a hard one in terms of height, but it worked out ok. (the rotaries are not screwed on, and the rest fits nicely.)

Thanks everyone for your help, build report coming soon.