DC-2 Dimension C clone - soliciting opinions on how to design the bypass

[aion]

So, I decided to tackle the Dimension C (!) and am about halfway through the design process for a project that I will release to the community in a couple months. It's going to be in a 125-B and no SMD. It's been a fun challenge and I've got a solid plan to make it all fit..

Now I'm getting to the point where I have to decide which bypass scheme to use. I was originally just going to use the Boss bypass—the DC-2 uses op-amps for the buffers, so as long as you use good-quality op-amps, the bypass should sound great. However, I just found out that the original Boss bypass doesn't work well with the mechanical momentary switches I was planning on designing around - it's made for a tactile switch that has much less bounce, and the circuit can be unreliable when using something other than a tactile. So that's a no-go.

This pedal is a mono->stereo splitter, so traditional true bypass isn't in the cards. It's got to at least pass through the first op-amp buffer stage to cleanly split the signal. I've seen others do senseless things in the name of true bypass (IIRC the Fromel Seraph would actually cut out the 2nd channel when in bypass mode... what?) and I want to make sure this is done right. Just looking for some opinions and feedback on what is the "rightest"

Original schematic here. The two circuit options, as I see it:

1) Preserve the original bypass path, meaning it goes through the input buffer, then pre-emphasis, de-emphasis, and out. Requires the equivalent of three SPST switches, plus LED.

2) Hard-wire the circuit "on" (omit & jumper Q1, Q11, Q12). Split the signal immediately after the input buffer. One pole of the switch grounds the circuit input from that point, and two poles switch the outputs, each going through an output cap. Still requires the equivalent of three switch poles not including LED, but this time they have to be double-throw.

Then there's also the actual technology of the bypass:

1) Mictester's latching relay, using a latching SPST switch and using two DPDT relays in parallel. One of the four relay switches will be unused. The relays are $3 each, but the other parts are pennies.

2) Optical bypass (H11F1/Optotron method), but with a 3PDT stomp switch instead of the usual DPDT. The one downside here is that the post-buffer input would not be grounded in bypass, just blocked by the optoFET, but that's probably fine.

3) 4PDT switch. (I really, really, really don't want to do this because of the inconvenience of sourcing them and the potential for hardware failure, but it's an option.)

I think either circuit scheme will work with any of the three bypass methods, so any combination of the above is an option. Whichever method I decide on will have to be built into the circuit and will be pretty much required, so that's why I'm trying to get some feedback before going full steam ahead on one specific method. What do you guys think? If you were building it, what would be your preference and why?

[czapa tranzystor 2]

regards 

[Yonatan]

How is it going?  Any update on this project?  I recently heard some stereo samples of the DC-2, and it it sounds fantastic!

[flanagan0718]

I've always wanted one of these. A standard optical bypass wouldn't work huh?

Coda FX has a great write up on how to flash a micro controller. I wonder if you could use that and have one of the soft switches too.


Sent from my iPhone using Tapatalk

[aion]

I ended up going with both 1 and 2 and using optical bypass with a standard 3PDT. I'm in the prototyping phase right now - it's 95% built but not yet tested. I've been held up a bit with a house move and prepping for a new baby. If all goes well in the prototyping, I should be ready to release it by the end of September.

[miha]

If you need any info on what's going inside the original unit let me know, I can pop it open and check

[brejna]

Definitely interesting project..

[daleykd]

Sign me up!

[dbp512]

A very important question that seems to not have come up yet: will this include top mounted jacks like your typical projects, or side mounted jacks like your L5?

[aion]

Side mounted. It's going to be very nearly identical to the original DC-2 in proportions and box layout, with a 125B enclosure, top DC jack, and side in/out jacks. The PCB will be a double-decker, with LFO & clocks being on the bottom PCB and the signal path & bypass being on the top. Even though it's 1/3 the enclosure size of the L5 project, it actually has 20-30% more parts total!

[m-Kresol]

Even though it's 1/3 the enclosure size of the L5 project, it actually has 20-30% more parts total!

that I have to see to believe it. are you sure you're not sneaking in SMD 

[alanp]

If it's a PCB sandwich like he said, I can fully believe more parts than the L5.

[aion]

No surface mount! It uses all 1/4w resistors (even the original used 1/8w) and standard box film caps. No arcane parts. There's even room to socket the ICs

The major space-saver compared to the L5 is that there are only three switches as external controls - not four switches and twelve pots! For the L5, the size wasn't so much due to the components as it was the space for the myriad of controls.

[Jebus]

This sounds awesome!

[Muadzin]

I wonder if you can do top mounted jacks with a 125BB? Sure its a little wider then a 125B and almost the same width as a 125 with side mounted jacks, but it might still be a centimeter or more less wide, depending on what kind of jacks you use. Hmm....