rotary switch wiring in small clone

[spenwah]

So, I'm building the Small Clone from TonePad. This is my third pedal build, and I still hardly know what I'm doing, just following the schematic. I want to implement the mod (included on TonePad's schematic) to use a rotary switch to vary the capacitance going into the CD4007. I have a sp12t, but I'm pretty clueless as to how to wire it up. If anybody could sketch me a rough diagram I'd be very grateful. Here's a picture of the insides of the finished product, but I still can't make out what's going on.

Thanks.

[grathan]

I am new at this too, but I will try to help.


You have a sp12t switch but the tonepad schematic says use a sp4t switch? so your just looking to substitute the 12 for the 4? Or am I looking at the wrong schematic?

[spenwah]

Yes, I'm pretty sure the sp12t switch should work... but it will have 8 settings that are all the same. I think it's a sp12t in moosapotamus's build that I linked to.
Thanks

[WLTerry]

Take a look in the math section of the cap switch mod, caps in series are:

Cresult = 1 / [ (1/C1) + (1/C2) + (1/C3) + (1/C4) + ...]

So, with the switching mod with those cap values, the whole resulting capacitance is indicated with every position. The trick is when you change positions, you are short-breaking capacitors. I made a quick drawing of this, the wires and caps on grey in the scheme means those are short-breaked, followed by a view of the rotary switch and the proper wiring for this case.




Hope it helps.
Will

[frank_p]

The grey lines are what are bypassed.

But you have to include the 3rd and 4rth wires (not indicated or underneath).
In other words : the common or/and the ground.

In a two leads in, two leads out fashion, you still have to wire a common contact.

Plus, the grey lines must not only be "grey ghosts", they must be inexistant.

And the common black leads on the A-B string (by under, must not be considerated, as they are: on the same potential, thus non existant or only a point). 

[spenwah]

Thanks so much! Exactly what I wanted to see, and more.

[awgearhart]

Bumping this post...

Going to build a Small Clone soon and want to add the SP4T rotary switch. The image that WLTerry posted is no longer available, but I want to understand how to wire the SP4T switch. The only switch I found available is on Amazon: https://www.amazon.com/uxcell-Position-Selectable-Selector-Rotary/dp/B07JLRM2L7/ref=sr_1_4?crid=1M6C4GQWQ2KHO&dchild=1&keywords=sp4t+rotary+switch&qid=1592399805&sprefix=sp4t%2Caps%2C142&sr=8-4

In this case, I would imagine wiring the input of C1 on the PCB to the COM terminal on the switch, C1 goes from COM to Pin 1, C2 from Pin 1 to Pin 2, C3 from Pin 2 to Pin 3, C4 from Pin 3 to Pin 4, then a wire from Pin 4 to the output of C1 on the PCB.

Does that make sense? Or is there a different way of doing it (with a different switch, etc.)?

[Mark Hammer]

Personally, I would caution against using a rotary switch.  Not that there is anything wrong with rotary switches, but if you're a novice, and you're building something that uses HF clocks, where the wires go starts to become important in addressing noise.

Use of a SPDT on-off-on (3-position) toggle switch will not only simplify wiring, take up less space, not require purchase of an additional knob, and permit shorter wiring (a good thing in HF circuits like this), but likely provide as much variation as you need.

There are two ways to do this.  The simplest is to install the lowest-value cap you want to use on the board, and use the toggle to add each of two additional caps in parallel.  In such instances, I'd select the "default" cap, and attach/solder slender wires to both leads first, before installation, up where that cap's leads enter the body of the cap.  One of those leads will go to the center lug of the toggle.  Now solder your additional caps to the switch, one end of each to the outside lugs.  The "free" end of those two caps are soldered together, and the second wire from the default cap goes to the junction of those two additional caps.  Now, solder the default cap to the board.  Ideally, the leads should be long enough for the toggle to go where you want, but still kept fairly short.

There is the risk of some minimal popping when you switch values with the toggle, but the math is simpler, and the installation easier for a beginner.  The Tonepad document, using a 4-way rotary switch shows effective values of 33pf, 100pf, 180pf, and 330pf.  I can't speak to whether the filtering in the SC is sufficient to keep noticeably longer-than-stock delays free of clock noise, so let's stick to delay ranges around stock and shorter.  Let's say your "default cap is 47pf, and the switch adds in 68pf (=115pf total) and 120pf (167pf total).  These will provide three easily differentiable delay ranges, with the shortest moving in the direction of flanging, sounding great for slow Leslie-type sounds, and the longest being a smidgen "thicker" than stock.

The second way of doing this would be akin to using the first 3 positions shown in the Tonepad 4-way wiring.  So, the toggle would add none, one, or two additional caps in series to achieve three ranges.  The math is a little trickier, though not impossible.  What may be trickier for you is the installation, since we can't make use of a "default" cap soldered to the board.  I vote for method 1 in your case.

[awgearhart]

Thank you Mark! That makes a lot of sense. In fact I'm doing that cap mod on a Brian May Treble Booster at the moment. I can't believe I didn't think of that! 

However, don't you mean a DPDT on/off/on switch? That's what I'm using. 6 terminals, that way I can properly wire the caps in parallel.

[duck_arse]

single pole izz all you need. wire one end of the caps together, and their other ends to the outer switch lugs, one each. then the middle switch lug and the two joined cap leads are put in parallel with your stock cap. middle possie - nothing added. top possie adds one cap, bottom position adds the other cap instead.

[Mark Hammer]

The duck speaks truth.

(Which sounds a lot more complimentary than "the arse speaks truth")

[bluebunny]

(Which sounds a lot more complimentary than "the arse speaks truth")

That's a whole 'nuther idiom, and quite the opposite meaning. 

[awgearhart]

But I can still use a DPDT on/off/on switch too, correct? Solder one cap to the top 2 lugs, other cap to bottom 2, then wires from the PCB cap to the middle...

I only want to clarify since I just ordered a bunch of DPDT's I would like to use and not have to buy even more switches...

[Mark Hammer]

Yes, no problem.  There is no sin in using a switch with more contacts on it than you need, and leaving some to twiddle their thumbs.  Once upon a time, the difference in cost between SPDT and DPDT would have meant something, but not anymore.

[awgearhart]

I appreciate the help! And the future knowledge with the SPDT.

[awgearhart]

I just finished the Small Clone... sounds great!! Thanks Mark for the simplified cap switch mod. I used a DPDT on/off/on instead of SPDT since I have a bunch laying around. Also added a rate LED by wiring the positive terminal to Pin 1 of the LM386 and negative to ground.

[Mark Hammer]

Excellent!

The temptation to dicker extensively with the clock rate on BBD-based devices is strong.  But here's something one needs to remember: all BBD-based circuits will have lowpass filtering based on the anticipated clock-frequencies used.  So if it aims for slower clocks, the filtering is set lower, and if anticipating higher rates, the filtering will be set fairly high; the principle being that one aims for most bandwidth with least clock noise/aliasing.

So, it is technically possible for someone to whip themselves up a cap-switching arrangement that either aims for too low a clock rate, given the lowpass filtering, or aims higher than what the existing filtering would really allow to be heard.

That's not what happened in your case, but is also what might have happened, had you gone nuts with a 12-position rotary switch.  Which is why I thought to recommend against it.

Sometimes, you come out ahead by settling for "enough".