Small Clone flange-mod switching popping noise

[lars-musik]

Hello all,
I recently build a bass chorus for a friend based on the Tonepad Small Clone Layout. I did some research on useful modifications specifically for bass (http://www.diystompboxes.com/smfforum/index.php?topic=67908.0) and did the Flange-Mod described here:  http://www.accidentalaudio.com/2010/10/small-clone-mods/
Now everything worked nicely until the bass player expressed to switch to flanging mode by foot. Unfortunately the switching causes a quite loud popping noise. Is there a chance to avoid that (possibly pulldown resistors, but where)?

Thanks very much in advance, Lars

P.S. I'll attach two imaged of the mods I made.

[Mark Hammer]

I've modded a few Small Clones to have multiple ranges, and the way depicted in your post is the way most prone to switch popping.  The reason is caps pop when they have no opportunity to drain.

If one is content with switching delay ranges (which is what the mod does) when the pedal is in bypass mode, the popping is not an issue.  But if you want to be able to step on a switch, mid-tune, and do it, then a different method is required.  A preferable way to switch between cap valuyes is to exploit the way in which capacitance is calculated/produced.  With resistors, series resistance is simply the addition of the various resistances involved.  With capacitors, it's different.  Effective capacitance is calculated as 1/Ca+1/Cb....=1/C.

So, in the simplest case, if I have two 1000pf caps in series, they're effective capacitance is NOT 2000pf, but rather 500pf (1/1000+1/1000 = 1/C).    If I were to use a toggle to shunt/bridge one of those capss, it would be as if it wasn't there, resulting in an effective capacitance of 1000pf, but the bridged cap has a way to drain off so that it doesn't pop when I bring it back in.

Turning to the Small Clone, if one replaced the 150pf cap with a 220pf and 150pf cap in series, the effective capacitance would be just under 90pf, resulting in a shorter delay range.  Bridge the 220pf cap and you're left with the stock 150pf.  Bridge the 150pf and you're left with 220pf, which produces a slightly longer delay range and a "thicker" sound.  A 3-position toggle, with the toggle's common connected to the junction of the 150 and 220pf caps, and the outside lugs connected to the "other end" of each cap, would get you 3 ranges, pop-free.

My suggestion for a dual-range, swirlier, unit would be to stick a 120 and 150pf cap in series to replace the stock 150, and use a stompswitch to bridge/unbridge the 120pf cap.  An effective capacitance of 67pf (the net result of the two in series) might be a little low, so a pair of 150pf caps, or a 150pf + 180pf, would get you 75 and 82pf, respectively.

Just keep in mind that the Small Clone has no means to introduce feedback, so I would not consider it a true flanger, when modded, but it IS capable of nailing some very nice "slow Leslie" type sounds.

The range of sweep speeds, is actually pretty usable for both, I find.  In particular, the really slow sweeps the described mod aims for are unlikely to be that audible in the complete absence of feedback.  So, from where I stand, your buddy can probably get away without the sweep-rate mod.  If he finds he really needs slower rates, just compromise with a 3u3 cap or 4u7, instead of switching between caps, and replace the 47k resistor before the Rate control with 39k or 36k.  This should get you a wider range of sweeps, including much slower, and most of the faster speeds...though maybe not the fastest.  But most importantly, there will be no pop.

[PRR]

When you dis-connect a capacitor, leave it "floating", it charges to some unknown voltage. When you re-connect that cap, this voltage must discharge. POP!

Use some large resistor to pull the open end of the cap toward the voltage it will have when connected.

In this case, you can put the resistor across the switch. The in-use side of the switch has the normal voltage. A resistor across the switch pulls the other side toward the right voltage.

The smaller caps can use 1Meg to 10Meg. The 10uFd cap has more leakage so needs a lowish resistance. But for simplicity I'd tack-solder 1meg at all points indicated.



Actually... the 10uFd is an LFO, so won't POP the signal. Likewise the one off the CD4047 isn't signal. So you may not need those two.

[lars-musik]

Thank you both very much for your clear explanations and straightforward solutions.

I feel honoured that both of you masterminds took your time looking into this problem. I always follow your posts and advices with great appreciation and trust.

I guess I'll try Paul's pulldown resistor(s) first, as it seems worth a try without disassembling the unit. However, the second Small Clone board I already etched will definitely be build up with Mark's words in mind.
This Small Clone definitely seems worth putting under the microscope as there is a lot to learn for me, especially now that I received my second shipment of mn3007 (this time fortunately no fakes). 

[Mark Hammer]

Paul knows MUCH more than I do, so don't rule out his suggestions yet (though I think mine will take up less space).

Alternatively, there is an old Japanese project here - http://experimentalistsanonymous.com/diy/index.php?dir=Schematics/OOP Japanese Electronics Book - for a combination chorus/flanger project, complete with PCB layout.  You may want to give that a look.  Unlike the Heladito, though, it does require the MN3101 clock driver, which the Heladito replaces with a cheaper and more available 4047.

[lars-musik]

Thank you Mark - that page seems to be a brilliant resource. I wonder whether I could get an early retirement to work my way through it. Have you ever tried that flanger-chorus? Looks promising indeed.

[Mark Hammer]

have I tried it?  No.  Ran out of MN3007s.  Trust me, I have enough flangers and choruses for one lifetime! 

[lars-musik]

I'm relatively new to that diy pedal addiction but - enough flangers of all effects? That I thought impossible.

However, I have got enough of choruses now already: No luck with the pulldown resistors. I put 1 Meg across the hanging 10uF cap and initially 1Meg and eventually 10Meg across the 180/33pF switch but the pop remains unchanged (this sentence by itself sounds quite nice "the pop remains unchanged", music-wise).

Obviously I am able to understand schematics and forum posts completely only when my soldering iron is heated up. Now that it is, I see that Paul wrote "LFO won't pop the signal". Somehow that's what's happening. The popping 3PDT footswitch switches only the "LFO-Bremse" (eng. LFO-Break) labelled parts, that's the 10 uF (sweep) and the 33/180pF cap combination for the clock/delay. Additionally the two resistors are switched that change the dry/delayed mix towards delayed.

I don't understand how - if the LFO cannot be the perpetrator –  the pop is generated. The dry/wet mix surely won't be responsible, right?

[PRR]

> dry/wet mix surely won't be responsible, right?

?? Do you mean Bypass? Is there a link between input and output to send signal around the flanger?

if so, the 0.033uFd cap at the input wants a pull-down resistor, 1Meg-10Meg.

[lars-musik]

No, not the bypass. As far as I understood the schematic, the 20K and the 22K after point A and before point D respectively, control the mix of the dry and the delayed signal to generate the chorus effect. I took one switch of the 3PDT (LFO-Bremse) to shift that balance in favour of the delayed signal, getting more of a leslie-ish sound.  

[Mark Hammer]

I'm relatively new to that diy pedal addiction but - enough flangers of all effects? That I thought impossible.

However, I have got enough of choruses now already: No luck with the pulldown resistors. I put 1 Meg across the hanging 10uF cap and initially 1Meg and eventually 10Meg across the 180/33pF switch but the pop remains unchanged (this sentence by itself sounds quite nice "the pop remains unchanged", music-wise).

Obviously I am able to understand schematics and forum posts completely only when my soldering iron is heated up. Now that it is, I see that Paul wrote "LFO won't pop the signal". Somehow that's what's happening. The popping 3PDT footswitch switches only the "LFO-Bremse" (eng. LFO-Break) labelled parts, that's the 10 uF (sweep) and the 33/180pF cap combination for the clock/delay. Additionally the two resistors are switched that change the dry/delayed mix towards delayed.

I don't understand how - if the LFO cannot be the perpetrator –  the pop is generated. The dry/wet mix surely won't be responsible, right?

The pop results from stored charge being unable to drain off before the next time that cap is connected to a viable path.  Once connected, the charge leaves suddenly, making the pop.  The suggestion that Paul provided, and is congruent with what you will see elsewhere, is that one picks a resistor value that is high enough to make the capacitor behave like it is not part of the circuit, yet drain off enough current that it will behave like it is completely uncharged when you connect it again.

And while I'm not disputing Paul's suggestion, I think that in trying to reuse the original schematic, and stuff as many changes in as possible, he misdrew the suggested speed-range switch.  The bleedoff resistor should be connected between the negative sides of the 10uf and 2u2 caps, so as to provide that discharge path for the 10uf cap.  Instead, the switch is shown as either connecting to the 10uf cap or the bleedoff resistor in series with it, with the bleedoff resistor essentially "hanging".  There should not be any point where the 10uf cap is not connected to the rest of the circuit, even for a brief moment, and the way it is drawn does not provide that.  Just connect the bleedoff resistor between the negative sides of the caps, and use the switch to either connect the 10uf in parallel directly...or not. 

Incidentally, since the additional cap is being summed with the 2u2 already in there, you may find that 10uf is too slow.  A total combined capacitance that is 6 times higher (2u2+10) will reduce your speed by the same ratio.  You may find that adding 4u7 in parallel (total of 6u9 or a little more than 3x as slow) may be your "speed".

[lars-musik]

Thank you again for your advice. Alas I am at a complete loss here.

I'll attach a figure just to make sure I did understand you correctly. There is absolutely no improvement with the pulldown resistors. I even attached a parallel 560K to the 1Meg to see whether the resisitance was too low. No success.

But at last I tried a 4.7uF instead of the 10uF. Honestly, it is too fast. Eventually I put a huge fat "vintage" looking 8.2uF non-polar cap in there. I hope that will support my new working hypothesis that goes somewhat along these lines "but...these pops are darn vintage...no idea why bass players nowadays use pop free machinery...sounds just great, doesn't it?"

[Mark Hammer]

Well, you did do it right, so I'm still not sure what's going on.  I guess the only sensible advice at this point is to recommend doing all the pop-susceptible switching while the unit is in bypass mode.

As for the adjusting of the resistance values between points A and D, the 22k should remain at that value, since it is calculated to achieve unity gain when switching between bypass and effect.  The Heladito is drawn under the assumption of true bypass switching, while the Small Clone uses a JFET to lift/block the delay signal and leaves all the rest of the circuitry in place and on.

Reducing the 22k results in more dry than wet signal.  If you lift one end of that 22k resistor, you have only wet, which will give you a vibrato effect.  Just bear in mind that vibrato is largerly inaudible at slower speeds.

If you want to lighten the chorus or flange effect, then you increase the value of the 20k resistor.  On mine I think I stuck a 25k pot in series with the 20k resistor, so that I could go from a 50/50 mix to fading the wet signal more into the background for a lighter touch (somethng that bass players often appreciate).

[lars-musik]

Thank you once again for your patience and advice. It seems like sometimes there just is no perfect stompbox. This one looks good and sound great in between the pops. Perfect enough.
Best regards from Bremen, Lars