FilterFX - 12dB/oct autowah thing with LP/BP/HP and 8 wave shapes

[jackwithoneye]

Taking a few voltages from the working unit on my workbench, the 7.6V seems to be about right. Running my pedal at 9.2V, I get a 1.2V drop across R23 (I used a 68R because I didn't have any 47Rs, so my situation is a bit worse). The pedal draws a fair bit of current (~25mA or so). This is perhaps not surprising with two optocouplers and two LEDs.

Consequently, with Vdd at about 7.8V, Vbias comes in around 3.8V. What's not right is that pin 5 voltage, which should be a lot closer to Vbias than 1.2V. Check the soldering on R3 and pin 5.

Incidentally, I now think I was being a bit exaggerated with the 2M2 values for R2 and R3. Other pedals I've done use 1M in those positions. I'm not entirely sure why I decided this needed a higher value. I was trying it out, I suppose. Anyway, it should work either way - 1M or 2M2.

i switched R2 and R3 to 1M resistors an checked pin 5 soldering, it gives me 1.9V on pin5.

[ElectricDruid]

i switched R2 and R3 to 1M resistors an checked pin 5 soldering, it gives me 1.9V on pin5.

That's still odd. You've still got 3.8V or so for the Vbias, right? So at one end of that R3 resistor we've got the voltage we expect, and at the other end, we've got a voltage which is significantly wrong.

Could C1 be shorted/a bad cap? That mean R2/R3 connect the Vbias to ground, effectively connected in parallel with R21/4K7. Since their values are so large in comparison with R21, you wouldn't see much effect on the Vbias itself. But the other effect would be that pin 5 would be connected to the middle of a R2/R3 voltage divider, so it would see half the usual Vbias level - probably about 1.9V for a 3.8V Vbias. That fits with what we're seeing.

A shorted cap would also explain why you get hideous pops and thumps when you switch the pedal. There'd be no DC blocking, so the input has a hefty DC voltage on it.

Try replacing C1/100n with a new cap, unless there's a bit of solder or something shorting its connections that you can fix.

[jackwithoneye]

i switched R2 and R3 to 1M resistors an checked pin 5 soldering, it gives me 1.9V on pin5.

That's still odd. You've still got 3.8V or so for the Vbias, right? So at one end of that R3 resistor we've got the voltage we expect, and at the other end, we've got a voltage which is significantly wrong.

Could C1 be shorted/a bad cap? That mean R2/R3 connect the Vbias to ground, effectively connected in parallel with R21/4K7. Since their values are so large in comparison with R21, you wouldn't see much effect on the Vbias itself. But the other effect would be that pin 5 would be connected to the middle of a R2/R3 voltage divider, so it would see half the usual Vbias level - probably about 1.9V for a 3.8V Vbias. That fits with what we're seeing.

A shorted cap would also explain why you get hideous pops and thumps when you switch the pedal. There'd be no DC blocking, so the input has a hefty DC voltage on it.

Try replacing C1/100n with a new cap, unless there's a bit of solder or something shorting its connections that you can fix.

i switch the C1 with a new cap. Same results.
Dumb question time : , what is the role of R3 (was 2.2M, and now 1M)? If shorten it, won't i have the good voltage on IC1 pin5?

[ElectricDruid]

i switch the C1 with a new cap. Same results.

Ah, dammit. I thought we were getting close!

Dumb question time : what is the role of R3 (was 2.2M, and now 1M)? If shorten it, won't i have the good voltage on IC1 pin5?

It provides the DC operating point for the op-amp - the "bias level". If you look at the schematic, no other pins have a defined DC level. The other input and the output are just connected together, without it being at all clear what DC level they are at. It could be anything, depending what the op-amp does. The +ve input is therefore crucial. It has its input held at the Vbias level by R3, and then the incoming signal is superimposed on that through C1.

Reducing the value of R3 further just reduces the input impedance. It shouldn't really do anything to the voltage.

Of course, that's a "it shouldn't - according to theory", so if you want to try it go ahead (stick a 100K in there instead) and see what happens, you never know, we might learn something interesting. I think it's a long shot though, to be honest.

[jackwithoneye]

thank you for the explaination, it's overpass my competences, but that's interestinng to understand.

i put a 100K in parallel of the 1M R3 (which gives 90K approx)
result : pin5 at 3.5V

[ElectricDruid]

Ok, I'll have to think about what that means!

Does it help with the thumping when you switch it? After all, that's what we're really trying to fix.

[jackwithoneye]

Ok, I'll have to think about what that means!

Does it help with the thumping when you switch it? After all, that's what we're really trying to fix.

no, it doesn't unfortunately.
And having more voltage on IC1 pin5 doesn't really affect the effect. It's still good, not better, not worse.

[garcho]

If you weren't slamming it with an amped up signal, maybe you were never hitting the power supply limits and distorting the audio, so it sounds the same now. It's still a good thing you got that voltage level where it should be, closer to V/2.

Are you certain that all grounds are connected, especially your input and output jacks? Are you certain there aren't intermittent shorts near any pots or switches? Try wiggling all the panel mount stuff, knobs, switches, led bezel, DC jack, see if that makes any crackling. Do you get any crackling when adjusting the pots? Move all the knobs from one end to the other and listen for any crackling or static sound. Just fishing for clues.

What's your signal chain when testing this? Guitar => pedal => amp? Anything else in there?

[ElectricDruid]

The voltage at pin 5 going up when you reduce the size of that resistor still only makes sense to me if it's part of a voltage divider. Can you do a continuity test across C1 please? I know you've replaced it, so it's probably not the cap, but maybe there's a PCB fault or something.

[jackwithoneye]

wiggling parts doesn't generate pops (only when switching, especially when resonance is on, but it does it the filter selector swotch too) and i checked grounds.
signal path :
guitar->FilterFX->HXStomp->powered Monitoring speakers
and
guitar->FilterFX->'67 Deluxe Reverb Amp
i sometime use a looper instead of guitar to ease testing

continuity across C1 : around 1.1M

thanks again for the brainstorming guys

[strungout]

Question:

So, I'm subbing the Resonance Switch for a pot wired as a variable resistor, connected between the 10K minimum resonance to the stage 3 output. I now have a minimum resistance, 10K, but I'm confused... The problem is the 1M pot I'm using cuts off at the extremes. The 10k takes care of the min resistance. Now, for the max resistance... how do I hook this up? Since I'm using only lug 3 and the wiper o the pot, essentially.
I thought about connecting lug 1 to the 680k resonance resistor (and making it a smaller value, 330k or so) and the lug 3 to that 10k and then the wiper to the stage 3 output... that would just blend between the two, no?

Like I said, I'm confused 

My layout:

[jackwithoneye]

continuity across C1 : around 1.1M

thanks again for the brainstorming guys

does it make sense to you tom?

[PRR]

...   

[strungout]

Thanks, Paul! I didn't think of putting a resistor in parallel with the outer lugs. Basically, I was looking for a setting that will allow some distortion cause by the 'resonance', on top of having a clean sound. I'll play around with that 2M value

[ElectricDruid]

continuity across C1 : around 1.1M

thanks again for the brainstorming guys

does it make sense to you tom?

No, not much, if I'm honest. 1.1M is too big to be a short, but not really big enough to be just board resistance. I'm stumped. There shouldn't be *any* continuity across C1 - hence my request for the test.

[jackwithoneye]

No, not much, if I'm honest. 1.1M is too big to be a short, but not really big enough to be just board resistance. I'm stumped. There shouldn't be *any* continuity across C1 - hence my request for the test.

excuse my french, i still haven't finsihed my "electonic for dummies" book but basically, measure conductivity over C1 gives the conductivity over the virtual ground (R2+R3+R21 through ground and vBios)? Am i wrong?

[ElectricDruid]

excuse my french, i still haven't finsihed my "electonic for dummies" book but basically, measure conductivity over C1 gives the conductivity over the virtual ground (R2+R3+R21 through ground and vBios)? Am i wrong?

No, you're not. Well spotted. So "around 1.1M" is *exactly* what you'd expect to see.
That's a good thing, at least!

[mdcmdcmdc]

Just bumping this thread to say that this is a very cool pedal! I ordered a board last year and finally got around to building and boxing it this week. Worked great on first go, lovely build instructions and documentation. Only spent a few minutes with it sweeping the pots but it did everything one would expect.

Just curious - what are the hi/med/lo resonance settings fixed at?

[ElectricDruid]

Just bumping this thread to say that this is a very cool pedal! I ordered a board last year and finally got around to building and boxing it this week. Worked great on first go, lovely build instructions and documentation. Only spent a few minutes with it sweeping the pots but it did everything one would expect.

Just curious - what are the hi/med/lo resonance settings fixed at?

Errrm...they're fixed at high, medium, and low resonance?!? Lol!

I have no idea what actual Q settings they give, I'm afraid. They just go from "less" to "more". More helpfully, the resonance switch picks one of three resistors for the resonance feedback. The schematic sets those as 10K, 100K and 680K:

https://electricdruid.net/wp-content/uploads/2018/03/FilterFXSchematicPg1.jpg

When I was experimenting with the prototype, I found that exponential changes gave a better jump in resonance. Also note that *bigger* values give *more* resonance. This happens because the resistor actually sets the damping in the circuit. It's set up so that it should oscillate (or close to it) with no resistor in place. Adding more damping (=lower value) reduces that resonance.
My first try was 10K, 100K, 1M, but I found the "high res" 1M setting was too thin and whistley and not that great, so I backed it off a bit to 680K. Of course, your taste may differ, and I always intended that people should experiment with these three resistor values which is why I laid them out in a row on the PCB with a little box around them so people could see what to play with. I recommend wiring a 1M log pot in one of the positions and trying out different settings. You can then choose three settings that suit your style and put those in as fixed resistors. Or if you can find space, wire the pot and leave the switch out.

Honestly though, I agree with Mark Hammer that actually for some settings, two, three, or maybe four (tops!) options is enough, which is why I went with a switch for this option. I don't really believe that Resonance is important enough to need endless tweaking. It's better to have three useful settings than an infinite number of ones that don't differ much. But I *do* totally accept that not everyone's three options need to be the same, so I would want people to tweak that to get the kind of range they're likely to use.

PS: Nice build, btw. I like the minimal panel. Glad to hear you're enjoying the pedal.

[Albo]

Hi guys!
Sorry for warming up this thread!
I just had a serious glance on the FilterFX project since it comes very close to my "ideal filter pedal" on the board of my dreams. Since it's Sync-Jack, the pedal could easily be tapped globaly by my Disaster Area Micro Clock. This is very nice!

One problem remaining:
Would it be possible for Electric Druid or any other human being to "update" the FilterFX PCB for usage with the TapLFO3 instead of the StompLFO?
Keeping all functions of the original FilterFX and adding the Tap Division (and the other waveforms) of the TapLFO3 would be a real dream in terms of a peal build for me.
Could anyone do this?

I personally haven't got the electronic knowledge to plan this, since I'm used to rebuild projects with at least documentation and pcbs. So sorry for that question coming from a noobie.

Cheers!
Albo