what creates this wah effect?

[vdm]

hey everyone,

you may have seen my posted layout for stellan's wah filter a couple days ago, but i've put it together and all i seem to be getting is a change in volume and some distortion.

because the circuit is not an inductor wah or not quite a twin-t wah either, i just want to know what actually creates the bandpass filter effect.

http://www.fortunecity.com/tinpan/humperdinck/223/effect.htm

if anyone can give me an insight as to what could be making the pot only affect volume it would be much help.

thanks
trent

[Transmogrifox]

Aha, I think I see your problem, " all i seem to be getting is a change in volume and some distortion" :

See that 2.2 Meg resistor that biases the noninverting op amp terminal on the input?  It is marked so to imply that it's connected to ground, and this would be fine if you had +/- 9V, but you're probably using a single supply, so you need to connect that node to a V/2 reference  (just a tap from the middle of a couple 10k resistors between ground and supply is fine).

You just need to have the two non-inverting terminals of the op amps biased to V/2 and that should fix your problem.

This filter is just a variation on the typical bridged - T filter.  It's buffered to be able to attain higher resonance, and a little bit better performance, but nothing terribly more special than the Morley pedals.  The input, then, looks like a pseudo-inverting terminal on an op-amp.

[vdm]

hey transmogrifix,

thanks for the help but i am using a bipolar supply and the bias resistor and non-inverting input of the second opamp are both hooked to ground, whilst pin 4 of the opamp is connected to -9V.

im about to attack the circuit with an audio probe, but i was curious that if it is a variation on a twin T filter... then shouldn't the 1.8M resistor across the two identical caps be 2 resistors of half the size and a capacitor to ground??

im still not sure what the problem is... but ive checked my layout on the computer many times... im guessing i've just mis-soldered something but i guess i could have a dud component somewhere, but this has never been a problem before.

if it's not too much trouble it'd be great if you or anyone could check my layout (it's on the first page of posts under "stellan's wah filter")

thanks
trent

[Transmogrifox]

Something that certainly would cause that effect would be if you had connected the 1.8M and 220k resistors to ground where the lines cross on the schematic...that's not a connection.  I don't want to take you for dumb, especially since you ARE using bipolar supplies, but I don't want to leave out an obvious because I often miss obvious things like that---and that is an easy to make possible error there.  The Bridged T (in this case) and Twin T filters don't require that particular ratio in order to make a band pass filter.  The designer apparently thought the resonance and bandwidth and frequency range resulting from the values shown were pleasing to the ear, so that's what he used.

Chances are there's something real simple but super hard to find in your soldering job (if you've soldered it as opposed to breadboard) that you keep overlooking every pass through the circuit.  Whenever I have a problem with debugging a circuit like this, the solution usually hits me at 2 AM in my state of insomnia and I have it figured out the next day.

The only other thing I can think of is that the IC's were damaged, whether you just happened to get the one of 99% that's no good, or you soldered it to hot, or zapped it with an ESD....or maybe it's just a cold-solder joint somewhere--or a hair-sized solder or wire strand that's shorting something you can't see.

[Michael Allen]

Layout looks good man, the cap values are wrong but I assume you changed those to modify the ciruit sound. I think I'll try this one in my vanishing point along with my morley wah because the wav files sound awesome. Lots of quack!

[Michael Allen]

Hey fellas, can't you convert it to a single supply and throw out all that split supply nonsense?

you just need to create a V 1/2+ point off of two 10k resistors, and connect it to the input of the opamp stage right? In this case you would need two separate 10k networks to create a seperate reference point for each input.

Might try it?

[Michael Allen]

well i just tried this on perf real quick using V1/2 reference points and a single supply and it doesn't work. no sound at all, I'm sure i've goofed something though.

[vdm]

hey michael,

i quickly tried the voltage divider and removed the bipolar supply to test if it works, and i too lost all the sound. so i've gone back to the bipolar supply and i have sound again...

the sound seems to be weaker after the first resistor and cap and is similar after the first opamp stage. but after the second stage it is very bassy and distorted.

the wah pot increases the volume greatly when the resistance gets to about or below 5-10k but for every other value including no connection at all the volume is much lower.

im guessing this means there is something wrong with my bridged T filter. I'll try reducing the 4.7uF cap back to the original 1uF and decrease the gain setting resistor in the second stage to around 500-600k.

im still unsure as to how to approach fixing the filter... but i might just try and replace the 1.8M  and the two 2.2nF caps. would it be an idea to use two 1M resistors and a 0.1uF cap to ground instead of the 1.8.

that would create a twin T filter.. so we'll see how it all goes.

trent

[Michael Allen]

Well mine works! Sounds pretty bad right now but it works.....

I've got a bit of volume shifting and distortion through the circuit when I turn the wah pot (500k Audio). I used a .1 input, 33pF instead of 4.7pF, 1uF between stages, and .0027 filter caps, left off the 560p on the output too.

It wahs, but only on the last 1/3-1/4 of rotation. I don't know if the sound is legitamate though, since i'm playing through a Ruby amp into some computer speakers and this whole setup has sounded murky and fuzzy lately. I don't have any other gear to test it on though.

Anyways, as is the circuit works. Oh I also went back to using a MAX1044 dual supply. I think the voltage divider schem i worked up would have worked had I left it. Only I realized just now that I was using a single opamp instead of a dual. If i had changed that before I bet it would have worked. I think you've got a constrution bug somewhere.

[vdm]

im glad you've got it working on some level, and thanks for checking the layout... if i can get it to work it will be my first successful layout.

im getting my first multimeter today so hopefully that will give me some help as to figuring out what's wrong. and i'm going to try and go back to the original circuit a bit more when i get the parts from the shop today.

i'm getting parts for the vanishing point, bar the pots and leds because i want to test some different values for them to make sure i can get a good range of brightness in the led.

i'll get back to you when there is some new developments

trent

[Michael Allen]

Ok, I switched the opamp from a TL072 to an OPA2604AP and the distortion is gone. I replaced the 500k pot I was using with a 100k and it was too much range. Replaced with a 10k and its usable all of it's range but I think i'm missing a little on the bass end so I'll try a 20k or 25k if I can find one. At the very treble end of the pot rotation I get this loud scratching. This happens with all the pots I use so it's a problem with the resistance getting too low in the circuit. I don't know man....

It sounds pretty good as I've got it now, if I could just get some range out of it without the crackle it would be good to go. Sucks that I don't have a wah pedal or have my Vanishing Point even started so if I want wah wah I've got to sit here and turn the pot while I'm playing .

[Michael Allen]

Ok, I put a 1k resistor between lug 3 and ground. That way there is always a minimum of 1k resistance between the caps and ground. It could probably be less that 1k but that's what I had within arms reach. Now there is not any scratching. Just a note: 10k is not enough pot range for this circuit. I don't have anything else right now....

BUT what i do have is: a little distortion. Previously my guitar was set on lead pickup which is a little lower output with less bass content. Therefore less signal to create distortion. I am using a ruby amp with some crappy little computer speaker (you know, dorm room and all) so I'm almost certain it is the setup that is causing the problems.

Anyways, I'm really liking this circuit! If I can find a wah shell I might just use this circuit to fill it in since it sounds so good to me! I can't wait untill I finish my Vanishing point to use this filter with. I was thinking of using a jack for a "sequencer input" on the wah wah pedal. That way I could put the LED/LDR combo inside the sequencer and then run the output to the jack in the wah pedal that breaks contact between the wah pot and the ground and replaces it with the LED/LDR resistance.

I remember John Hollis doing something like this with his Omnidrive. He used a little 2.1mm or something, jack to make/break contact with the clipping diodes. That way he could put different arrangements of clipping diodes into a little 2.1mm plug and just plug in different arrangements. I thought it extremely clever so I plan on trying to incorporate it into my wah wah / vanishing point joint project!

Good luck vdm, I hope you get it working.

[vdm]

hey michael..

glad you're really enjoying it... and glad i could give something back with a layout that can be used.

i got my first DMM today and it's so much fun checking all the resistances of my pots and seeing how large the tolerance is amongst them all... i had a 1M pot at only 810K... but i also found that from +ve to ground was around 8.3 volts.. but i was only registering 3.5V from negative to ground.

after looking at the max1044 datasheet, i was curious if the 10uF capacitor from negative rail to ground was necessary. i should try it anyway.. but it's late and i thought i'd ask if you used it.

thanks
trent

[Michael Allen]

Yeah I used a 10uf from negative rail to ground. I wired it up just like the one shown at Geofex.

[The Tone God]

That circuit can runned on single supply. Following the steps outlined in the "Opamp-eration" article there is a non-inverting and a inverting opamp stage so attaching the non-inverting inputs together to ground will not work in single supply operation. In both cases you will need a 1/2 V+ bias network.

For the first opamp you attach the 2.2M to one bias network instead of ground.  The second opamp stage will have its non-inverting input attached to the second 1/2 V+ network instead of ground. That should get things working.

In theory you should be able to substitue other opamps but there maybe some oddness depending on what you substute in.

This circuit appears to be a slightly more complicated version of the opamp based twin-T. It looks weird since one stage is non-inverting and other is inverting hence the tapping off the input. I haven't built this but my thinking would be towards plundering something from Morely's website. Simpler and known to work both electrically and sonically.

Andrew

[Michael Allen]

I think this filter is much simpler than a morley one. But I do plan on trying out a morley design (Power wah) when i throw together the Vanishing point.

I wired up bias networks as in your article, but at the time I was using a single opamp instead of a dual so it didn't work. I never bothered to check the opamp, I just went to a bipolar supply..

Here's a drawup of what I'm talking about with the sequencer In on a wah pedal..

[The Tone God]

The wiring on intial inspection look ok to me.

I would stick with the morely just on the basis that you don't have to deal with a 1M pot or all the extra parts to make it single supply or the need of a dual supply. To each their own I guess.

Andrew

[Michael Allen]

Just curious. What morely circuit are you going to use?

[The Tone God]

I was looking at using something like a slightly modded "Pro Series PWA" power wha. Pretty simple thing.

Andrew

[vdm]

hey!

i finally got this thing to work! (all hail the new DMM)

turns out after reading RG's article on the max1044 that the 10uF cap is essential to get the circuit to work, and as soon as i plugged it in, it has started working for me.

i used a 50kA pot with a 470ohm resistor in series. the sweep is in only about half the pot's rotation, so i might try putting a parallel 47k resistor across the pot to bring it down to around 23k max resistance.

it does get quite noisy as the sound goes towards the treble end of the sweep and it is also distorting when i play on the lower strings with my les paul. but it's also interesting to note there is a very big volume boost, so i might try decreasing the gain in the second opamp stage, to around a 500k resistance.

i do have a question though as the only LDR's i can get locally go from around 5k up to 200k with my thumb over the end, and then more than 2M in complete darkness. i was wondering if you guys know of any filter effects that would work with this kind of resistance, or maybe use a parallel resistor.

thanks,
trent