help debugging a MXR EF with mods

[Nico666]

Hi!

After some tinkering on simple drive and boost projects (RM, FF, TS and the kind), I decided to take a step forwards to more ambitious circuits. I tried the MXR EF with mods from GGG:

http://www.generalguitargadgets.com/index.php?option=com_content&task=view&id=113&Itemid=26

I strictly followed the layout, using PCB obtained from GGG ; no part substitutions, ICs are socketed and oriented according to the small dot visible on top of them, so the pinout should be correct.

Here is the problem: bypass works without noise, hiss, etc... LED lights up when switched ON but the pedal makes a high level whining noise (ambulance siren comes to mind). I can hear the unaffected sound of the guitar through this noise but it is very faint. When turned, the pots modify the frequency of the noise, as well as the toggle switches.

Using a 9V wall adaptator, I measured the following Vs:

at the DC input, around 10,5V

at the ICs

A1
1: 2.32
2: 1.94
3: 1.94
4: 2.30
5: 2.20
6: 1.94
7: 0
8: 0.40
9: 2.91
10: 3.88
11: 1.37
12: 1.50
13: 2.28
14 4.26

A2:
1: 2.27
2: 2.27
3: 0.4
4: 3.86
5: 2.27
6: 2.31
7: 0
8: 2.17
9: 2.26
10: 2.32
11: 2.26
12: 2.26
13: 2.27
14: 4.26


U1/U2:
1: 0
2: 0
3: 0
4: 0
5: 0
6: 3.86
7: 0
8: 2.26
9: 2.26
10: 2.26
11: 2.26
12: 3.86
13: 2.26
14: 2.26

D1:
A: 3.6
K: 4.28

D2:
A: 3.8
K: 3.6


Any clue about what's going wrong there?
Any help much appreciated. In the meanwhile, I'll return to simple things (Audio probe maybe?)

Thanks a lot!

Nico

[tehfunk]

I remember there is something special about the ICs you can only use a certain brand; it has something to do with one being buffered and the other not, smallbear offers the one that doesn't work so you need to order from elsewhere. I think he sells the TI ones which are buffered, try fairchild semiconductor's. If you already made sure of this I'm not quite sure, it's on my list of things to build. Hope everything works out well for you!

[tehfunk]

Also, this might not help, but check the build reports at tonepad's projects.

[Nico666]

Thanks for your help.
I will check the ICs, dunno which brand they are.
I checked Tonepad build reports but they refer to a different layout/circuit.
Will build an audio probe ASAP and go back to debugging.

[tehfunk]

I am aware they are different layouts, but they are both clones of the Envelope Filter, and therefore, have the same circuit.

[Nico666]

Sorry, no offense meant.
The circuits are indeed the same with some minor differences in parts value.

[mdh]

The story with the ICs is that for some reason that no one has quite pinned down yet, the circuit tends to not work with Texas Instruments ICs in some positions.  It's been on my to-do list for awhile to try and figure out what's up with that, but I got mine working with another brand of ICs, and haven't gotten back to the TIs.  As far as I know, you want non-TI, unbuffered 4069s (I think mine are from STMicro).

However, I'm not sure if I've heard of an audible whine as a symptom of the wrong IC syndrome with this circuit.  In my case, it just sounded kind of gated.  This makes me think that maybe the clock (made up of IC1a-c in the GGG schem) is oscillating at too low a frequency, and getting into the audio range.  If I were you, I'd get out my DMM and measure the frequency on pin 9 of IC1 (the input of IC1d on the schem).  If that frequency is in the audio range, it's probably the source of the whine.  In that case, it might just be a matter of figuring out how to increase the frequency of the clock a little to get it out of the audio range.  Someone more knowledgeable than I might jump in with exactly how to do that, or to tell me I'm full of it.  But IIRC, last time I played with the bits of this circuit, different ICs produced clock frequencies in the range of 18kHz to 50kHz, the low end of which is just at the top of the audio range.

[tehfunk]

Sorry, no offense meant.
The circuits are indeed the same with some minor differences in parts value.

I read my reply again and I realized, "man that came off kind of tersely"  Yeah, no offense taken, but if your ICs are TIs, then there is almost no question your ICs won't be right, but as mdh said, it could be some other more unique problem. good luck!

[Nico666]

@mdh: the chips are labelled CD4069UBCN and CD4066BCN, pointing to Fairchild origin. Not certain if they are buffered or not. I checked the frequences as you suggested. I found freq in the audio range at pins 2,3,4,6,8 and 9 of A1, pins 3 and 4 of A2 and pins 6 and 12 of U1/U2. Might be a problem with the clock indeed. Should I try other ICs?

@tehfunk: it's allright, I myself find it hard to speak my mind without being harsh with a keyboard.

[tehfunk]

Those ICs will work, the 4069 is UB= unbuffered, so I'd point my focus towards mdh's suggestions.

[Nico666]

OK but the 4066 seems to be buffered. Is it relevant?

[tehfunk]

I don't think so.

[Nico666]

OK thanks

[mdh]

You could try other ICs, but if you have a breadboard, it might be worthwhile to breadboard just the clock part of the circuit (A1a-c, and all of the resistors and capacitors associated with them), and play around with that portion of the circuit, seeing how changing parts values effects the frequency of the oscillation.  I wish I knew more about how this particular oscillator works, but I never came across anything exactly like it in a datasheet.  In my MXR EF thread it was suggested that I might up the value of C8 to 1nF or so to lower the frequency and induce whine (to make sure the clock was oscillating), so it would seem reasonable that using a smaller cap there would get you a higher frequency.  You could even just place another 100pF cap in parallel with it temporarily and see if the frequency goes up.  I just measured the clock frequency in my build, and got 43kHz, safely out of the audio range.  I believe someone said in the past that the official MXR schem quoted the clock frequency at ~35kHz.

[tehfunk]

a little off topic, but mdh, how do you like your build, is it modded? did you get to hear it before mods if you did mod it?

[mdh]

I built the Tonepad version, and didn't do any of the mods.  I might in the future, but it's not in a permanent enclosure or anything, so I didn't bother with mods.  I like the stock sound just fine, but I haven't played it in quite awhile.   As I recall, I didn't even really play it enough to really get a handle on the controls.  This isn't to say that the effect was underwhelming or anything, it's just that I think I built it near the end of my last envelope filter phase, and I started getting distracted with modulation effects.  Partly that's a function of the fact that I don't get to play guitar much in my band, and when I do, swooshy atmospheric stuff is more in order.  I think if I brought in the EFs, they'd probably get annoyed with all the wokka-wokka, and I wouldn't ever get to play guitar

[Nico666]

Thank you for your input.
I will try the C8 thing and see how it works.
I'll keep you updated.
Looking at all the threads about the MXR EF, I (retrospectively, as always) think this was not the best choice for a first modulation project. Anyway, it is a good project to learn something new.