Tonepad MXR envelope filter issue

[supercolio]

Hi all,

I'm building an MXR envelope filter from Tonepad, and am having some issues with it. I've built dozen pedals already, so I'm not that newbie 

At the moment, the effect sounds like a low-pass filter; imagine an autowah effect with attack/threshold all the way down. I found some info about how this effect works, IC2 (CD40469) works as oscillator that switches IC3 (CD4066) rapidly on/off, which I believe lets signal to IC1 (CD4069) and that creates the effect. If I'm mistaken please let me know! I used an oscilloscope to check 4069 and yes, it oscillates as it should, potentiometers affect the oscillation. I guess the problem lies somewhere between IC1 and IC3...? I have re-soldered the ic's and used several different chips (not from different manufacturers though).

I have made the reverse sweep mod, and here's the funky part: In position 1 "clean" signal is heard at 7 o'clock position of threshold control, and in position 2 it is heard at 5 o'clock (pm  ). There's really no signal in between!

Does anyone have ideas what could be wrong? Should I try different IC1 chip? I might have some minor differences on some resistor values, so I'm checking them asap.

Thanks!

[GibsonGM]

It won't be a MINOR difference that would do that (49k vs. 51k, say)...please check this thread and try to get us the info it asks for, and post a link to the schematic (NOT layout...).  Then we can probably chip in and make some suggestions!

Welcome to the forum - once you get over these little glitches, it's a lot of fun, so hang in there!

http://www.diystompboxes.com/smfforum/index.php?topic=29816.0

[Govmnt_Lacky]

I built this circuit several years ago and had the same issues as you. The auto wah was very lame and not very "quacky"

I did read somewhere that the IC chips used were important. Something about certain manufacturer's chips not working well. Try looking into the Tonepad build reports on the site (if they are still there) for more info. I think I found it there.

Good Luck 

[supercolio]

It won't be a MINOR difference that would do that (49k vs. 51k, say)...please check this thread and try to get us the info it asks for, and post a link to the schematic (NOT layout...).  Then we can probably chip in and make some suggestions!

Welcome to the forum - once you get over these little glitches, it's a lot of fun, so hang in there!

http://www.diystompboxes.com/smfforum/index.php?topic=29816.0

Alright!
1) MXR envelope filter fron Tonepad
2)here's the pdf file, http://www.tonepad.com/getFileInfo.asp?id=113 (To download file... ->)
3) some resistor substitutes I made were likes of using 200k resistor in place of 180 and 220, I think (I populated the board couple of weeks ago!). Sorry, can't be more specific.
4) I feel I answered the rest in my first post.

Yeah, it is a lot of fun! These filter pedals are difficult to get right

I built this circuit several years ago and had the same issues as you. The auto wah was very lame and not very "quacky"

I did read somewhere that the IC chips used were important. Something about certain manufacturer's chips not working well. Try looking into the Tonepad build reports on the site (if they are still there) for more info. I think I found it there.

Good Luck 

Yeah I read all the build reports already, and the problem was TI chips required different voltage to get oscillating, fixed that with a trim pot. I guess it's not the only problem..
I too heard it's not the quackiest filter out there, but I'd still like to get it working Thanks!

[GibsonGM]

Super!  Where are your voltages?  Can you post pin voltages?  Something may show up there.... Totally doubt those kinds of R subs will affect anything audible...

Do you have an audio probe? Maybe listen to the signal path, see where things start to go wrong?

[idy]

I built this from the tonepad pcb and like it; it may be that it has a built in "blend" with the unfiltered signal that makes it sound more natural.

Don't be surprised that threshold behaves more or less backwards when you flip the direction switch. But the fact that there is no signal in between suggests the filter is not being triggered.

I opened mine up tonight and found that I had added a little transistor preamp to mine with a 'pregain' knob.  Many times envelope followers work better if you feed them with a boosted signal.

Not sure I follow your explanation of how it works. Where did you read that?
I read the schematic the top 3 invertors are a filter. The 4066 is an electronic switch that opens and closes the filter. Its control voltage comes from the envelope follower (the middle row of four inverters. I did not understand the bottom group of three inverters but they do look like an oscillator... or an oscillator and a mixer to mix the wave with the envelope. I don't know.

I can get you voltages and such if you are still working on it.

[supercolio]

Super!  Where are your voltages?  Can you post pin voltages?  Something may show up there.... Totally doubt those kinds of R subs will affect anything audible...

Do you have an audio probe? Maybe listen to the signal path, see where things start to go wrong?

I had an audio probe, not sure where it is. Thanks for reminding about it, somehow I had forgot to use it at all 

Voltages are:

Vs 9.366v

IC1

1) 2.127
2) 2.094
3) 2.124
4) 2.152
5) 2.123
6) 2.129
7) 0
8 ) 3.944
9) 0.229
10) 2.106
11) 2.120
12) 2.150
13) 2.044
14) 4.192

IC2

1) 2.143
2) 2.027
3) 2.027
4) 2.142
5) 2.145
6) 2.352
7) 0
8 ) 0.231-3.220, potentiometers adjust this.
9) 3.067
10) 3.795
11) 1.509
12) 1.288
13) 1.991
14) 4.190

IC3

1) 0-0.220-ish
2) 0-0.7 oscillation, my multimeter is pretty slow so it can't keep up very well.
3) 1.082
4) 1.094
5) 1.090
6) 0.8
7) 0
8 ) 0-2.1
9) 2.156
10) 2.121
11) 0-2.15
12) 0.808
13) 1.086
14) 4.2

IC3's voltages seemed to swing quite bit. However with a multimeter it seemed pretty random at times.

I built this from the tonepad pcb and like it; it may be that it has a built in "blend" with the unfiltered signal that makes it sound more natural.

Don't be surprised that threshold behaves more or less backwards when you flip the direction switch. But the fact that there is no signal in between suggests the filter is not being triggered.

I opened mine up tonight and found that I had added a little transistor preamp to mine with a 'pregain' knob.  Many times envelope followers work better if you feed them with a boosted signal.

Not sure I follow your explanation of how it works. Where did you read that?
I read the schematic the top 3 invertors are a filter. The 4066 is an electronic switch that opens and closes the filter. Its control voltage comes from the envelope follower (the middle row of four inverters. I did not understand the bottom group of three inverters but they do look like an oscillator... or an oscillator and a mixer to mix the wave with the envelope. I don't know.

I can get you voltages and such if you are still working on it.

Sorry my explanation was a bit of a ramble, and I can't find the source atm However, if you could post voltages that'd be super! I tried to find them everywhere with no luck.

[Govmnt_Lacky]

Definitely have a power rail problem.

All of your Pin 14s should have about 9VDC on them. Yours show 4.1V

Check R4, R17, D1, etc. All of the components that feed directly from the 9VDC power rail. Check for proper values and shorts.

[GibsonGM]

Yup^     
Thanks for posting the voltages, Super...now track down why you don't have that 9V!   Usually it's a solder bridge - a blob of solder that crosses from one pad to another, that it shouldn't.   If you find one, just re-melt and push it away with a piece of wire or something.    Gotta check every single part/connection - part of "paying your dues"!
Good luck, let us know if you find anything.

[Mark Hammer]

This is why cleaning off all the solder rosin gunk from the board is a frequent first step for me when troubleshooting one of my own builds.  It's a question of being able to visually separate the shiny gunk from the solder bridges and cracks.  I won't claim it is 100% foolproof, but a valuable part of troubleshooting is being able to say "Nope, that's not it".

[duck_arse]

measure your supply volts at one of the IC's whist you work each pot through its range. see if 9V springs to life then dies.

[supercolio]

Thank you all, somehow I didn't think too much of it!

There was a ~4v drop over R4 (3k, is it even possible?), after bypassing it Vs rail shows correct 9v. It did not fix the effect though, now there's more "clean" signal, and attack potentiometer affects initial transient, just like an attack control would do. I will get back to this first thing tomorrow (I live in Finland, hence the time difference). Might check how's the oscillation doing with a scope, whether it's there or not.

measure your supply volts at one of the IC's whist you work each pot through its range. see if 9V springs to life then dies.

I'll do that tomorrow, thanks for the idea!

[Govmnt_Lacky]

There was a ~4v drop over R4 (3k, is it even possible?), after bypassing it Vs rail shows correct 9v.

Where did you check this "correct 9V' on the circuit? Even though bypassing it showed 9V... did you check the voltage at every IC Pin 14?

Also, is that truly a 3K resistor in there? Did you verify the proper resistor markings? a 3K resistor should NOT have had that much of a voltage drop.

If I were you, I would replace R4 with a known 3K resistor and re-post voltage readings.

[supercolio]

There was a ~4v drop over R4 (3k, is it even possible?), after bypassing it Vs rail shows correct 9v.

Where did you check this "correct 9V' on the circuit? Even though bypassing it showed 9V... did you check the voltage at every IC Pin 14?

Also, is that truly a 3K resistor in there? Did you verify the proper resistor markings? a 3K resistor should NOT have had that much of a voltage drop.

If I were you, I would replace R4 with a known 3K resistor and re-post voltage readings.

I measured 9v on every pin 14 after removing R4. I can't say I'm sure it is exactly 3k , that goes to my to-do list tomorrow

[duck_arse]

now that you mention "the 3k resistor", I remember this coming up in previous threads. was it that you shouldn't be getting 9V on the Vs connections, or was it the R14/R16 divider values? can't remember.

[Govmnt_Lacky]

also, did you jumper the "F" and "E" pads if you did NOT add the Emphasis control?

[supercolio]

now that you mention "the 3k resistor", I remember this coming up in previous threads. was it that you shouldn't be getting 9V on the Vs connections, or was it the R14/R16 divider values? can't remember.

Don't know, R14 is connected to the Vs. According to datasheet 4069 can take up to 15vdc, so I guess 9v is not fatal...?

also, did you jumper the "F" and "E" pads if you did NOT add the Emphasis control?

I did!

[duck_arse]

have you seen this thread?

http://www.diystompboxes.com/smfforum/index.php?topic=106627.0

[supercolio]

have you seen this thread?

http://www.diystompboxes.com/smfforum/index.php?topic=106627.0

I hadn't! There was some info yeah, but wasn't the end result "make another one"? Well, it seems like a common opinion is that mxr-ef isn't that great... If I can't get this baby alive in a reasonable amount of time I might just abandon it. But I'm not giving up yet!

[Mark Hammer]

Given how old some of the layouts are, you can imagine that there is more than one thread here about building an MXR-EF.

Apart from all the usual building bugs to look out for, one of the things that we collectively observed over the years was that some brands of 4069 chips "worked" and others didn't.  At least not at first.

The "engine" of the  circuit is the HF clock made up of 3 invertor sections.  The envelope voltage essentially governs the duty cycle of that clock; setting how long it is on vs off for each clock tick.  One of the invertor sections has a 62k/100k resistor pair dividing down the supply voltage to provide a bias voltage at the input to that invertor.  It would seem that different brands (or maybe  even different chip batches) need a slightly different bias voltage to kickstart that invertor to  work.  Raising the bias voltage would involve either making the 62k lower or the 100k higher, or both.  Lowering the bias voltage would involve making the 62k value 64k or 68k.  Some folks found success with their existing board and chips by tinkering with that bias voltage.

FWIW my own pedal that I bought in 77/78 or so had RCA chips.  In any event, one more thing to consider if nothing else seems to make it go.