MXR 70's Envelope Filter // 4069 ICs

[Svebster]

Hi,

so I've built a MXR Envelope filter clone with an original NOS PCB. I built everything but it doesn't work. I'm troubleshooting it but don't have a multimeter at the moment but I found out that the problem is definitely on the board. I think it might be the IC's, especially the 2 4069.

The schematic calls for "4069UBP". I used CD4069UBE. Do you know these chips? Is that an ok replacement?

Also, there are 2 of these on the board. When turn the effect on I can only hear a clean guitar sound and when I switch the places of the 2 IC's the sound changes to a loud pop whenever I hit the strings hard (the attack knob works here)

So I guess it's either one of these 2 IC's. Theres also a third IC on the board "CD4066" (I used CD4066BE)

[Plexi]

I have to search about: but if I'm not wrong...is a buffered/unbuffered question

[Svebster]

thanks, but i looked at a view pictures ans in some of them exactly the IC i used was used (CD4069UBE)

So I think they are the right IC's but that would mean at least one is not working properly?

[Mark Hammer]

You have used the right chips.

One of the endemic issues with such builds is that different makes/issues of the 4069 invertor can require different bias voltages to kickstart the clock subcircuit that uses them.  The design varies the duty-cycle of a high-frequency clock that opens and closes a pair of CMOS switches, and ends up mimicing two key resistances in the filter section that determine the corner frequency.  No clock, no sweep.

There is a pair of fixed resistors dividing down the supply voltage to provide that bias to feed the HF clock formed by one of the 4069s.  Stock, they are 62k and 100k.  Reducing the value of the 62k side willincrease the bias voltage, and increasing will drop that value.  As I understand it (or misunderstand it), there is a modest range within which the "ideal" bias might fall.  So it is highly unlikely to involve, say, a 43k and 120k, or a pair of 82k resistors.

Make sense?

[Svebster]

Hi,

ah yeah thanks, I saw a video about biasing the same circuit and there is a 3k limiting resistor right after the 9V. So I took that out and replaced it with a 10k pot as a variable resistor. When I turn the resistor close to 10k i can hear the popping sound that is dependent on the attack and when i turn the resistor to minimum, theres is just the guitar signal and maybe a faint hint of the effect far in the background. Could my chips need even more current??

Just to clarify here is my schematic and layout:

[Svebster]

OK!

Thanks everyone for helping! I just figured it out. Every IC manufacturer/batch is different and there are 2 essential biasing points. The 3k at the power supply and the 100k for the 2nd 4069. Turns out the 3k was fine and the 100k was just a bit too large so I replaced it with a 50k.

Even swapping the 2 4069 IC's made a difference, so even 2 identical IC's need to be biased differently!

[StephenGiles]

Good morning/afternoon all - if I may chip in here, don't forget the mod we did to include a second filter in parallel - was it really 17 years ago!!!!!


Same biasing rules apply of course, I used 4049s which is all I had at the time!

[Mark Hammer]

Now that you have it working, it's time for simple-but-useful mods.

There is a 200k feedback resisor (shown in the schematic) in the filter section that sets Q/resonance/emphasis.  I bought an EF somewhere around 1978 and loved it, but had to sell it when I moved across the country in 1980 (tracing it out first).  Around 1986 I bought another new one, and found I liked t more than the other one.  Turns out the original used 200k and the later issue used 240k, increasing the Q a bit more.

Flash ahead a quarter century, and through the combined efforts of many here, we realized that 240k did not have to be the upper limit.  So, consider replacing that 200k value with a 500k pot (wired as variable resistance) in series with something like 82k-100k.  The very-low-resonance settings, in tandem with a slow attack and some chorus, does a very nice job making almost any bass sound more "Jaco-like".  It obviously doesn't make a fretted bass a fretless one (or endow anyone with his skill), but the slow buildup of more harmonic content can resemble the way in which unfretted bass notes can drone.

Conversely, while the sweep of the EF is not as dramatic as any of the Mu-Tron clones, and people traditionally bought the EF for its more vocal sound than for dramatic quack, increasing the resonance can achieve a nice quack.

Stephen Giles figured out how to reverse the direction of the sweep, from up to down, via clever use of an unusedinvertor section.  While it "works", and quite well, I find that the start and end points of a downward sweep need to be different from those of an upward sweep, simply because of the differing harmonic content at different points in the plucked note's lifespan.  One also needs to adjust the Attack time a bit slower when sweeping down, compared to up, for the same reasons.  I've not done the work required, but I suspect the "right" direction reverse would entail Stephen's trick PLUS some change in one or more components that set the default duty-cycle in the clock.  In other words, it would require more than the DPDT toggle that some drawings indicate for it.

Stephen, where have you beenhiding that 2nd-filter mod all these years?  I'm hearing this for the first time.  You've been holding out on me, bro!! 

[StephenGiles]

I'm sure that mod fell into the pot. Who was it that designed a kit and pcb based on all these mods?

[Mark Hammer]

What does the 2nd filter do?  I'm not seeing any substantive difference.  The two filters are configured identically.

[anotherjim]

Does anyone know why the PWM oscillator was designed like that, rather than the basic 2 inverter oscillator that does not require Schmitt trigger technique...