Madbean's Current Lover - works but not quite as wished..

[Longer713]

Hello everyone!

I'm in big need to debug my Current Lover (Electric Mistress) builds. Yes, I have 3 boards in progress and none of them works properly. I mean that before calibration I do get a clean signal (which is good) but during calibration sometimes I reach nasty clock clicking noise or very poor flanging comparing to what you can find on YouTube. I made a run through every single resistor and cap, tried to swap ICs and I'm kinda hopeless.



Here are my voltages:

https://ibb.co/dSRsBF

And the PCBs itself:

https://ibb.co/dqcM5a

Is there anybody able to debud this board? Can it be faulty MN3007 chip?
Thank you soooo much in advance

[EBK]

Your URL has to end in .jpg or .png (or probably .gif would work too) in order to use the [IMG] tag.  I don't know if I'll be able to help, but I can at least take a look once it becomes possible. 

[Longer713]

Strange, it works on my side.

Maybe try these:

https://ibb.co/dSRsBF
https://ibb.co/dqcM5a

[EBK]

Just got a chance to look at this.  Your voltages look OK to me.
Do not be alarmed about your values for pins 12 and 13 of IC4. That part of the chip is unclocked and just sits at arbitrary output values.

Have you tried replacing Q1?

[Longer713]

i just reminded myself that I had a problem with R32. I couldn't find 62k resistor so I can choose between 56k or 68k. Does this resistor have a big impact on overall circuit? I can see that it's close to Vc which goes to Clock trimmer and later MN3007..

All other parts are according to spec, I've checked them few times each one of them.

[EBK]

Maybe the resistor value matters.  Not sure.  You could consider a series or parallel combination to get it closer to spec.  Perhaps 47k in series with 15k or try 75k in parallel with 360k.

And now, the potentially important question:
Where did you get your MN3007 from?

[Longer713]

I bought MN3007 in Musikding, guys on Madbean forum says that if I get any sort of flanging that means BBD's works. Also they are not cheap and have good rates so I don't think it's a faulty one.

I will really try this resistor with proper value, according to instruction Clock trim settings are very important and difference of 6k in value isn't small.

[DrAlx]

R32 should not be critical, but given the choice prefer bigger value to  smaller.
R30 of 15k is a design flaw in my opinion. It is lower than the value used on the original 9V EM which used 39k.  The lower value is an attempt to expand the sweep range but has the drawback that it can end up feeding too low a control voltage to the clock generator and can give drop outs at the top of the sweep due to the VCO trying to go too fast and simply freezing. If the clock freezes the BBD wont pass signal. That has been reported before. R30 should not have been made lower than 22k in my opinion. In filter matrix mode, vary the range pot and measure the voltage on pin 2 of the LM311. If it drops lower than about 0.85 volts you will get a dropout in the sweep. The fix for that is to increase R30.
You haven't given a clear description of what your problem actually is. If you can, post a sound sample on Soundcloud and send a link that would help.
I would first take voltage measurements in filter matrix mode and record some samples in that mode with the colour turned up. That will at least give a clue if the BBD/VCO is working.

[Longer713]

I swapped R32 for a 68k resistor but no significant difference.

So voltages on Pin 2 of LM311 are:

In matrix filter mode: 1,03V with pots fully counter-clockwise;
7,38V with pots fully clockwise

In matrix filter mode: 700mV-1400mV with pots fully counter-clockwise;
3,1V with pots fully clockwise

I'll try to upload some soundclips too.