MXR Stereop Chorus - SAD5101 to MN3008 Conversion

[pedalbob]

Hi everyone-
I have an MXR big yellow stereo chorus that has a dead SAD5101 chip (the big 22 pin one).  I know that is the issue because I also have a big green MXR Delay with the same Reticon chip and when I swap them the Chorus works fine and the Delay doesn't and vice versa.
The 5101 chips are like impossible to find now as you all know unless I find another working unit but what fun is that ?
I've studied the schematic for the 5101 version and also the MN3008 version and other than the MN3008 being an 8 pin chip the other parts of the circuit are nearly identical. They both use mc14013 which just appears to need a minor jumper and then the 5101 has a couple extra transistors in between to a couple of the additional pins.
Am I crazy or can I attempt to mod my pedal to use the MN3008 which I can still get ?
I've built a lot of pedals from scratch so I can handle the actual workmanship.
Just not sure I'm missing something from a circuit standpoint.  Assuming I may need to mess with some biasing with the trimpots but should be able to find some settings online for the MN3008 version.

Thanks for any thoughts on this and sorry for all the words. If I try this and it works I will obviously post some results and diagrams because I can't be the only one in this situation but I've scoured the web and only yfound one conversion for the small micro chorus with the small SAD BBD chip (512). 

[ElectricDruid]

How many stages is the R5101? Is it equivalent? There's basically nothing out there about that chip, as far as I can see. If you've got a link to a datasheet for it, I'd love to see.

What about the schematic for this box we're supposed to be fixing too? That'd help us help you, right?!?

In principle, yes, you should be able to replace one unobtanium BBD with another that's easier to find. In practice, there might be some gotchas, but it's not impossible.

[Govmnt_Lacky]

How many stages is the R5101?

Fairly certain it is 2000 stages. Where as, the MN3008 is 2048 stages.

EDIT: Found this: https://www.synfo.nl/datasheets/Reticon.pdf

[ElectricDruid]

Fairly certain it is 2000 stages. Where as, the MN3008 is 2048 stages.

Close enough! That's about 2% extra, hardly a big deal. If you were getting really picky, you could adjust the clock calibration frequencies upwards a tiny bit to compensate for the extra delay.

EDIT: Found this: https://www.synfo.nl/datasheets/Reticon.pdf

Ooh fantastic, thanks! I notice the R5101 has a single phase clock input. Still, if it's being driven from a 4013, that's no problem since that chip has the biphase output we'd need for a MN-series chip. Unfortunately it does make a pure daughterboard solution impossible - the 4013's Not-Q output won't appear anywhere on the socket for the R5101.

[pedalbob]

Thanks for the Reticon datasheet I hadn't found that.

This is the only schematic I have found for the MN3008 box ... the quality is horrible but it seams to be the only one on the web ... a copy of a copy of a copy scanned in.

The second better image is the SAD5101 version.  I will upload some pictures of the guts as well. 

[pedalbob]

Hang on attachments didn't attach ... 

[pedalbob]

[pedalbob]

Here are respective gut shots.

[Mark Hammer]

Bucket brigade devices require a two-phase clock.  The Panasonic and early Reticon chips expect a two-phase clock at their clock pins.  THESE Reticon chips include an internal flip-flop to transform a single-phase clock into a dual-phase clock.  So, the question that I'm too tired to poke around the datasheets for is: Does the clock pulse one feeds the chip have to be double the derived frequency?

[pedalbob]

Thanks for the insightful question Mark.

Not an expert at this being a mechanical engr. but if you have any jet engine question ... lol

Anyway looking at the MN3008 datasheet it does show 2 "Clock Pulse" pins for CP 1 and CP 2 at pins 6 and 2 respectively.  The associated diagram shows the pulses to be 180 degrees out of phase.
The schematic for the MN3008 version shows pin 1 of the CD4013 chip going to pin 6 of the MN3008 and Pin 2 of the CD4013 chip going to Pin 6 of the MN3008.
On the SAD5101 schematic it shows pins 2 & 5 of the CD4013 connected together (with nothing coming out) and Pin 1 feeding pin 21 of the SAD5101 "Fc".  So if I'm interpreting this correct based on your comment the second pulse is getting generated internal to the SAD5101 but the MN3008 is getting fed two out of phase pulses from the CD4013.

I could be all wet on this. 

[ElectricDruid]

It looks to me like the answer to Mark's question is "yes". In the schematics above, the Reticon version has a 30pF cap next to the 4013 which I guess is the clock timing cap. The MN3008 version has a 100pF in the same position, which is not exactly double but rather double-and-then-some. That would imply the Reticon clock runs much faster than the MN3008.

But I might be way wrong about that being the clock cap.

So if I'm interpreting this correct based on your comment the second pulse is getting generated internal to the SAD5101 but the MN3008 is getting fed two out of phase pulses from the CD4013.

Yes. The 5101 is using only one of the 4013 outputs, and the MN3008 uses them both. That's what we'd expect to see, and it looks that way.

[pedalbob]

Thanks.
So that piece should be easy to modify ... just change the capacitor and run lead from pin 2 of the CD4013 to the MN3008.

The only other difference I see is that the MN3008 version doesn't have Q1 and Q2 just the Q3 which all take signal coming out of pins 12 & 8 of the SAD5101 version and pins 3 & 4 of the MN3008 version. Not sure what those transistors are actually doing but a trimmer is in that circuit too on both so it is probably key. I guess my point is do I need to exactly replicate the MN3008 version or can I leave that alone as long as I modify it to come from the right pins on the BBD chip?  Easy enough to lift Q1 and Q2 out of the circuit and "jumper" them appropriately to make it look like the MN3008 version.