Do We Need/Want a BBD Tester?

[Cybercow]

UPDATE: The schematic is now corrected & updated.

A few years back Dimitris Smith Diamantidis created a neat little BBD test circuit board. It was available thru DSMCZ. But since, Dimitris has pretty much moved on and DSMCZ is out of stock on those PCBs. I had been in contact with Dimitris in working up a more comprehensive BBD test circuit to incorporate also testing the MN3011, but he just dropped off. His PCB was simple yet quite workable, allowing several of the 8-pin BBDs (MN3007, MN3207, MN3101, MN3102 & a few others) and the MN3005/MN3205 to be tested.

I was able to trace his original PCB and added a bit to the schematic for testing the MN3011. His original PCB looked like this:


The 4 pin holes above the 8-pin BBD footprint is designated for installing jumper pins for changing between the two primary clock chips used with BBDs - the MN3101 & MN3102. By changing the jumpers from parallel vertical to parallel horizontal, one could easily test the MN30xx or MN32xx series devices.

So without being to reach Dimitris any longer, I've made some changes to the schematic but not being at all proficient in PCB design, I am stuck at this point in progress to re-vamping the PCB to include the MN3011 chip.

Here is what I have written up so far. (It just needs a PCB layout.):


I have verified the MN3011 addition with the help of a little test jig to augment the original BBD tester . . . .


Any comments, suggestions and/or other ideas are welcome.

 

[Stickman393]

I like this idea! Honestly, I've got a couple of BBDs that I bought early on that I'm very skeptical of. Would be cool to have a test circuit for 'em.

I'm willing to give the PCB design a shot. Granted, I design PCBs like a blind man, but If we're chill with that I could use the practice.

 

[Cybercow]

I like this idea! Honestly, I've got a couple of BBDs that I bought early on that I'm very skeptical of. Would be cool to have a test circuit for 'em.

I'm willing to give the PCB design a shot. Granted, I design PCBs like a blind man, but If we're chill with that I could use the practice.

Please go for it. When you get to Gerber creation, I'll be happy to submit them to JLCPCB for a trial fabrication, build one out and test it.

 

[jwin615]

All for it here

 

[Stickman393]

On it. Working on the schematic now. Thinking a board mounted DPDT would be easier to deal with than jumpers.

Considering alternatives for sockets too. Normal sockets have a tendency of getting loose after a few uses. ZIFs would be cool, infinitely reusable. Dunno though. Gonna look at what's reasonable and available.

Probably just gonna stick with DIPs.

 

[Cybercow]

On it. Working on the schematic now. Thinking a board mounted DPDT would be easier to deal with than jumpers.

Considering alternatives for sockets too. Normal sockets have a tendency of getting loose after a few uses. ZIFs would be cool, infinitely reusable. Dunno though. Gonna look at what's reasonable and available.

Probably just gonna stick with DIPs.

Yeah, a DPDT would suffice to replace the jumpers. I'd prefer DIPs over ZIFs if you please. Thank you!

 

[Cybercow]

On it. Working on the schematic now. Thinking a board mounted DPDT would be easier to deal with than jumpers.

Please be sure there is room on the underside in the corners to put some of those sticky little rubber feet so it can sit evenly and raised slightly from whatever surface it's resting on.

Looking forward to seeing how this goes.

 

[Stickman393]

No prob, wasn't able to do much on this this week, back on it today.

I'm seeing something a little strange on the schematic with the clock, anybody with a second set of eyes feel like checking me on this?

Pin 3 on the clock is meant to be "ground". Granted, the MN3101 uses a negative rail on VCC, meaning most of the time pedals are gonna invert the PSU in order to have a negative voltage across the rails. Thus the jumper arrangement in the original board to invert the power supply. Savvy?

It looks to me like the schematic has this connection wrong. It should be on the switched PSU leg, and should not connect to pin 6 on the DIP8 BBD. Pin 6 on the Dip8 BBD should connect to Pin 4 on the clock. Yeah?
------------------------------------------------
Also: got a little concern:

It seems to me that this design makes it remarkably easy to fry a BBD.

The MN30xx series uses a negative Vgg/Vdd. Absolute maximums across Vgg to ground is (-18Vdc) to ~0.3Vdc. The MN31XX series uses a positive Vgg/Vdd to ground, absolute maximums are (-0.3CVdc) to 11Vdc.

Basically: pop in a BBD and power this thing up without checking that the switch is in the right position, boom. There goes your clock and BBD. Expensive mistake. Which is...well, troubling. And really easy to do. I know I'd probably do it.

I'd like to do something here that idiot-proof this thing to a certain extent.

Ideas:

Double up on the sockets for everything but the MN3011? Have one dedicated row for MN30XX, one dedicated row for MN32XX, clearly labeled.

Have some kind of visual feedback of what mode the board is currently in (could be done with a coulple of LEDs). Use a separate on/off switch?

 

[Cybercow]

Ack! Oh no! You're right. That schematic does have some connection errors. I will have it triple-checked and corrected by tonight. I will replace the image attached to the OP.

But as far as the potential for toasting a BBD or clock chip, it would be just as easy to insert the wrong chip into one of two sockets. It's a common "hazard" of most testing boards as well as basic building of most any production circuit board. (IMO) A line has to be drawn somewhere as far as 'idiot-proofing' goes. Otherwise the project just keeps growing.

 

[Cybercow]

No prob, wasn't able to do much on this this week, back on it today.

I'm seeing something a little strange on the schematic with the clock, anybody with a second set of eyes feel like checking me on this?

Pin 3 on the clock is meant to be "ground". Granted, the MN3101 uses a negative rail on VCC, meaning most of the time pedals are gonna invert the PSU in order to have a negative voltage across the rails. Thus the jumper arrangement in the original board to invert the power supply. Savvy?

It looks to me like the schematic has this connection wrong. It should be on the switched PSU leg, and should not connect to pin 6 on the DIP8 BBD. Pin 6 on the Dip8 BBD should connect to Pin 4 on the clock. Yeah?
------------------------------------------------
Also: got a little concern:

It seems to me that this design makes it remarkably easy to fry a BBD.

The MN30xx series uses a negative Vgg/Vdd. Absolute maximums across Vgg to ground is (-18Vdc) to ~0.3Vdc. The MN31XX series uses a positive Vgg/Vdd to ground, absolute maximums are (-0.3CVdc) to 11Vdc.

Basically: pop in a BBD and power this thing up without checking that the switch is in the right position, boom. There goes your clock and BBD. Expensive mistake. Which is...well, troubling. And really easy to do. I know I'd probably do it.

I'd like to do something here that idiot-proof this thing to a certain extent.

Ideas:

Double up on the sockets for everything but the MN3011? Have one dedicated row for MN30XX, one dedicated row for MN32XX, clearly labeled.

Have some kind of visual feedback of what mode the board is currently in (could be done with a coulple of LEDs). Use a separate on/off switch?

The schematic is now corrected/updated in the OP. (Sorry about that. Doing electronics with lysdexia is an added challenge layer.) Good catch on your part. Thank you!

Yeah, the jumpers simply determine the power supply polarity for which chip set (MN30xx or MN32xx) is to be tested.

The visual cue for a quick glance to know whether the test board is currently set up for MN30xx vs MN32xx is a fair enough idea. But to my thinking, that's what jumper positioning both determines and shows. I've tested just over a hundred various BBD & clock chips and have yet to fry one. Not sure if it's luck or over-cautiousness on my part.

I also think it's important to be aware that the end result board size will determine the fabrication costs. Either way, still looking forward to see what you come up with.

 

[Stickman393]

Easy enough to add a pair of LEDs. Don't even need an extra pole.

Just an update: I'm working through this one at the moment. Unfortunately KiCad doesn't have the MN3011 as a model as-is. I'm gonna need to see what I can do to either find one or design around that. Work's been nuts recently and I haven't had much time to noodle around on this stuff.

Might be another weekend or two. Sorry for the delay, working it in with life and stuff.

*edit* WELL, that was easy. Found the stuff for the MN3011.

 

[Stickman393]

Got a schematic worked out here:

Need a second set of eyes, as mine are now pretty well crossed.

I took some liberties: like adding a rotary switch for routing the MN3011. The idea here would be to use a 1p12t like what stombox parts uses, just limited to 6 positions so each output can be individually tested. Though...if I keep this setup, it would be remarkably easy to cut back 5 of those resistors and just have a single 4.7k on the output of pin 13 on the switch.

I also made some changes to the grounding scheme as some of what was drawn was going to be at odds with the switchable positive/negative ground setup.

I'm a little perplexed on the MN3007 side: are we not going to use out 2? I could always route that to out 2 on the MN3005 slot.

 

[Cybercow]

Easy enough to add a pair of LEDs. Don't even need an extra pole.

Just an update: I'm working through this one at the moment. Unfortunately KiCad doesn't have the MN3011 as a model as-is. I'm gonna need to see what I can do to either find one or design around that. Work's been nuts recently and I haven't had much time to noodle around on this stuff.

Might be another weekend or two. Sorry for the delay, working it in with life and stuff.

*edit* WELL, that was easy. Found the stuff for the MN3011.

No worries, no hurries. I prefer it done right and without any duress. Life, work and play is always a juggling act.

 

[Cybercow]

Got a schematic worked out here:

Need a second set of eyes, as mine are now pretty well crossed.

I took some liberties: like adding a rotary switch for routing the MN3011. The idea here would be to use a 1p12t like what stombox parts uses, just limited to 6 positions so each output can be individually tested. Though...if I keep this setup, it would be remarkably easy to cut back 5 of those resistors and just have a single 4.7k on the output of pin 13 on the switch.

I also made some changes to the grounding scheme as some of what was drawn was going to be at odds with the switchable positive/negative ground setup.

I'm a little perplexed on the MN3007 side: are we not going to use out 2? I could always route that to out 2 on the MN3005 slot.View attachment 93199

Nice! I'll look it over more thoroughly Monday and report back. But the first thing I notice is that the D1 diode (1N5817) should be in series - not parallel.

Thanks!

 

[Stickman393]

My bad! Been using zeners a bunch recently. Will fix.

 

[Cybercow]

And yeah, another great catch! After looking at things, it's clear that you are correct. We need to use Out2. I dbl-checked the connections on my existing PCB, and pin 8 of the MN3007 is directly connected to pin 4 of the MN3005. See the bit circled in red in the attached image.
(I've corrected and updated the schematic in the OP too.)

 

[Feral Feline]

A commendable joint effort!

 

[Cybercow]

Thanks Feral!