Ok, I've searched through the other threads on this board but still haven't come up with a solution. Jumpers are correct, C5 is jumpered... this is an MN3007 version currently running on a 9V supply. Here's my IC voltages at the moment:
INPUT: 9.5
IC1
1: 4.7
2: 4.7
3: 4.3
4: 0
5: 4.6
6: 4.7
7: 4.7
8: 9.4
IC2
1: 8.8
2: 4.7
3: 5.0
4: 0
5: 0
6: 4.7
7: 6.5
8: 6.5
IC3
1: 9.4
2: 4.7
3: 4.7
4: 4.7
5: 4.7
6: 4.7
7: 4.7
8: 0
9: 4.7
10: 4.7
11: 4.7
12: 4.7
13: 0
14: 4.7
15: 4.7
16: 0
IC4
1: 4.7
2: 4.7
3: 9.0
4: 0
5: 4.7
6: 0
7: 0
8: 0
9: 0
10: 0
11: 0
12: 0
13: 9.4
14: 9.4
IC5
1: 4.7
2: 4.7
3: 4.6
4: 9.4
5: 1.0
6: 1.0
7: 5.2
8: 8.0
9: 4.2
10: Varies
11: 0
12: 4.2
13: 4.2
14: Varies
IC6
1: 0
2: 5.2
3: 2.9
4: 0
5: 9.4
6: 9.4
7: 9.0
8: 9.4
So... what's off? Should half of IC4 be dead like that?
Look at the schematic to see that those pins on IC4 are grounded. You are good there.
You are getting signal through, just no effect, correct?
Do you have any signal going into pin3 of the MN3007? Any signal coming out of pins 7 or 8?
Have you tried any other values for C18 other than 22pF, such as 15pF or 47pF?
Since is the MN3007 version, have you omitted D4 and D5?
Thanks for posting voltages first. They seem to be good, in general.
D4 and D5 omitted....
I am NOT getting signal on pin3 of the MN3007. What does that mean?
UPDATE: I am, actually... but not pins 1 or 7.
I means you are not even getting signal into the BBD. That's good, because it means the BBD may actually be working correctly, but it has no signal to process.
So, you need to trace back and see where your input drops off. Probe pin1 of IC1 and pin7. See if you have output in both of those places.
Crap. I measured wrong earlier. Turns out I do have signal on pin3, but not on the others.
I just switched out all of the ICs with spares one at a time... no change. Still baffled.
I just socketed C18 and swapped in a 47pF. Still no dice.
I got my MN3007s from Mojo Musical. They're reputable, right?
Any chance of a board photo.
Can you try running it at 12 or 15v and rebiassing ? I had a lot of trouble getting my MN3007 PorkBarrell to bias at 9v
Sure - here ya go. I will try running it with higher voltage on my lunch break and see if that changes anything.
Anyone? The only thing I can come up with is that I might have two bad MN3007s (came from same source). Any other theories?
That's probably what you're looking at. Really, if you've used your audio probe and DMM to get to this point and you're coming to that conclusion, that's what you're dealing with.
The only time I'd doubt that is if I knew that the parts I had worked in another pedal already.
Jacob
Yeah, I was afraid of that. I do not have a circuit to test these in, sadly. Over to Small Bear I go...
Grrrr.
Ok, popped in a brand new MN3007 from Small Bear. STILL no effect. I've tried biasing at both 9V and 15V. I've swapped out every single IC one by one. I've double checked my soldering.
I can get it to flange intermittently if I tweak the bias and clock knobs carefully and increase the rate a bit... it will sort of pop in and out of flange every couple of seconds, but won't stay there continuously. I can also hear a clock-like knocking at the extreme of the clock trimmer and a particular setting on the bias.
Any help would be appreciated at this point.
Intermittent flanging is flanging, your BBD works and the audio path seems fine.
Which leaves us tracing further down, the 4049 looks fine, as does the 4013 so we're looking at the LFO & Clocking section.
LFO output isn't running right uness you have it stuck in the Filter Matrix mode for those voltages (I suspect you do).
311 is also stuck. Can you take voltages for the 324 & 311 in both modes for me?
What transistor have you used for Q2? And what diode for D1 & 2?
You're probably right...
Q2 = 2N5087 / D1 = 1n4148 / D2 = 1N4007
Hmm... it appears I've subbed for D1. Aren't 1n4148 and 1n914 interchangeable? I have some 1n914s in stock now if it would be worth swapping that out.
Here's voltages:
DPDT Switch "Up" Position:
IC5
1: 4.7
2: 4.7
3: 4.6
4: 9.4
5: 1.5
6: 1.5
7: 2.6
8: 4.3
9: 4.3
10: 4.3
11: 0
12: 4.3
13: 4.2
14: 4.3
IC6
1: 0
2: 2.6
3: 1.7
4: 0
5: 9.4
6: 9.4
7: 6.3
8: 9.4
DPDT Switch "Down" Position:
IC5
1: 4.7
2: 4.7
3: 4.6
4: 9.4
5: 1.0
6: 1.0
7: 4.3
8: 4.3
9: 4.2
10: 4.2
11: 0
12: 4.2
13: 4.1
14: 4.2
IC6
1: 0
2: 4.3
3: 2.6
4: 0
5: 9.4
6: 9.4
7: 7.6
8: 9.4
You're not getting any varied pins at all now? That means both have locked up.
Check the pin out of your 2N5087, it may be worth pulling it and trying higher gain transistors along with other diodes, this section is infuriatingly picky.
D1 substitute is fine but try a 1N914 in D2, the higher capacitance of the 1N4007 diode seems to be the cause of issues in that section.
I swapped a 1N914 into D2, no perceivable change. What transistor would you recommend for Q2? Pinout on my 2N5087 is E-B-C.
Does the 2N5087 match the board though? Have you put it in the right way... there are different brands with different pin outs sooo...
A 2N5087 is fine, just depends which one.
The 2N5087 has the pinout actually printed on it (which is nice), and I double checked the orientation with the schematic and it does appear to be in there correctly. I swapped it with another 2N5087 I had and perceived no change.
Is there anything else I should try before attempting to swap out D1 as well? That little bugger is going to be near impossible to get to...
Quote from: eldanko on March 28, 2012, 10:59:10 PM
The 2N5087 has the pinout actually printed on it (which is nice), and I double checked the orientation with the schematic and it does appear to be in there correctly. I swapped it with another 2N5087 I had and perceived no change.
Is there anything else I should try before attempting to swap out D1 as well? That little bugger is going to be near impossible to get to...
D1 shouldn't need changing.
So... you're sure all the values you're using are correct... you've swapped the chips... you've swapped the diode and transistor.
Here's some working voltages Lacesensor posted
LM324
1-4.6
2-4.6
3-4.2
4-9.2
5-1.4 - 2.0 ish
6- varies
7- varies
8-jumps 0.6 up to 7.5 and back
9-4.2
10-varies
11-0
12-4.2
13-4.3
14-varies
LM311
1-0
2-varies
3-varies
4-0
5-9.2
6-9.3
7-varies
8-9.2
Here's what I'm getting now:
LM324
1-4.7
2-4.7
3-4.6
4-9.4
5-1.0-2.0 ish
6- 1.1-2.0
7- varies
8-jumps 0.6 up to 8.1 and back
9-4.4
10-varies
11-0
12-4.4
13-4.4
14-varies
LM311
1-0
2-varies
3-varies
4-0
5-9.4
6-9.4
7-varies btwn 7.0 and 9.0
8-9.4
So... Not locked up anymore?
I double checked all resistor values in that area and found no problems...
Lookin' better! Now try playing with that cap value and readjusting the clock and bias.
I just tried a series of values ranging from 10pF to 220pF in C18... still can only get a very intermittent flange, and that's generally only when I bump up the Rate a bit...
I've been trying to bias at 9V and 12V. Would bumping to 15V help at all?
Quote from: eldanko on March 29, 2012, 03:42:40 PM
I've been trying to bias at 9V and 12V. Would bumping to 15V help at all?
I doubt it, it shouldn't be that tricky to bias, while clock and bias do interact a bit it's not massive.
This has got me a bit stumped... can you 'manually' flange it in filter matrix mode?
If so we can either rule out or pin point it to the clock/LFO
Yes, this is easily the most vexing debug I've ever had to deal with. I'm learning a lot, but I sure wish I could get the blasted thing to function.
No luck manually flanging, after twiddling the trimpots and controls several different ways.
Quote from: eldanko on March 30, 2012, 03:03:35 PM
Yes, this is easily the most vexing debug I've ever had to deal with. I'm learning a lot, but I sure wish I could get the blasted thing to function.
No luck manually flanging, after twiddling the trimpots and controls several different ways.
Hmm.. might have spotted something, check the values of R20/21/25/26 C16 and make sure the joints are good.
I'm wondering if it's the Filter matrix switch itself or the opamp feeding the 311.
Hey ED, as I prepare for this build, did you ever find the issue?
Sadly.... no.
This thing has been the "Moby Dick" of my pedal building experience thus far.
Scruffie generously offered to take a look at it, but after calculating the costs of shipping it across the Atlantic and back, I think I may just take another hack at it with a new board since the more expensive parts are socketed anyway.
I'll certainly keep this thread updated if there's any new developments! It breaks my heart, as the enclosure I have planned for this one is pretty freakin badass.
Bummer man. I'll build mine in your honor :)
Dum-dum question, but are there any solder bridges from signal to ground anywhere? Haven't seen it mentioned in the thread is all...
Never a dumb question, and I appreciate you jumping into the fray! Nevertheless, I've triple checked my soldering and it all seems very clean.
Sorry to open an old wound ED :P
Oh, make no mistake Blueshawk... I WILL beat this thing!
The first MBP project I built was a Grapevine. It didn't work. I posted in the forum, re-flowed joints, tried every transistor orientation I could... just didn't work. I gave up, built another, and it worked great.
I picked the original board up once every 6 months or so just to re-inspect it. The third time I picked it up - literally a year and a half later - I noticed that D1 was backwards... that's it. It just took a more experienced, wiser version of myself to know what to look for and catch it.
All that to say, I'm hoping that becomes the case with this one... that someday I will have the mad skillz of Scruffie or other forumites here and figure it out, plug it up, and happily play Police songs all night.
The peseverance and will of the DIY'r !!!! Can't be it with a stick ;D
Hi Ed, just posting to commiserate. I am having the same problem. I am also having an issue with the collusalus!?! I will not let the flanges beat me though. March on brother. I will post if I have an epiphany.