Rustbucket Issues

[Timko]

I had mentioned in a post in the Shelter In Place sub forum that I am working on a Rustbucket.  I can essentially confirm that there's no effect signal making it out of the circuit as I can turn the blend knob counterclockwise and hear clean signal, then clockwise and hear nothing.  I would also mention that all ICs, transistors, and electro caps were acquired from either Mouser or Smallbear so at the moment I'm going to assume my parts are good.

Using my audio probe, I can confirm that I am getting signal out of pins 7 and 8 of the MN3007. 

I can also confirm that I'm getting signal through the portion of the schematic marked A  However, the signal gets very faint on the Q7 side of R50. 

I traced the audio signal down to D5.  On the anode side I hear signal.  However, no signal on the cathode.

Now for the voltages.  I'm only going to list the voltages of the things that are off as the voltages for everything else looks pretty much in line for the moment.

IC7
-
10.51
.34
0
2.18
0
4.115
0
0
0
0
0
.293
12.48
12.53

IC8
-
.436
.433
12.35
12.35
10.49
4.114
0
0
12.34
12.35
0
.350
0
12.53

IC9
-
12.36
12.36
12.33
-14.36
0
.013
-.473
14.10

IC10
-
0
12.36
10.02
0
12.43
12.43
2.18
12.54

Q3
-
12.50
1.27
4.4

Q4
-
4.14
4.92
4.14

Q5
-
0
-.474
-14.41

I noticed that IC7 was getting warm too which sure isn't good.  The voltages look to be a mess there honestly and I may try re-flowing all of the IC sockets and transistors that don't look right.  But rather than just re-flowing a bunch of parts, is there a particular place I should start looking in terms of the signal path?

[Timko]

Using a 1-Spot power supply so amperage shouldn't be an issue.

[madbean]

The section marked "A" is not part of the audio path. The audio and envelope generator split off at R18 and R12 resp.

If you are getting audio out of the MN3007 then the next place to look is pin6 of IC11 which leads to the effect output. However, no ICs should be getting warm at all. Here's what I do in these situations before anything else: remove all the ICs and put them back in one by one. Start with IC7 and see if it heats up. If so, that's definitely a problem area. If not, keep inserting the other ones and check to see when IC7 heats up (or if any other do the same).

I had this exact problem with my very first Rustbucket build but in that case it turned out to be one of the 3080s that was bad.

I don't have time to review all your voltages tonight but I will take a look at them tomorrow.

[Timko]

I pulled all of the ICs out of their sockets, and put IC7 back in.  I'm still seeing pin 1 read over 10v, and it's still getting warm.

I also did a little more testing on IC7 without the chip in (just the socket).  All of the sockets read 0 other than
4. 14.99
6. 7.17
14. 15.02

Also, without any ICs in, I measured Q3 and Q4 again.

Q3
14.98
2.21
7.17

Q4
7.17
7.94
7.16

I assume Q3 and Q4 should read closer to the expected voltages, even with no ICs present?

[ferdinandstrat]

The voltages are way off which suggests there's a short somewhere or a wrong component value

Sorry for stating the obvious, mine fired up the first time

[Timko]

The voltages are way off which suggests there's a short somewhere or a wrong component value

Sorry for stating the obvious, mine fired up the first time

Totally agree.  The trick is rather than testing every possible connection or value, what logical way can I try to narrow it down to smartly test?  I suppose I should start checking around Q3 and Q4 because it seems like that even without any ICs in the voltages around that should be accurate.

[Govmnt_Lacky]

I would also mention that all ICs, transistors, and electro caps were acquired from either Mouser or Smallbear so at the moment I'm going to assume my parts are good.

Maybe back to basics?

Verify ALL resistor values, cap values, electro orientations, etc.

[Timko]

I would also mention that all ICs, transistors, and electro caps were acquired from either Mouser or Smallbear so at the moment I'm going to assume my parts are good.

Maybe back to basics?

Verify ALL resistor values, cap values, electro orientations, etc.

Yep.  With so many parts I was hoping to see if I could hone in on a particular part of the pedal to begin investigating.  I have a feeling I'm going to start with the network around Q3 and Q4 and go from there.

[Timko]

Update!

I was able to find a resistor that was indeed the wrong value in that Q3/Q4 network.  Now the voltages of IC7 and IC8 are in acceptable ranges!  The bad news is I'm still not getting any effect.  There is nos signal coming out of the output side of the 2nd CA3080 .

I noticed my voltages on the first part of the LM311 don't exactly match what Bean had

-4.7mV
-3.83mV
-1.14mV

Other than that everything else appears to line up now.  Pin 7 of I9 reads -146mV, but since that pin connects to the Base of Q5 I'd expect that value.

Is there part of the envelope I can examine with a scope?  I may start poking around in there to see what's happening tomorrow.

[Timko]

Welp...I tried a couple other CA3080's I had in my L5 build, and those didn't work.  I then decided to re-flow every IC socket connection to see if there was something that wasn't connecting correctly.  Success!  I am still fiddling with the pedal a little as I don't hear a large difference when changing the Attack and Decay controls, but the good news is there's now effect signal coming out!  The Blend, Harm, Edge, and Sensitivity controls all work from what I can tell.  Now to listen to some demos to see what the Attack and Decay are supposed to do...

[Timko]

Yeah, I'm pretty sure the envelope isn't enveloping.  I can hear a signal that's significantly louder than the bypassed signal, but I'm definitely not getting any of those violin swells that I expect.

I'll spend a little more time studying the schematic to see if I can understand how the envelope itself works.  Before diving into the envelope itself, the output of the envelope should affect what's happening at pin5 of IC11 (the 2nd CA3080), right?  That's the bias pin, and the pin I assume that controls how much sound goes to pin 6 of that IC.

I'll try again swapping in some other CA3080s from my L5 build to see if I get some different results too.

[Timko]

Just a little searching on reddit and I find a post from Madbean himself:

The trigger signal isn't delayed - the audio input is (when the blend knob is turned up). It's necessary to do so to allow the envelope to lop off the initial attack. It works like this - the trigger is tied to a gate. The trigger is formed by rectifying the input signal and feeding it into a network of flip-flop and switch chips using some magic I don't totally get myself.

The gate open and closes according to the user setting. The audio is fed through to the BBD which produces a very slight delay to allow the gate to act before the audio gets to the output. This is the basic envelope around the note. Like I mention in the Rustbucket project doc, it's not a true reverse because the pick attack doesn't swell at the closing of the note like you would have in an actual "tape" reverse. But, that's fine because it sounds bad-ass as is.

[Timko]

I'm going to use this as a running log of my investigation.  This morning I spent some time with the schematic, trying to understand what each pin of the IC7 and IC8 do.

IC7 is a CD4013, which is a Flip/Flop device.  Only one of the two flip/flops is used.  Here's what Ch 1 (the first part) is connected to.

Pin 1 (Ch 1 Output) -> IC8 B Control
Pin 2 (Ch 1 Output (inverted)) -> IC8 D Control
Pin 3 (Ch 1 Clock) -> Ground
Pin 4 (Ch 1 Reset) -> IC10 Output (Pin7)
Pin 5 (Ch 1 Data Input) -> Ground
Pin 6 (Ch 1 Set) -> Q3/Q4 Network
Pin 7 (Vss) -> Ground
Pins 8-13 (2nd channel) -> not used
Pin 14 (Vdd) -> +15v rail

I've been reading through this (http://www.learningaboutelectronics.com/Articles/4013-D-flip-flop-circuit.php) to understand what this circuit does.  There's a handy truth table here.

IC8 is a CD4066 which is a quad gate device.  We only use 3 of the gates (A is not used).

Pin 1 (Input A) -> Not Used
Pin 2 (Output A) -> Not Used
Pin 3 (Input B) -> 15v via Attack Control
Pin 4 (Output B) -> Input D & IC9 pin 3 through resistor network (is the 2nd part accurate?)
Pin 5 (Control B) --> IC7 Ch 1 Output (Pin1)
Pin 6 (Control C) -> Q3/Q4 Network
Pin 7 (Vss) -> Ground
Pin 8 (Input C) -> Ground
Pin 9 (Output C) -> IC9 Pin3
Pin 10 (Input D) -> Output B
Pin 11 (Output D) -> Gnd via Decay Control
Pin 12 (Control D) -> IC7 Ch1 Output (inverted) (Pin 2)
Pin 13 (Control A) -> Not Used
Pin 14 (Vdd) -> +15v rail

Here's a great resource I found on this IC (http://www.learningaboutelectronics.com/Articles/4066-quad-bilateral-switch-circuit.php).

One of the immediate questions that I have after this analysis is what is the purpose of the resistor network aroun pins 9 and 10 of IC8, what is the purpose of IC9_A, and what is the prupose of IC10?  I have found some good resources on how to use a CA3080 as a gate (https://www.experimentalistsanonymous.com/diy/Datasheets/AN6668.pdf) but I can't wrap my head around the purpose of those particular parts in producing that gate.

More to come.  The good news is I have plenty of time to think this through .  Learning what a circuit really does, not just how to put it together, is more fun anyway.

[Timko]

Somewhere in the envelope path (which I assume is R12 and past) the AC audio signal should get rectified to DC.  I can definitely see that on pin6 of IC7 that is not happening on my oscilliscope.  I assumed the D2/3/4/5 network formed a full wave rectifier but it's still a distorted AC signal coming into Pin 6 of IC3.  I also checked at R36 after those diodes but I'm not seeing a difference.

Should the signal be fully rectified when it enters pin6 of IC3?

Also, can someone explain the purpose of IC2 (the first CA3080) to me?  it appears to be affecting the audio path with some part of the envelope path but I had assumed that the envelope effect was on IC11, not IC2 (and I think I may be wrong).

[Timko]

After a lot of testing and probing over the weekend, I think something is wrong with one of the CD chips.  I would see the voltage go back to the wrong values, then remove the IC and re-socket it, and watch the values behave correctly.

Mouser is still shipping so I'm going to order new ICs for everything I can (except the CA3080 and the MN3007 since those are obviously out of print).  I was also seeing some weird values on the transistors (again) so I'll order a new pile of those as well (I am running out of 5088s anyway).

To be continued next weekend once the package gets here and I wait a couple days to open it .