I have too many questions Re: Rustbucket, ne OG Vintage EHX Attack Decay pedal!

[Carlsoti]

   New caps came in! I verified my order and proceeded to plop the new caps right back where the old ones were, based on the pics I posted previously. As I clicked off my soldering iron for the day, it dawned on me that I should probably verify that the ones that were in it were installed polarity-correct. I think I got lucky and it appears as if they all were. I'll double-check tomorrow before firing it up to verify the voltages.

   IIRC, in another thread, Brian(madbean) had specified a procedure for the order in which to put the ICs on the board, checking voltages for each as you go. That was for a Rustbucket, which I suspect most people are building with sockets in each position. I don't have that luxury, so is that procedure relevant for diagnosing this pedal? Should I just keep them all in, fire it up slowly with the variac, and start taking and recording the V data?
   
   I'm looking forward to getting back to this, but I've also got a mid-80 MIJ Jazzmaster with some custom pick-ups and wiring to finish setting up and a Wilson 707-3 "widow-maker" to sound-test at wall voltage before each of them go back to their owners. This little amp is a real champ-killer with the input voltage dialed back a bit. I sure hope the owner bites Re: my insistence that he have me put in an isolation transformer on this amp. Anyways, these seem like better jobs to finish the day with.

 

[Scruffie]

First things first, post a full set of voltages for every chip, you either need to trace it and correspond the IC numbers to the rust bucket or photo it and annotate the image.

Your next step no order to worry about.

[Carlsoti]

I don't feel like my head is in this today, but I've pushed myself and fumbled through it.

In tracing through the circuit, I found that Q3 and Q4 HAD been reversed when it arrived. They've been changed back and forth a few times now. I tested the components and put them right before taking these measurements.

The voltage chart has columns labeled RB V. These are the voltages from the Rustbucket docs, for reference. I did my best to match component labels to those docs, as well. Attached are the two documents I've created.

I forgot to add it to the chart, but my voltages out of the W02G regulator are -17.29V and +18.29V.

[Scruffie]

IC4 is dead, looks like IC11 probably is too with that output pin voltage.

[Carlsoti]

I've got a replacement for IC4, I'll have to hunt down a 3080 for IC11. Do you figure the other voltages will come back to where they should be with these two fixes, or is this just step two?

[Scruffie]

I haven't thoroughly checked it, but most of the other voltages looked just fine to me.

[Carlsoti]

New parts arrived today! As usual, the place I ordered the hard-to-get part put on a 22% sale the day AFTER I put in the order. Oh well.

New IC4(4558D) and IC11(CA3080AE) parts added to the board. I took voltages from these, but not the rest.

Should I re-check ALL the voltages or get brave and run a guitar through it?

[Scruffie]

A guitar ain't gonna hurt it, time to find out if I missed anything!

[Carlsoti]

Well, the hum is gone, but so is any signal related to the effect. As before, raw signal passes with the effect off and on, and less so as the blend is rotated clockwise. I'll pull it apart, looking for any obvious mistakes. If I don't find anything, I'll take a full-panel voltages again, and start signal tracing.

[Carlsoti]

I found the grounded leg of the Sens Pot had broken, likely due to being flexed back and forth while troubleshooting. Mending that did not fix the issue, so here's a new set of voltages I've just taken. I've highlighted sections that seem to have reversed polarity voltage readings, and others that just seem fishy. These lead me to believe I've installed something incorrectly, or that some other part has failed.

I'm starting to think that building fresh is much more fun that repairing, even when it's something vintage like this thing.

[Scruffie]

Those are differences in the mV... I wouldn't worry about them.

Have you got an audio probe? If not, build one and it's time to start tracing through the circuit to find where the signal is getting lost.

Follow this path for the wet signal.

IC1 Pin 1 > IC1 Pin 7 > IC4 Pin 7 > R28/29/30 Junction > IC11 Pin 6 > IC12 Pin 7

[Carlsoti]

I've got a scope and sig gen. I'll dig into it tomorrow. Thanks again, Scruffie.

[Carlsoti]

   I started tracing this out this morning. It's noisy, but I've got signal up to pin 3 of the MN3007.

I had to back-track as there's nothing on the R28/29/30 junction. I also tracked further down-stream to verify any wonkiness; the 4.7Uf cap is oriented properly. Signal off of pin 7 of the MN3007 is all kinds of crappy, but I'm not certain what I'm supposed to be seeing. The following three images are from pin 7 of the 3007 with the bias at one extreme, centered, and the other extreme.

I thought I was dead in the water until I got another 3007, but then remembered I DID have another one of those, and a CD4047, to boot. I'm not really certain of their operational relationship, but figuring all that out likely won't benefit getting this pedal fixed. I've got the 2 new chips in, but found I had accidentally shut off my scope. I'm letting that warm back up now.  ...and now having to run out for the GF.

Does anyone know where to get replacement CTS pots with the appropriate stand-offs for this and other EHX pedals from the same era?

P.S. I'm hoping to figure out, this weekend, how to format my posts and pics how'd I like them to show. Many works in progress. :/

[Scruffie]

Noisy signal on pin 3 is to be expected, it's being fed by a compressor and depending on where your harmonics pot is set, that'll add some distortion to proceedings.

The 4047 is a clock that tells the MN3007 BBD to work, pins 10 & 11 of the 4047 should have opposing square waves at some frequency in the ~10-100kHz region (too hot to calculate closer).
The bias trim will have a range of about 10-20% somewhere around the middle where it passes signal, you need to adjust it while scoping that junction until that happens and find the spot with the least amount of clipping.
Both chips have appropriate looking voltages and I don't think I have ever seen a broken MN3007 from natural causes... check the bias trim is actually adjusting the voltage on pin 3.

CTS pots you're probably out of luck, the best you can usually do is scour ebay for similar old CTS pots and hack together new ones from the bits you have.

[Carlsoti]

   I'm going to take a break from this project for the rest of the day. There's got to be an obvious problem that I'm missing and it's making me so frustrated that I doubt spending more time on it will be fruitful. I did swap out both the MN3007 and CD4047.

   Bias adjustment is changing voltage on pin 3, from near zero to -15.5-ish V. This appears correct based on the RB schematic. I checked the 4047 for the square wave signal from 10 and 11, which it has with a frequency near 36kHz, IIRC, but I couldn't find another probe to see both at the same time. With my cheap digital o-scope, the trigger function alters the phase, so it appears as if they are IN phase, as opposed to out of phase. There's probably a setting that will help me figure that out, but the broken English that the help-guide is written in is equally frustrating, as it often contradicts itself.

   None of the signals to or from the 3007 look any different than with the original IC. Attached is the "cleanest" signal I can get from pin 7 of the NEW 3007. This is at one extreme of the bias adjustment. It's more or less the same through 30% of the travel. From there, the waveform gets considerably more deformed. With the bias set somewhere in the middle, regardless of waveform, I can get the tiniest signal at the 28/29/30 junction, but it's totally swamped in noise that's an order of magnitude greater.

I'm kinda wishing I had a part-time plumbing job. I could go for some pre-construction demolition right about now.