A litany of dumbassery

1142

My story seems very close to @craigyb’s. I currently have 10+ boards that have 80% of their components, some are awaiting another components order, others just waiting on me to decide they’re up next. That’s how it stood just 25 minutes ago.

Rewind three weeks. I had just put pots and jacks into one of my 4 prepped precision adsr’s (thanks to @analogoutput) but for the first time in 15 completed modules, it turns on to ~nothing~. What. The. what.

So, I walk away, and next morning pull out the multimeter, and can make no sense of it. No output anywhere, even though there’s a positive output, and a complete inverse output.

I’ve grown though, instead of posting, I go and dive into the schematics, only to realize that the LED on the output isn’t parallel, but actually in-line of the circuit. I’ve skipped LEDs in the past, often because I don’t have any on hand for long reaches from the board to the panel, but here’s one I skipped that’s critical to the circuit! I go back to bed confident I’ve figure it out.

Fast forward a week, install LED. Success! LED is lit! (Nope, no success, still no response difference through the OBA module) OK, I wonder if this is a time that Rich is clever enough to tell us all to get the low wattage 555
CMOS, but bought the right parts from the start annd didn’t realize he needed to tell us that our knockoff chips won’t work because of power draw. Fine, I’ll go buy CMOS chips.

Fast forward until 10 minutes ago, begrudgingly installing the new fancy 555 chip instead of the bulk chips I bought to populate an Oskitone Poly555 for my god daughter…

The 555 chip was in the TL071 socket, and the 71 in the 555, the entire time. Too far back to have a sense of why, but I do remember that night, and it being the only module I worked on (just like I always do when I finalize a module). So much debugging and self doubt, all over swapping two 8 pin chips for who knows what happened in the moment.

Now to go finish the others (starting tomorrow).

7 Likes
1143

yep been there done that …

3 Likes
1144

Well now! that could certainly be a reversal of fortune!

1145

I’ve NEVER done this before. Well, …almost never. :rofl:

That must be why Sam B scribbled the bit about the Yellow Snow right there.

I made the realization a while ago that some of the random art is actually purposeful beyond whimsy.

1 Like
1146

<Breadboarding stuff>

Why is this shift register giving me a whole lot of garbage?

<15 minutes later>

Oh. Huh. How about that.

Adding bypass capacitors really does improve stuff.

6 Likes
1147

Oh how I’ve been there!
Does anyone ever bypass using one of those ground plane sheets that come with the larger breadboards?
All my old wooden breadboards (yes , I’m old) had a sheet of cookie tin *riveted to the underside for just that purpose, especially with radio projects. Most of the kids in my highschool, having no cars would mount their CB DV27s on emptied *biscuit tins.

*Biscuits, (cookies if you were colonized), would be eaten and their volume replaced with a brick so you could balance the whole antenna rig precariously on a windowsill or gutter for a better reception.

*Riveting; an ancient mechanical skill for building bigger butchier Vias.

As you were!

3 Likes
1148

“Mend it Mark” on Youtube did a recent video replacing a ripped pad with a rivet. Turned out it was not even used .

Expensive Pre-amp!!

1 Like
1149

I have just built the MFOS ADSR. Now it doesnt mention anywhere (that I could find) if the pots should be A or B. Can anyone clarify this for me please. I have posted this in here because I am feeling like a dumb ar$e.

1150

No need to feel dumb, different people have different answers and my first guess was exactly backwards from what at least two designs specify:

Maker A D S R Notes
AI Synthesis log log lin log
LMNC lin lin lin lin but this uses the Electric Druid chip which might have different requirements
Schmitz log log log log
YuSynth log log lin log

I wouldn’t think the MFOS design would need anything a lot different from the AI, Schmitz, or YuSynth designs. So I’d go with log (A) except maybe lin (B) for sustain (S).

4 Likes
1151

Thank you. I appreciate the help.

1 Like
1152

I would have said LIN for the lot as your controling a linear progression of a control circuit

But hey , whoever designed the circuit should have specified it ( Tom Whiltshire has it in the documentaion as LIN. )

Rob

1 Like
1153

My take is log for attack, decay, release and linear for Sustain.

Electric druid is a micro processor so Lin for all controls as it can switch between linear and log responses internally in the chip.

2 Likes
1154

It’s log log lin log, which is the standard for pretty most analog ADSR designs. LMNC uses a microprocessor and, similar to other microprocessor designs, all potentiometers are linear 10k or 100k, whatever is considered the most common value.

1 Like
1155

That is, guessing instead of looking up what I did with my ADSR, which was a modified version of the Kassutronics Precision ADSR, which was based on the YuSynth ADSR, which, as mentioned above, uses log log lin log; so does Kassutronics and so does mine.

1 Like
1156

I have trouble sourcing A1M potentiometers so I looked everywhere to see if there is a design that recommends B1M potentiometers instead. There is not. It’s log log lin log all around, with a couple of exceptions that specify 500K for the log potentiometers, and a few others that go up to 2.2M (which are even more difficult to source).

The only exceptions are the ADSRs based on microcontrollers (PIC, ATTiny, Arduino) and those based on the CEM/AS3310 all of which specify common linear 10k or 100k for all the potentiometers.

1157

You could halve the values of the A, D, and R pots, if those are easier to find as log pots, and double the value of the main capacitor. In the YuSynth design for instance those pots are 1M and the capacitors (switch selectable) are 1 µF and 10 µF — you’d get about the same behavior with 500k pots and 2 and 20 µF.

Another option is to fake an audio pot by adding a fixed resistor from wiper to the bottom of the pot:

You get something like log behavior with a resistor 10% of the pot value, so 100k with a B1M pot behaves kind of like an A1M. Note, though, the above link is talking about using pots as attenuators, but in an (analog) ADSR they’re used as variable resistors — not sure if the same rule applies, I haven’t checked.

1158

Unfortunately, the fixed resistor trick works only when the potentiometer is wired as a voltage divider, which is not the case for the A D R potentiometers in ADSR.

I also tried the resistor/capacitor tradeoff, but it is not optimal. Maybe you can go down to 500k (as some designs already do), but not lower than that, and even then the response that you get is not quite the same as using A1M.

In the end, I ordered a bunch of A1M potentiometers from Tayda, but when you factor all the taxes, etc, the cost is similar to AE. In retrospect, I should have probably gone for an AS3310 design, as I can get these locally on the cheap. Too late now, already built an AR with 500k, a Kassutronics ASR with some AE A1M, and now in the process of building a couple of AO precision ADSRs with the potentiometers from Tayda (although I’ll have to kluge those in as their footprint is different) :slight_smile:

1 Like
1159

Really? The PCB provides footprints that should work with both 16 mm and 9 mm Alpha pots.

image
image415×468 4.24 KB

1 Like
1160

I got this footprint. I assume it does not fit?

1 Like
1161

Ah, different orientation. Right then, kludge away.

1 Like