News:

Forum may be experiencing issues.

Main Menu

Afterlife TL072

Started by dwhacks, January 01, 2018, 04:12:27 AM

Previous topic - Next topic

dwhacks

So I built an afterlife a few weeks ago, with all the stock values, no subs. I thought it was working, but seemed very minimal so I looked at some video demos, which made it seem a lot more apparent. I checked it my Vactrol is correct, and it is, so I took a standard LED and bent the leads out so they would contact the LED leads on the vactrol, and the led lights up. No guitar signal present, nothing. the brightness doesn't change with guitar signal, so obviously the effect is not working. I tried numerous other TL072s with the same result. These are all from the same seller and I've used them in many projects (eurorack synth and stompbox) without previous issues.

So I tried a tl082 from my parts bin, and low and behold the LED is off, until a signal is present, and it seems to react as expected. The effect is a bit noisier as well, but sure works as I have expected.

This made me think maybe my supply of DIP TL072's is at fault, so I made an adapter for a SOIC tl072, and it has the exact same response as the DIP ones.

So what could possibly be at fault here? Why would these chips work in so many other projects but not the afterlife? What can I test to give some better ideas? I've checked voltages, and don't see anything obvious.

TL082:
4.41v   -|1   8|- 8.76v
4.40v   -|2   7|- 4.40v
2.22v   -|3   6|- 4.40v
0.01v   -|4   5|- 4.40v

TL072
5.39v   -|1   8|- 8.79v
4.60v   -|2   7|- 4.20v
0.79v   -|3   6|- 4.43v
0.01v   -|4   5|- 4.43v

All measured with respect to the ground on the DC jack

fair.child

Have you checked with Audio Probe? It could be cold solder joints or your 3PDT wiring. If you could post some pictures, it will be a huge help for us to understand the problem. I've done two builds of Afterlife and they are amazing optical compressors.

dwhacks

I didn't test with an audio probe because I don't lose audio anywhere. With a signal or not present (guitar not even plugged in, or volume down) the LED in (parallel with) the vactrol is always on.

I will un box it and take some pictures. As I figured this effect was working I have boxed it up, it was only after watching videos that i realized something was wrong.

I have already re flowed all the solder joints, but the cold joint was also a possible culprit in my mind. Still doesn't explain why it works with a TL082 though.

dwhacks

Here are some pictures. The last picture is with the TL082 and the led is off. If I strum the guitar the signal effect the brightness (as expected) and the effect works.

jimilee

Have you tried using a diffused red led instead of water clear?


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

dwhacks

Quote from: jimilee on January 01, 2018, 10:24:06 PM
Have you tried using a diffused red led instead of water clear?

Well the LED is only there for testing, to show that the opamp isn't doing its job. I am using the recommended vactrol for the effect.

somnif

Out of curiosity, how does the resistance across the LDR look in each case? Its in parallel with a 220k, and in series with some stuff that will probably make it look different than the spec values, but it should still be able to tell you if something is going on over there.

I find it interesting how different the voltages are between the 82 and the 72. In theory the two chips should be essentially identical in function, just with a little more stability (ie lower noise) on the 72. And in either case, the LED is being drive by pin 7, which in your two chips is almost identical.  Pin 5 (the power for the LED side) is the same on each chip, so no worries there.

The weirdness is on pins 1 and 3, which are on the sustain side of the circuit. Ok yes, pin 1 does talk to pin 6, with a few resistors in the way (and your two pin 6s are almost the same). But it still seems odd. Why is the pin 1 output on the 72 so hot, why is the pin 3 input so much lower than on the 82. Only thing I can think of is running the little red LED across the vactrol is sucking up another volt and change, so MAYBE thats where the extra juice out of pin 1 is going?

If you've got an extra 75 cents lying around you could grab another '72 from Arrow, just to see if your batch is a bit funky, otherwise... I have no idea. Consult a voodoo priest.

Boba7

I may be off topic but I remember trying a ne5532 in an Afterlife build, and it didn't work at all. It may be an impedance issue. That same ne5532 chip worked in another build (overdrive I think) with no issue.
So my guess would be you have a batch of fake/bad tl072 and the Afterlife can be picky with the opamp used.

jimilee

Quote from: dwhacks on January 02, 2018, 03:53:08 AM
Quote from: jimilee on January 01, 2018, 10:24:06 PM
Have you tried using a diffused red led instead of water clear?

Well the LED is only there for testing, to show that the opamp isn't doing its job. I am using the recommended vactrol for the effect.
The other one between the two pots. The effect uses an LED and a Vactrol for compression. I was having probably let the same issue you were, I bet I read the BOM about 10 times before I realized what it was saying.


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

dwhacks

Quote from: jimilee on January 02, 2018, 03:54:50 PM
Quote from: dwhacks on January 02, 2018, 03:53:08 AM
Quote from: jimilee on January 01, 2018, 10:24:06 PM
Have you tried using a diffused red led instead of water clear?

Well the LED is only there for testing, to show that the opamp isn't doing its job. I am using the recommended vactrol for the effect.
The other one between the two pots. The effect uses an LED and a Vactrol for compression. I was having probably let the same issue you were, I bet I read the BOM about 10 times before I realized what it was saying.

Maybe MadBean himself will chime in on this, but according to the schematic, the indicator led is in no way connected to the circuit.

I've ordered some tl072's from Tayda (usually a good supplier) and I'll look through previous projects to see if I can find one that might have come from mouser or digikey.

I will also take all the measurements again now that its out of the box, and check out the vactrol resistance, although I seem to remember it being ~2m.

jimilee

My apologies . The schematic doesn't and the pcb picture with the audio path  looked like it does, that's my mistake.


Sent from my iPhone using Tapatalk
Pedal building is like the opposite of sex.  All the fun stuff happens before you get in the box.

dwhacks

Well the tl072 is indeed fake. I found one that came in a kit and BAM works great. It even works better then with the TL082 so thats an even greater thing.

So, now we know, the Afterlife is good for testing fake chips!

somnif

Quote from: dwhacks on January 03, 2018, 04:43:41 AM
Well the tl072 is indeed fake. I found one that came in a kit and BAM works great. It even works better then with the TL082 so thats an even greater thing.

So, now we know, the Afterlife is good for testing fake chips!

Excellent.

My guess is its some other dual opamp, likely one with a non-Jfet input (4558, etc). That would change the input impedance and wonk up the carefully balanced control the compressor relies on.

Do you happen to recall where you purchased your chip? I know Tayda has, on occasion, been caught selling poor quality/mislabeled chips and transistors.