Debugging EQD The Depths clone

Started by punchy, July 30, 2023, 07:53:50 PM

Previous topic - Next topic

punchy

Hi, I've recently built a clone of the EQD The Depths pedal, using the pedalpcb Abyss, and the schematic is here: https://docs.pedalpcb.com/project/Abyss.pdf

For the most part the pedal works as expected, the only issue that I've been debugging for a while now is some noise that seems to be some sort of distortion introduced by the circuit. I captured here the noise of the pedal with basically no signal going through it: https://soundcloud.com/user-981938832/the-abyss-noise

I checked all the components to make they're the correct value, and they seem to be. I traced the circuit and it looks to be flowing as it should be. I checked that things that are grounded are indeed grounded, and the VCC/VREF values at each spot in the schematic and the values are good.

I used an audio probe and it looks like the noise is introduced in the third LFO stage, where I can hear some distortion at pin 1 of the TL074. The same noise is also present, but amplified, in the next stage at pin 7. I also used an oscilloscope today that my friend brought over and I'm seeing a what seems to be normal looking sine wave in the first two stages (at pins 8 and 14 of the TL074), and the inputs of stage 3 (pins 2 and 3 of the TL074). Probing the output of stage 3, pin 1, I'm seeing a square wave. I could be wrong in my assumption, but I take this to mean that the signal is getting clipped somehow and possibly causing the distortion. To note, I used the oscilloscope to generate a sine wave that I used as the input into the pedal.

Just to rule things out, I did try other TL074 and TL072 ICs with the same result.

I checked the voltages of the TL074 socket (without the IC) and pins 10, 12, 3, and 5 are around 4.7V (withe VCC being around 9.44V using a Boss power adapter). The ground pin (pin 11 I believe) is 0V and pin 4 shows VCC as expected. The other pins are close to 0 (showing a bit of voltage, around 0.01 or so). With the TL074 in the socket, the voltage of each I/O section of the TL074 is around 4.7V at each pin (+/- 0.1V)

My apologies if I missed something, but any help would be much appreciated. I've hit a dead end here and I'm not sure where to go to next (I'm just getting started in this hobby, so I have lots to learn).

ElectricDruid

Well, it sounds like you've narrowed it down pretty accurately to that third stage. If there's too much gain there causing distortion, you might be looking for a bad joint on one of the components around that stage.

With a sine wave input into a phaser, you should see a sine wave at each stage. At the final output, you should see a sine wave that varies in volume - a phaser is basically a tremolo for a sine wave!

duck_arse

also welcome. please be posting some photos of your build. component side and copper side, and all offboard connecteds. it's the rule. use your meter to test the resistance of each of those 47k resistors. please.
I feel sick.

punchy

Quote from: duck_arse on July 31, 2023, 11:26:07 AM
also welcome. please be posting some photos of your build. component side and copper side, and all offboard connecteds. it's the rule. use your meter to test the resistance of each of those 47k resistors. please.

Thank you for the welcome, and my apologies for not including photos. I've attached a few to this post. Please forgive the two messy LDRs (I just removed the two I had initially installed just on the off chance that they were introducing any noise), and also as you can see the capacitors I had to use were a bit bigger than the board was designed for so I installed C7 on the "back" side.

I measured all 47k and 4.7k resistors just now and their value is as expected.








ElectricDruid

Did you solder that board from the component side?

Also a belated welcome - sorry, I didn't notice it was your first post.

punchy

Quote from: ElectricDruid on July 31, 2023, 02:30:16 PM
Did you solder that board from the component side?

Also a belated welcome - sorry, I didn't notice it was your first post.

No I didn't, I'm guessing you're asking because of the blobs of solder on the component side? If that's the case then I'm guessing I used too much solder and a lot of it flowed on the other side.

idy

The noise on your sound cloud sounds like hiss being phased, and like you are adjusting the rate of your phaser.

We can't tell how bad the noise is without a reference, like a guitar. We hear the bypass click, which should be pretty quiet.

And we don't know how powerful that signal generator you used is. It is probably hotter than a guitar. 1v?

Phasers reinforce and cancel out "notches" in the spectrum. We hear the cancellations more, but the reinforcing can start to distort. You are hearing (or seeing) the distortion after the third phaser stage.

Some phasers are delicate that way. You need to get some perspective on how it works with a real signal.

punchy

Quote from: idy on July 31, 2023, 07:51:42 PM
The noise on your sound cloud sounds like hiss being phased, and like you are adjusting the rate of your phaser.

We can't tell how bad the noise is without a reference, like a guitar. We hear the bypass click, which should be pretty quiet.

And we don't know how powerful that signal generator you used is. It is probably hotter than a guitar. 1v?

Phasers reinforce and cancel out "notches" in the spectrum. We hear the cancellations more, but the reinforcing can start to distort. You are hearing (or seeing) the distortion after the third phaser stage.

Some phasers are delicate that way. You need to get some perspective on how it works with a real signal.

Thanks for taking a listen and for your insight. It does make sense and honestly I've been trying to figure out the source of the noise that maybe I lost track what I'm looking for exactly.

I tried to capture the current state of the pedal, and I tried to record something quick. The chain is guitar -> pedal -> audio interface (AXE I/O) -> pc (using Ableton and the Tonex plugin to sim a clean amp).

There are two recordings, one the guitar, and the other is using an audio probe on each output stage of the TL074. In the guitar recording, the actual guitar signal is clean but there is a background noise that pulsates in time with the LFO. I don't know if this is just "noise" that's being fed into the pedal that's then amplified by the phaser. I kind of hope that's not the case as I'm not sure how to fix that as the current setup is pretty basic. I'll note that the noise is much worse when playing through my real amp.

On the probe clip, the signal is pretty clean in the first two stages (the probe itself does introduce some noise), but there's a very obvious noise introduced in the third stage and that's what I've been trying to figure out. If it's "normal" at that stage in the circuit then I'd really appreciate it if someone could explain it to me why that may be the case.

Thanks again for your help, and I do hope this won't be a waste of your time where it turns out there's no real problem.

https://soundcloud.com/user-981938832/sets/abyss-pedal-debug

idy

On a guitar that sounds like a nice phaser. And not too noisy. It might sound different to you though...

It's advertised as kind of a univibe, which is to say a "diffuse" phaser. Those caps (2, 3, 5 and 6) are Univibe values. I have not seen C4, a 100p in the feedback loop of the second stage. That should only tame super highs.

It sounds OK. I do hear, on the second track, I guess you are probing each phase stage, and three sounds louder and distorted, but four sounds softer again. But that is with an artificial signal.

Hopefully you followed duck_arse's advice: to triple check the resistors and caps around that stage and reflow the solder there.

In the pictures you have an odd idea  of placement for the LDRs. Usually they go close to and facing the LED, not waving in the wind. It sounds like you are putting the works in a dark place (like inside a box) before using it? Usually you can't really make out the effect on these things unless they are in darkness.

It also seems the controls leave some room to go wrong, and are confusingly named. Both Depth and  Voice should affect the brightness of there LED. Maybe voice biases it so the phasing never goes too deep. Throb seems like, at max it could cause your problem. It changes the gain or frequency (or something) of the second stage. It is supposed to make things dirty.

And the fact that it is advertised to give unity gain at 1 o'clock on the volume knob may mean it is designed to give more gain than a classic vibe... I think.

punchy

Thanks again @idy for taking a listen and your detailed response. I'll be honest in this context, as it was recorded, it does sound pretty clean and the noise in the background I can ignore. I'll try it again with my amp to see what it's like again and I'll capture that as well.

Just want to address a couple of your comments. The LDRs are a mess right now just because I recently replaced two of them, just in case they were a source of noise (it didn't make sense that they would be, and they weren't, but I wanted to eliminate that). When I'm done debugging it, I will trim the LDRs and place them facing the LED up close.

The sound clip of me probing each stage (going in order) is with a guitar signal as well, not artificial (same chain as the other clip, except I'm using a probe not the pedal's output jack). If it's expected that there would be some noise introduced at the third stage I would be really appreciative if someone could explain why that would be the case.

I also did recheck the resistors and caps the other day (thankfully there's not that many) and the values are good. I'll check the solder joints and reflow as best I can.

idy

#10
If the throb is at max that would help the third stage clip. It either gives stage two more gain or resonance, gain at one frequency.

Did you double check the color codes on "all those" 47k Rs in the phase section?

How about the number codes on the Caps? Are you sure you know how the number codes work?

duck_arse

here's a funny thing - because Q1 has its emitter grounded, its base will be at +600mV about. and because resistors R19 , R17 and RATE.2 all pull it to ground, C11 is shown backwards on the circuit diagram. even funnier, as C12 has grounding resistors at both ends, it's backwards no matter which way you install it [if electro/polarised]. neither point matters too much, but, well, y'no, these things. also, they won't be affecting the audio.

the real best way to ensure those resistor values and connections is with the Ohm-meter, measuring from each mt IC socket pin to socket pin. that way you test the connections and values are correct, not yust the colour bands. so IC1 pin 1 to IC2 pin 9, IC2 pin 9 to pin 8 etc along that chain.
I feel sick.

punchy

Thank you again to both of you for the help.

@idy I did double check all the resistors based on their color codes and they matched. I don't know how the number codes on the caps work but I used an online lookup tool to verify that they're correct.

@duck_arse, I haven't done your way of checking the resistor values but I will do so as soon as I can. A few days ago I did go with a multimeter through each resistor to measure the value and it looked good (there's not many spots where resistors are wired in parallel so it seemed easy enough to get a measurement this way without having to do any calculations).

I hooked up the pedal to my amp just to remind myself of the issue. My setup is guitar -> pedal -> front of JCM 800 -> Fryette Power Station -> cab. The power station has a balanced out so I used that to send to my audio interface to record the following two clips. Both clips are the same, just one is the raw signal from the power station and the other is with a cab sim IR applied (I find it easier to hear what's going on without the IR applied, but just in case it didn't seem like a good comparison I included both). The recording is clean sound -> turn pedal on and strum the guitar -> increase Voice knob during the silence in the middle up to 80-90% -> strum the guitar -> turn pedal off.

This reminded me that the noise was much more apparent with the Voice knob turned quite high, but not at max. I think you can hear that there's a spot (around 80-90%) where a swishy noise becomes very apparent. Even with the Voice knob turned to a much lower value, as in the initial clip, there still seems to be some distortion present in the signal.

Without cab sim: https://on.soundcloud.com/Ym7ox
With cab sim: https://on.soundcloud.com/XoUwd