Phantom Octave Troubleshooting

[MothGarden]

Hi, this is my first pedal that I am trying to build and I cannot seem to get it to come close to working after multiple attempts.

This is the layout I am using: https://tagboardeffects.blogspot.com/2013/04/fuzzhugger-phantom-octave.html

I am getting no output and the LED is dimly lit. Also to note is that according to my bench supply I am only drawing around 3 volts instead of 9. The way I am testing it may be causing the issues. I am using alligator clips to connect the wires coming off of the board to my inputs and outputs, although from my limited knowledge this shouldn't be the issue. I have tried just disconnecting the input completely to see if somehow that was causing issues and in that process I found that if I touch the input wire with my finger the led lights up brightly and the circuit starts drawing 6 volts. To my knowledge the LED shouldn't light up at all unless there is signal going through the pedal.

I have not made any changes to the original layout besides just extending the leads to the LED and combining to resistors in series to make 1 2.2m ohm resistor (I didn't have any on hand). I have tried to swap out the IC to see if that was causing the issue and I am having the same results.

My final guess is that it might be my soldering that is causing the issues. I'm not the best at it, but I have reflowed everything and checked over and over for accidental bridges with a magnifying glass.

I'm about ready to give up on this as I just cant seem to make any progress on building this pedal.

Attached below are some images of the build. Please let me know anything else I need to include.

[idy]

welcome to the forum

We never know for sure what a newcomer does or doesn't know. About electricity or soldering or....

So you are using a bench supply. It provides 9v when not "loaded" by the circuit?
But with the circuit attached you measure 3v?
That means a near short from power to ground. I hear you that you found no shorts on the rows; now use a meter (power off on circuit) and make sure there is no continuity around each cut. You can't tell by looking. Check again all the rows that that carry that 9v.

What voltage do you read on both ends of the diode the orange (+9v) wire goes to?
It should read 9v on one side, and about 8.4 on their other.
You should also see that voltage on pin 6 of the IC. and 0v on pin 4.

This is a weird circuit, some question about the diode orientation. I don't see how an LED with its + end attached to ground would light. But we can't see your LED so don't know if you have it that way. The layout shows the "flat" anode or - side going away from ground.

It is an unbiased opamp, so we would expect it to be "off" when no signal is present and to lurch into being a rectifier when signal starts....?

[MothGarden]

I solved the issue with the supplied power being way less than 9 volts. Somehow I accidentally limited my bench's current supply. We are now drawing 9 volts at around 200mA. The LED is now lit up brightly at all times, getting slightly brighter when something contacts the lead for the input.

Here are some of the voltages I am reading. I will also note that the IC is getting quite hot. This could be normal or could be a sign of something not very good.

D4: Reading 9.007v before and 8.001v after

IC:
Pin 1 - 0.825v
Pin 2 - 0.727v
Pin 3 - 0.000v
Pin 4 - 0.001v
Pin 5 - 1.473v
Pin 6 - 8.039v
Pin 7 - 4.061v
Pin 8 - 1.000v

The positive of my LED is attached to ground and is lit brightly when the circuit is on. There is a voltage drop of around 2 volts when the circuit is on and nothing is attached to the input lead, but when something is attached that jumps to around 2.6v.

I did a quick scan of all my cuts to make sure there was no continuity and I found nothing, which leads me to believe there could be a short elsewhere? Any idea where might be the best place to check given that information? If you need anything else let me know.

Thanks for the help and warm welcome!
 

[MothGarden]

Also as far as knowledge goes I feel like I have an okay-ish grasp on how electricity and circuits work, before I even started building pedals I would just read about how everything works. When it comes to physically building the pedals and troubleshooting, that's where I am having trouble. However don't be afraid to double check on some fundamental stuff, as you can see with that bench supply problem, sometimes I make silly mistakes!

[MothGarden]

Attached is a schematic I just tried to make based off the layout. Going to try to use it to help troubleshoot in the morning but thought I would post it just incase it has some really obvious mistakes. Ignore the extra lead coming off of the power supply, was drawing in pen and accidentally make a mistake.

[antonis]

200mA current draw and pin 5 low voltage call for heavy DC loaded amp..
(maybe due to a short across 100μF - the one between pin 5 and 47k/LED/D2/D1 junction..)

[duck_arse]

your linked page says the circuit is similar to this:


and your circuit drawing matches pretty well. the led would lite if you were injecting humm into the input with your finger, and the amp amped it way up. could happen.

and welcome to the forum.

[MothGarden]

200mA current draw and pin 5 low voltage call for heavy DC loaded amp..
(maybe due to a short across 100μF - the one between pin 5 and 47k/LED/D2/D1 junction..)

Checked for a short in this area and found nothing. Even took a knife and did a few light scores in-between the rails that the cap is across for good measure. Triple checked the values of everything as well. With everything plugged back in I'm still drawing 200mA and the blue LED is lit brightly. Considering plugging this circuit into a sim and just comparing voltages at every point to see where something might be going wrong, although I'm not 100% sure if this will help me. The amp being DC loaded seems like the potential issue. If that were the case would that cause the IC to heat up a lot?

[MothGarden]

and your circuit drawing matches pretty well. the led would lite if you were injecting humm into the input with your finger, and the amp amped it way up. could happen.

and welcome to the forum.

That was my guess as to the reason the LED was lighting up brighter, but it doesn't explain the LED being lit when there is nothing attached to the lead at all. To my knowledge the LED should be almost off when no signal is passing through.

[idy]

How is the LED anode getting one blue LED drop (3v?) below ground? I know opamps can sink current. But where is that current coming from?

[MothGarden]

How is the LED anode getting one blue LED drop (3v?) below ground? I know opamps can sink current. But where is that current coming from?

I have done some probing and I am getting around 1.4 volts coming out of pin 5 and that is going into a 100uF cap, and then on the other side of that cap there is -5.5 volts? This would explain the LED lighting up since its below ground, but I have no clue what is causing this. Gonna do some more probing and see if I can get any more info.

[MothGarden]

New Update. I took the IC out just to start probing around and I found that with the IC out (as in I'm probing the sockets) there is some stuff that doesn't make sense. Pin 6 is getting 9 volts which is expected. Pin 5 reads as -0.2 volts? And pin 2 is reading as 2 volts? If I'm reading the diagram correctly shouldn't both of those be 0v? This would explain the DC loading behavior, but I'm lost on where to start looking for these shorts. I checked all my cuts to see if anything was bridging and didn't find anything. Given this information is it possible to track down where the short is?

[idy]

with the IC out, probing where pin 5 goes, you might be reading a voltage in C3, 100uf. If you discharged that cap (low value R to ground) you might find 0 there.
Not sure why you are reading -v anywhere.
Maybe you are not getting a good ground for the black probe on the meter?
I don't see where 2v is getting to pin 2.

Again, we don't know how much you don't know. We can't see: Do the input and output wires go to jacks that have ground wires attaching to the black wire on the board and the ground from the power supply?

I have looked at your board vs the layout and didn't see misplaced components or missing cuts. It seems this circuit is the same as the "finger print" with its knobs maxed. And your schematic does seem faithful to the layout.

[MothGarden]

Again, we don't know how much you don't know. We can't see: Do the input and output wires go to jacks that have ground wires attaching to the black wire on the board and the ground from the power supply?

My black probe is attached to my power supply's negative terminal via an alligator clip and I am probing with the red probe. Input jack is grounded through the power supply as well. The output is just plugged into a breadboard which is holding my pot at the moment. I have the pot turned all the way down (as if the volume were on minimum). Ground on the breadboard is same ground as everywhere else. I looked pretty hard to find where this phantom 2 volts on pin 2 is coming from with no luck. I will have another go after class but I am thoroughly stumped. I think if I can't figure this out I may just throw in the towel and move on to trying a different pedal build, maybe come back to it later. But I am really determined to try and figure this out. Only error I found in my wiring so far was a 100nf cap that was incorrectly wired. It was one of the caps that filters the DC voltage in and I determined it was fine to remove for now. Didn't solve my issue though. Good thing is there is only so many components and this 2 volts has to be coming from somewhere.

Not sure why you are reading -v anywhere.

I'm gonna throw a stupid guess out there and maybe I am misunderstanding how inverting opamps work, but if pin 2 is getting this phantom 2 volts DC, couldn't it in theory be inverting that after the gain causing this negative voltage that is turning on my LED? This is a wild guess and I'm here to learn.

[idy]

An opamp can't go past its power rails. Yours are 9v and 0v.

If you were using a bipolar supply, with +9v and -9v and a ground at 0v, then yes, the opamp would turn 2v into -2v. But that is not your circuit. When we want an opamp to put out negative voltage, we need to feed it with bipolar; charge pumps (or timers used as charge pumps)m are the *only* way to make this happen from a single end supply.

Your work looks ok. You would maybe have better luck with a first circuit that is well known and easy to understand. Yours is deliberate abuse of the canons of normal design, not an example of how 99.9% of opamps are used. The common cases we "all" know what to expect. There are mysteries here...

[MothGarden]

You're totally right. I remember learning about that and completely forgot. I probably am going to come back to this one  in a little while. Although it didn't end up working out, I still think it was a good learning experience. I'm going to for sure take your advice and move on to something that is a little more well known, just so I can get one successful build under my belt. Thanks for all the help and walking me through some troubleshooting steps! I'm sure it will come in handy when I start trying to design some pedals of my own.

[antonis]

I DO insist that your audio power amp (not op-amp, as mistakenly called..) is heavily loaded...

P.S.
Did you take any measurements with IC out of its socket..??

[amptramp]

It is easy enough to understand the negative voltage at the ungrounded diode junction.  The capacitor going to the diodes prevents DC from coming through.  As the signal goes positive, it is clipped at the 0.65 volt diode drop of a silicon diode.  As it goes negative, it is clipped by the 1.9 volt diode drop of the LED (if it is red - 2.05 volts if it is green).  The junction sees the average of the signal if it is a square wave but still negative if it is a typical signal.

[MothGarden]

I DO insist that your audio power amp (not op-amp, as mistakenly called..) is heavily loaded...

P.S.
Did you take any measurements with IC out of its socket..??

Yes, pin 2 is being loaded by around 2 volts. Couldn't see where it is coming from though, but its there. Pin 6 is 9 volts of course and pin 5 is around -0.2 (I thinki).

It is easy enough to understand the negative voltage at the ungrounded diode junction.  The capacitor going to the diodes prevents DC from coming through.  As the signal goes positive, it is clipped at the 0.65 volt diode drop of a silicon diode.  As it goes negative, it is clipped by the 1.9 volt diode drop of the LED (if it is red - 2.05 volts if it is green).  The junction sees the average of the signal if it is a square wave but still negative if it is a typical signal.

Ahh, I see. Assuming I have my LTSPICE configured correctly it should like something like this (ideally). This is putting a 300mV amplitude sine wave at 440hz going through the circuit. It behaves very similarly to how you describe. This is the probe at the diode junction.

[idy]

Sorry Moth. I have spotted the difference before and should have known LM386 is power amp not opamp.

Also I was being stupid in not seeing why the LED was just an asymeteric clipper to ground, with audio going up and down from 0v. No mystery there.

Glad I didn't ruin your pedal or scare you away...