DOD 230 photocell

[Ronan]

Marcus, if you can give us the voltages at all the pins of the TL074, we should be able to sus it out, or at least point you in the right direction.

[mwynwood]

Thanks Ian This one has been bugging me for a while! It'll be my 11th build when it works 

Here are the voltages with no input and with a Red LED and just a jumper for the CELL

Power Supply: 9.67v

TL074CN:
1: 4.44v
2: 4.43v
3: 4.39v
4: 8.88v
5: 6.71v
6: 0.01v
7: 8.01v
8: 4.46v
9: 4.44v
10: 4.44v
11: 0.01v
12: 2.21v
13: 4.44v
14: 4.44v

Here's a bad, blurry photo 

[Pyr0]

The voltages on pins 5, 6, 7 & 12 don't look right. The 8v on pin 7 is going to keep that led full on.
Is pin 6 going to ground via that toggle switch ? What happens when that's the other way.
I'd also expect the voltage on pin 12 to be closer to 4.4v

[Ronan]

Given that pin 6 is at or near ground, then the voltages at pins 5 and 7 look OK to me.

The problem is that pin 6 is near ground, either through that switch, or because the transistor is fully turned on. If the transistor is the right type and in the correct orientation, then it must have 0.6 to 0.7V or more on the base to turn it on. With no input signal, the base should be close to ground via the 220K resistor (going from base to ground).

One question, does the board pass signal from input to output as it is, with a resistor in place of the VTL?

I finally found the BOM on that fancy website - sorry about that! Its under the BOM tab 

[mwynwood]

Ok, so I'm running it exactly as the layout, except for these changes:

R12 looks correct, but R13 should connect from R12 to ground (row 7 to row 10 instead of 7 to 9). Then R11 has to somehow connect to the TL074 pin 5 to pin 7.

I have a normal Red LED and just a jumper where the CELL should go (you can see it in the pic I posted before)

Here goes...

Is pin 6 going to ground via that toggle switch ? What happens when that's the other way.

Pin 6 changes between 0.00v and 0.14v when you flick the switch.

The problem is that pin 6 is near ground, either through that switch, or because the transistor is fully turned on. If the transistor is the right type and in the correct orientation, then it must have 0.6 to 0.7V or more on the base to turn it on. With no input signal, the base should be close to ground via the 220K resistor (going from base to ground).

The transistor is the one listed on the BOM.
I'm not sure which is the 'base', but orientated as per the layout, the 3 legs read
Top: 0.01v
Middle: 3.15v
Bottom: 0.01v

One question, does the board pass signal from input to output as it is, with a resistor in place of the VTL?

Yep, with a LED and a Jumper, it passes audio fine.

[Ronan]

Good to hear the rest of the circuit is working fine, passing audio.

The middle leg of the tranny is the base, it should have no voltage on it with no input signal from the guitar.
Even with a signal going in, it will be AC not DC.
R13 should no longer be connected to the base, so it would be worth trying to find out where that 3.15V is coming from, that is causing your problems at this stage. If you can't see or find anything obvious, it might be worth removing the transistor and then seeing if there is voltage at the base. If the voltage is gone, then try a new transistor, if the voltage is still there, you need to find out where it is coming from. If/when you remove the transistor, you should have around 9V at the upper leg (collector), 0V at the base, and 0V at the lower leg (emitter). That is, if all is well. If you don't get those readings, then let us know and we'll go from there.

[mwynwood]

Good to hear the rest of the circuit is working fine, passing audio.

The middle leg of the tranny is the base, it should have no voltage on it with no input signal from the guitar.
Even with a signal going in, it will be AC not DC.
R13 should no longer be connected to the base, so it would be worth trying to find out where that 3.15V is coming from, that is causing your problems at this stage. If you can't see or find anything obvious, it might be worth removing the transistor and then seeing if there is voltage at the base. If the voltage is gone, then try a new transistor, if the voltage is still there, you need to find out where it is coming from. If/when you remove the transistor, you should have around 9V at the upper leg (collector), 0V at the base, and 0V at the lower leg (emitter). That is, if all is well. If you don't get those readings, then let us know and we'll go from there.

I've just taken the transistor (2N3904) out and replaced it with a socket.
I have a red LED and a jumper in where the VTL5C10 is supposed to go.
With the transistor out, the LED stays ON.
Here are the readings at each hole of the socket with no input:

collector - top leg: 0.02v & 1.41v (switch)
base - middle leg: 0.02v
emitter - bottom leg: 0.02v

R13 (1M) is no going from "2 Along, 7 Down" to "2 Along, 10 Down" , which puts it from the 5th leg of the IC and at the bottom leg (emitter) of the transistor.

Thanks again for your help!

[Ronan]

What you have done is good. Looks like the R11 is causing the LED to be permanently on, because it is pulling pin 6 of the IC to ground, and it shouldn't be in that position on the vero - it is another mistake in the layout. Hopefully the last fix is - remove R11 (1 Meg resistor), it needs to be soldered between pins 5 and 7 on the IC. There may not be room on top of the pcb, if not, maybe solder it underneath on the copper side. Then the LED should be off with no signal, and anywhere between dimly to brightly lit with a guitar signal, and if so, try fitting the VTL5C10 and while praying to the pedal gods, apply power and a guitar signal and see what happens.

[mwynwood]

I've moved R11 so that it now goes from "6across-7down" to "6across-9down" which was a trace cut, but I put it through and bent it and soldered it on the IC side 

Still... the LED stays ON...

I'm wondering if I might need another trace cut there somewhere?

Anyway, it's late - I'm off to bed to ponder this layout and to pray to the pedal gods! haha

[mwynwood]

I've just drawn up how my build looks right now - in its "not working" state haha

[Harold]

Hi, I got a notice that my layout isn't working. I object to that "opinion"! 


DOD 230 Noise Gate

I forgot to update the description after building one: it works, but it's not very good ... It closes too late and abrupt, not anything like the iSP Decimator, which I cloned right after the DOD. I also own a original Coron Noise Gate (MXR Clone) which behaves the same, so I guess that's common with these type of gates.

IIRC, I still have a complete print somewhere ... Do you want me to check something?

[Ronan]

On the vero there is R13, a 1M resistor, connected to Q1's base. This does not match the schematic.

On the vero, R11, a 1M resistor connects pin 6 to gnd, does not match the schematic.

There is no 1M resistor between pins 5 and 7 of the TL074, feedback that locks the gate in or out.

There is doubt that a TL074 will deliver the output to ground that an LM358 can, which may or may not cause latch-up when power is switched on. I'm wondering if this is what is stopping Marcus' circuit now.

If the vero is correct then the schematic is wrong. Or otherwise I'm wrong. Any of these is a possibility

Marcus, I'll have a good look at this in the aussie morning, go from there.

[Ronan]

I can't see any problems with your layout Marcus. As far as I can see it should work. Can you measure the following resistances with the IC out and power disconnected?

pwr to gnd, should be around 44k
pin5 to pwr, should be around 500k
pin5 to gnd, should be around 500k
pin6 to pwr, should be around 220k
pin6 to gnd, should be around 264k
pin7 to pwr, should be around 1.5 meg ohms

Here's a schem showing TL074 pinout and component designators:


I breadboarded the opamp section driving the LED using a TL072, it worked fine.
Now that R11 and R13 are in the right places, could you measure the voltages on pins 5, 6 and 7 again, with the IC fitted, power on, and no guitar signal (as in guitar plugged in and guitar volume turned to zero - or just short the input socket).

[mwynwood]

I can't see any problems with your layout Marcus. As far as I can see it should work. Can you measure the following resistances with the IC out and power disconnected?

pwr to gnd, should be around 44k
pin5 to pwr, should be around 500k
pin5 to gnd, should be around 500k
pin6 to pwr, should be around 220k
pin6 to gnd, should be around 264k
pin7 to pwr, should be around 1.5 meg ohms

IC out, power unplugged, LED and Jumper in:

pwr to gnd, should be around 44k = 42.2k
pin5 to pwr, should be around 500k = 153k
pin5 to gnd, should be around 500k = 182k
pin6 to pwr, should be around 220k = no reading...
pin6 to gnd, should be around 264k = no reading...
pin7 to pwr, should be around 1.5 meg ohms = 1.14M

Something is weird there...

I breadboarded the opamp section driving the LED using a TL072, it worked fine.
Now that R11 and R13 are in the right places, could you measure the voltages on pins 5, 6 and 7 again, with the IC fitted, power on, and no guitar signal (as in guitar plugged in and guitar volume turned to zero - or just short the input socket).

Pin 5 = 6.70v
Pin 6 = 0.04v (1.43v with switch flicked)
Pin 7 = 8.00v (7.87v with switch flicked)

[Ronan]

All looks good except the 220k resistor R5 appears to be not connecting pin 6 to pin 4 of the IC. With the IC out, and no power, you should be able to measure 220k between pins 4 and 6. Until that happens, the circuit will not work.

pin5 to pwr, should be around 500k = 153k
pin5 to gnd, should be around 500k = 182k

Those measurements are fine if the switch was closed, sorry, I didn't specify the switch to be open...

R5 is somehow not connected at both ends, and that appears to be the problem. When pin 6 goes high voltage (maybe 8V give or take), then the gate will close. While ever pin 6 is below a few volts, the gate will always be open. R5 is needed to pull pin 6 high when there is no guitar signal, to close the gate.

[mwynwood]

It works! 
I traced back to R5 and found that it indeed wasn't going from Pin 6 to Pin 5 - fixed that, and not it's going!
My VCL5C10 has broken off a couple of legs because it was in and out of the socket so much... I have ordered a new one, but I found a VTL5C4/2 laying around and that seems to work fine.


Thanks so much for all your help Ian and others, and thanks for making the layout Harold - hopefully this thread helps out others with this one

Now, the next thing is that it adds a bit of volume and a slight amount of distorted fuzz compared to the bypassed sound.
It's kinda like a "boost/fuzz/gate" pedal at the moment.
Is there a way to tame the volume boost and remove the slight distortion?

[armdnrdy]

Ian, 

You deserve an award for perseverance!


mwynwood,

You should present the award!


Congrats to both!

[Ronan]

Thanks Larry, Marcus is very tenacious, I think he should get an award

Now, the next thing is that it adds a bit of volume and a slight amount of distorted fuzz compared to the bypassed sound.
It's kinda like a "boost/fuzz/gate" pedal at the moment.
Is there a way to tame the volume boost and remove the slight distortion?

If you increase R14 (10k) to 22k or 33k (just guessing) it should bring the gain back to unity. I'm not sure exactly what size to choose, it will have to be trial and error. Or temporarily fit a 25k or 50k pot in series with the 10k (R14) and find what value you need.

Be careful, if you fit a pot without the series 10k resistor, and turn the pot to minimum resistance, it will probably burn out the VTL5C4/2, and maybe damage the TL074. BTW, R14 will be different for different types of optocoupler, so if you swap out the VTL5C4 for VTL5C10 at a later date, you will probably have to readjust R14 for unity gain.

[mwynwood]

Thanks Larry, Marcus is very tenacious, I think he should get an award

Now, the next thing is that it adds a bit of volume and a slight amount of distorted fuzz compared to the bypassed sound.
It's kinda like a "boost/fuzz/gate" pedal at the moment.
Is there a way to tame the volume boost and remove the slight distortion?

If you increase R14 (10k) to 22k or 33k (just guessing) it should bring the gain back to unity. I'm not sure exactly what size to choose, it will have to be trial and error. Or temporarily fit a 25k or 50k pot in series with the 10k (R14) and find what value you need.

Be careful, if you fit a pot without the series 10k resistor, and turn the pot to minimum resistance, it will probably burn out the VTL5C4/2, and maybe damage the TL074. BTW, R14 will be different for different types of optocoupler, so if you swap out the VTL5C4 for VTL5C10 at a later date, you will probably have to readjust R14 for unity gain.

Thanks!

An 180k resistor in series with R14 brings it back to about the same volume when bypassed with the VTL5C10 - I guess this means I can replace R14 with a 190k resistor.

The only thing now is a slight distortion/fuzz/farty/clicky noise that is always there.
Have a listen here: https://soundcloud.com/mwynwood/dod-230-noise-gate-problem
This noise happens no matter how the switch is set, and obviously disappears when the whole effect is bypassed.

[Ronan]

Marcus, can you measure the voltages on all the pins of the TL074 again (with the switch open, and a guitar connected with the volume turned to zero)?  Just wondering, was there ever a time when this noise was not present? The sound clip is interesting...it seems to have a slow close and opening time, might be a good thing, I don't know, but there would be ways to modify that if required. That noise definitely should not be there. If you have a really strong magnifying glass it would be worth checking for dry solder joints, where it looks like the solder is not flowing onto a component leg freely.