aion refractor just built, but works only for 1min

[DragonSF]

I was so glad, that the pedal immediately worked with great sound. But, after some seconds the volume goes slowly down and after that no output signal anymore. I checked the signal flow and it seems the signal gets lost after pin1 of IC2. Strangely the voltage at pin 1 is 15V.
Looks to me as a bad capacitor, but which one? The voltages at the other ICs are fine. Any ideas?

[antonis]

Normallly, on pin 1 should appear a voltage of 5V (+ - )

It seems to me that op-amp is hitting the supply rail but it should only happen in case of "open" clipping diodes..
(if your signal tracing is OK till pin 2..)

P.S.
If your signal is lost just after R13, your switch is malfuctioning..

[DragonSF]

for some time (~1 min), everything is fine. After that, the signal is lost after pin 1 of IC2 as if the second op-amp lost the power.
Watching the signal at pin 7 shows the desired signal going down from full amplitude to zero within some time.
Just for the record: if I remove power for about 1min, the sound comes back again. AFAIK, the clipping diodes are connected to IC1, which is just working fine.

[antonis]

Just a reference to talk about...

Just for the record: if I remove power for about 1min, the sound comes back again.

This leads to supply problem - but you shouldn't measure 15V on IC2 pin 1.

AFAIK, the clipping diodes are connected to IC1, which is just working fine.

Maybe IC1 is working fine but diodes come after that and if they are "open" it may result in "square wave" distortion..
(meaning that IC2 will work as an ON-OFF switch, hitting the rails for every possitive & negative period..)

[duck_arse]

squinting at the circuit suggests to me that IC1 is only 9V powered, but IC2 is connected to the volts pumper. and you say you loooose signal around IC2? check the supply voltages while good, wait for it to go bad, and then tell us what's happening.

[PRR]

Lift one end of C10 and C11. Watch that U2 pin 1 voltage. Does it stay at 9V or go off toward +15V?

If happy without C10 C11, put C11 back, check. If OK, put C10 back.

Polarity is critical and not clearly shown on plan. The solid-black end of the cap is NEGative.

[DragonSF]

I checked the supply voltages and they are all OK and don't change, when output signal is going bad.
Regarding C9/C10 (coupling caps): the AION refractor calls for film caps (1uF), but I replaced them with electrolytic caps to no avail.
Removing C9/C10 clearly shows the same behavior (working for a couple of seconds and then goes to zero signal) at the output pin 7 of IC1.
So the problem is either related to IC1 or both.
I swapped ICs of course with new ones to no avail.
Further testing: checking  pin 6 of IC1 shows  a voltage below 3V when having the desired signal, but as soon it's over 3V (up to 4.4V), the signal goes low.
Just to make sure that the culprit is not the dual-gang poti (GAIN) I removed it completely. Behaviour at the output is the same, but Pin 7 of IC1 (6 as well) look fine now.
I also replaced the clipping diodes - no changes in the output signal behavior.
Another thing I noticed: VB is not constant: it varies with the gain control and jumps between turning forwards and backwards. Is that the way it's supposed to do? Range is like 4.0 -> 3.9 -> 4.1 -> 4.3 V.
I measured also the voltage at Pin1 U2:
it starts at 4.3 V and goes up to 15V within 3 minutes. After that the output signal is lost. Below 9V the signal is very good but get gets worse (amplitude gets lower) until 14.8V. Then no signal anymore.

[TejfolvonDanone]

Could you please list all the voltages at each pin of the two ICs and at Vb for about three position of the gain pot (turned all the way down, half way and all the way up) and measure the resistance between the two lugs of the gain pot? Obviously when the problem occurs. That would help a lot. Like this:

pin number (IC1) |	Gain 0 |	Gain half |	Gain max
1
2
3
4
5
6
7
8
Vb

pin number (IC1) |	Gain 0 |	Gain half |	Gain max
1
2
3
4
5
6
7
8

(click quote and you can copy the table)

Another thing I noticed: VB is not constant: it varies with the gain control and jumps between turning forwards and backwards. Is that the way it's supposed to do? Range is like 4.0 -> 3.9 -> 4.1 -> 4.3 V.

This only means that the current flowing out of Vb isn't constant through the rotation of the gain pot. It means about 30uA change from about 15uA to about 45uA.

[DragonSF]

pin number (IC1) |	Gain 0 |	Gain half |	Gain max
1	4.1	4.4	4.2
2	4.1	4.2	4.2
3	3.4	3.7	3.5
4	0	0	0
5	4.1	3.5	4.2
6	4.1	4	4.2
7	4.1	1.5	4.2
8	9.5	9.6	9.6
Vb	4.2	4.2	3.9

pin number (IC2) |	Gain 0 |	Gain half |	Gain max
1	16	16	16
2	3.9	3.8	4.15
3	4.1	4.4	4.2
4	-8.71	-8.67	-8.78
5	4	4.4	4.2
6	4.5	4.5	4.5
7	-7	-7	-7
8	17.1	16.98	17.1

Gain pot resistance: P1 from 0-100KB
P2 from 0-15kB (in circuit), outside 0-100KB
Just to make sure that the problem isn't the power supply part, I replaced the 1044 with an external power supply (using +18V and -9V). The behavior is exactly the same.

[TejfolvonDanone]

For me it looks like the bias currents mess up everything. I encountered an issue like this with my one op-amp overdrive.
Is the build breadboarded? Because we can suggest wilder ideas trying to fix it.

[DragonSF]

no, it's an Aion PCB. I also contacted Kevin about this this issue, but because of the time zone he's living in, there will be a delay. Anyway, I'll look into the bias voltage, I might nbe able to create it from a different source directly.

[slacker]

Something's wrong around pin 7 of IC2, it should be 4.5 volts. I would start by checking the component values and connections around that opamp stage, a bad connection  to pin 7 could also cause the problem so if you've used a socket check the opamp is seated properly.

[DragonSF]

For me it looks like the bias currents mess up everything. I encountered an issue like this with my one op-amp overdrive.
Is the build breadboarded? Because we can suggest wilder ideas trying to fix it.

That's possible. I replaced the Vb part with an external variable power supply and measured the same voltage at pin 1 of IC2 when set to around 4V. Having a lower setting (3V) gives 'normal' voltages at pin 1, but I can achieve the same effect by setting gain to 0.

[DragonSF]

Something's wrong around pin 7 of IC2, it should be 4.5 volts. I would start by checking the component values and connections around that opamp stage, a bad connection  to pin 7 could also cause the problem so if you've used a socket check the opamp is seated properly.

I replaced every component around pin 7 to no avail. I'm concentrating now on pin1, because after removing R20, still the voltage is 16V.

[anotherjim]

Removing R20 would I think cause pin1 to hit a power rail, It could be R20 isn't doing it's job, as if there's no DC feedback. Check your R20 value and prove it is well connected at the amp pins. Pin 7 is doing what I would expect with the high voltage on pin 1 ; driven to the negative rail.
If R20 is ok, you're looking for some voltage more negative than +Vb before either of R16,17 & 19. Don't get caught out worrying about small differences in V you measure - your meter is loading the circuit down some, and changing the gain pot may just be altering the effect meter loading has.

[DragonSF]

Removing R20 would I think cause pin1 to hit a power rail, It could be R20 isn't doing it's job, as if there's no DC feedback. Check your R20 value and prove it is well connected at the amp pins. Pin 7 is doing what I would expect with the high voltage on pin 1 ; driven to the negative rail.
If R20 is ok, you're looking for some voltage more negative than +Vb before either of R16,17 & 19. Don't get caught out worrying about small differences in V you measure - your meter is loading the circuit down some, and changing the gain pot may just be altering the effect meter loading has.

I replaced firstly R20 with different values and @ 47K the voltage at pin1 (7V) and Pin 7 (1V) seems more like it. Maybe I'm on to something.
After taht, I replaced R20 with a 500K trim-pot, but the result is strange: I can keep pin 1 and 7 within reason, but the gain is not as expected (i.e. lower at gain full open and higher at gain set to zero).

[slacker]

Assuming you've used a socket, if you remove IC2 and with it powered on measure the voltages on the socket do they look right. Should be about 4.5 volts on all except the the power pins, maybe a bit lower on some depending on your meter?
Apologies for my comments about pin 7 earlier, somehow I'd completely missed the problem with pin 1.
Have you tried a different opamp?

[DragonSF]

Without ic, the voltages look fine. No problem with pin 7 : I'm grasping at every straw I can get.Yes, I already tried 3 different op-amps. Same result for each, and all work fine as IC1.

[anotherjim]

It looks like R20 is doing its job.
If pin 1 is being driven too high, then an input to the amp must be some voltage below that on pin 3 (the reference +VB). Increasing R20 increases the gain, which makes it amplify an input voltage feeding an input resistor - this will NOT be measurable at pin 2, because the chip is trying, and succeeding, to make pin 2 the same voltage as pin3.

You will not find the faulty input on the chip pins when this happens, the chip is working properly. It's being told to drive the output high (positive). It's an inverting amp, so the control that's making pin 1 too high must be an input voltage too low (negative). The terms positive and negative refers to voltage above or below the reference, which is the voltage to the + input pin 3 = +VB.

You must measure the voltages feeding the input resistors to pin 2. There are 3 of those as I said before - R16,17 & 19. The ends of each resistor furthest from pin 2 is where you should measure. They should read nearly the same as +VB.

[aion]

At this point, I would get out the soldering iron and just touch every single joint with a fresh bit of solder to reflow. It's probably a bizarre cold/open joint or a bridge someplace, but one that's impossible to see visually.

I had a similar thing happen with an OD-820 (which also has a charge pump and dual gain pot). Everything was correct, no wrong parts anywhere, the soldering was very good (all joints shiny) - but it would stop working after a bit. After debugging for a couple of hours, I was considering just trashing it and starting over, but then on a whim I decided to just reflow everything, and after that it worked perfectly.

Sometimes random issues happen, and these can only be fixed by equally random repair attempts!