Flintlock Flanger not Sweeping

[bluebunny]

Not sure how to apply that to a pcb kit build.

Ah...  Dissecting tracks might be a bit tricky...

[duck_arse]

well let's wander through some circuit here. R7 and R8 divide a voltage from VR and provide a weak reference to beefy and strong IC2.2, which buffers it against all comers and calls that voltage VB. it will bash them up with this voltage, no arguments - that's why it's been buffered. so any point connected to VB must be hard hard hard VB volts, it can't waver or vary. those opamp pins should all read the same, VB [* note meter loading effect on resistors to (+) inputs *]. so, find !C2.2 output pin - I'm not going to try any more pin numbers, they are stupid small - and measure the resistance to each and every other point on the board that connects to VR on the circuit dia. this will prove nothing, unless you find some resistance ...... but will rule out bad connections at least.

and see - 9V comes in and gets messed about by IC11. it doesn't matter how badly that performs, because it gets regulated by IC7 to produce another of our strong-man voltages, the toughest of all, +15V. so you could again disprove problems by pulling IC11 and carefully providing 18V to the anode of D9. does VB still waver?

next I'd pull IC7 as well, and then carefully provide +15V at C38. any waverings sighted?

all this also applies to IC2.1 and its output, which appears at T, which connects to I don't know or care what, but that voltage should be strong and steady. is S C T a switch? that would make the output of IC8.1 "follow" [with gain] R39 or MANUAL. but the voltage at that end of the RANGE pot should be steady and strong, nevermind the oscillations at its other end.

and of course, if your opamps aren't what we think they are, you might get funny things happening. I dunno if any of this will get you flanging, but you might be at least able to say it's not the reference voltages.

[bluebunny]

As you were.  I actually looked at the schematic this time and I can see that the LFO does indeed have its own supply ("VA").

But does your IC4 (LM1458) pull as little current as the one specified (TL062) in the build doc?  It might yet be the source of the ticking.

[pacealot]

I can anecdotally add that replacing the LM1458 with either a TL062 or TL072 improved (different — mainly symmetry-related) issues I had with the LFO on my Flintlock build (same PCB), for anything that may be worth here...

(I seem to recall that the lectric-fx build docs somewhat cloud the issue of which chip to use for IC4, which is why I started with the 1458 myself...)

[matopotato]

If it's a power-environment related clock noise, maybe the 18V power supply trick could work. You just have to remove IC11 first.

First some "proof" that I tried to clean it up a bit (and @duck_arse likes pictures) :










The I did try the remove IC11 and run on 18V. I can see how it helps. The ticks became more like thuds, but it sounds wrong anyway.
Now the Manual pot had some kind of impact that was difficult to hear on 9V+IC11
At one moment the sweep changed speed by itself (yes, I know, it can't happen, but I'm just the messenger...)

Still it does not sound as I remember it, and not so "flangy" as I'd hoped. But I could be having the wrong expectations.
If you get a moment and feel very generous, I would appreciate some sound samples + notes or photos of knob positions. If possible some sample with extreme (?) flange, and some more musical. then I can compare.

Merci

[matopotato]

all this also applies to IC2.1 and its output, which appears at T, which connects to I don't know or care what, but that voltage should be strong and steady. is S C T a switch? that would make the output of IC8.1 "follow" [with gain] R39 or MANUAL. but the voltage at that end of the RANGE pot should be steady and strong, nevermind the oscillations at its other end.

I will try to test the other things you suggested. Just wanted to clarify: S is sleeve, T is tip and C connects to "ring" of a stereo jack. Used for expression pedal.
Those points are used when calibrating frequencies.

EDIT: I will try to dial in a setting where the "main" voltages do not sweep and take new readings. It still seems strange that voltage fluctuates in places where it should be steady, but maybe finding a more stable condition to do reading could point somewhere and then look into the fluctuations later if they remain after any first adjustments.

[matopotato]

I can anecdotally add that replacing the LM1458 with either a TL062 or TL072 improved (different — mainly symmetry-related) issues I had with the LFO on my Flintlock build (same PCB), for anything that may be worth here...

(I seem to recall that the lectric-fx build docs somewhat cloud the issue of which chip to use for IC4, which is why I started with the 1458 myself...)

Thanks,
I had a spare TL072CP. I put it in, and the ticking stopped, but so did the sweep.
I put the LM1458 back in and still no sweep.
So fearing the second 4047 was toast I put the 4047 back in the Countdown Phaser and it does sweep and sounds as it should. So thankfully the 4047 is ok.
I put it back in the Flanger and still no sweep, but with some mild pressure on the LM1458 it comes back. So I will try to reflow the socket see if that makes anything better.

EDIT: I reflowed the socket. And verified continuity to the nearest components. Then with the LM1458 in it was still a bit wobbly, so I "adjusted" the legs a little bit and not it sits in the socket and there is sweeping. Still got the thump-clicks though.
Re-empowered I tested again the TL072 with a bit more spread legs, and it works but the thump-click is still there.

[matopotato]

...
and see - 9V comes in and gets messed about by IC11. it doesn't matter how badly that performs, because it gets regulated by IC7 to produce another of our strong-man voltages, the toughest of all, +15V. so you could again disprove problems by pulling IC11 and carefully providing 18V to the anode of D9. does VB still waver?

next I'd pull IC7 as well, and then carefully provide +15V at C38. any waverings sighted?

I breadboarded a voltage divider trimpot setup, fed it 18V, took out 15V, then pulled IC7 and IC11. But there were only bypass. No effects signal, LED does not come on, but not sure I expected that. So can't say if waverings are there.
Should it be fed 9V the usual way as well? If some other point is subscribing to that and I took out those two ICs?

I will try to measure for any resistance between the various Vx points. Not sure if you meant VBs VRs or all of them. But might as well measure all.

[matopotato]

well let's wander through some circuit here. R7 and R8 divide a voltage from VR and provide a weak reference to beefy and strong IC2.2, which buffers it against all comers and calls that voltage VB. it will bash them up with this voltage, no arguments - that's why it's been buffered. so any point connected to VB must be hard hard hard VB volts, it can't waver or vary. those opamp pins should all read the same, VB [* note meter loading effect on resistors to (+) inputs *]. so, find !C2.2 output pin - I'm not going to try any more pin numbers, they are stupid small - and measure the resistance to each and every other point on the board that connects to VR on the circuit dia. this will prove nothing, unless you find some resistance ...... but will rule out bad connections at least.

I think I went through all
VA, VR, VC and VD show 0.2 or 0.3 ohm between their points
And for VB most of them do as well,
But where R18 is involved I get fluctuating resistance. R5 to R18 shifting from 0.2 to 13 ohm
And R21 to R18 6-16 ohm.
Also R5 to IC4 pin 5 was jumping around a lot, but I also got it to settle at some point, so that one is more likely me not getting the probing pins right.
But for R18 I felt it quite conclusive. Tries several times. So I will resolder it and see if that does anything. The R18 itself measures a stable 68k (ish), so the R is probably ok (IMHO).

Unless there is someone who could tell if there might be adifferent reason for this fluctuation. Not sure if IC1 or IC3 is most affected by this anomaly. If 13-16 ohm is an anomaly in this case...

[ElectricDruid]

I wouldn't read anything into a few ohms difference in readings, even 0 to 13. That can be explained by how hard you're holding the probes on the pads while you're doing the readings. It doesn't have to mean anything about the actual board itself.
The important bit is that those points are joined by a low resistance, which means the track is generally intact and the solder joints are generally good!

[eh la bas ma]

-Edited- (didn't read your last posts)

[eh la bas ma]

I would appreciate some sound samples

Here is a short demo.

I swapped the LM1458 for TL72 as suggested by Pacealot. Not sure if I notice any change, but it didn't hurt.
On the other hand, I noticed that my builds were a bit more noisy when the BF-3 was connected after them. Some white noise appeared at the top of the sweep, maybe it's the buffers in the BF-3, or something with the electrical environment. Putting the BF-3 before both Flintlocks solved it.

First 2 minutes it's one Flintlock alone, then I briefly compare with Boss BF-3. Finally, I conclude with 2 Flintlocks modulating the signal, demonstrating that 2 Flintlocks are working great together. That's why I'd humbly suggest to build an other one, as soon as you finish your first Flintlock unit : L'union fait la force.

Control settings in description.

https://soundcloud.com/ehlabas-ma-chaussure/flintlock-demowav

It looks much better, but there're still some suspicious pads :

[matopotato]

I wouldn't read anything into a few ohms difference in readings, even 0 to 13. That can be explained by how hard you're holding the probes on the pads while you're doing the readings. It doesn't have to mean anything about the actual board itself.
The important bit is that those points are joined by a low resistance, which means the track is generally intact and the solder joints are generally good!

Thanks,
Ok, I understand. But it was a flucuating resistance 0.2 to 13-16 ish that felt more strange than just a somewhat higher value. I felt I pressed similar as the other checkpoints, but I am aware there could be an effect of varying values depending on how I hold on to propes etc.
So in some measures the swimg was even bigger.
Still I might resolder it for my own peace of mind and "well, at least I tried that"
The cycle thump is still there and the flanging sounds as if it is lacking something.
But I will compare with the sound samples below.

[matopotato]

I would appreciate some sound samples

Here is a short demo.

I swapped the LM1458 for TL72 as suggested by Pacealot. Not sure if I notice any change, but it didn't hurt.
On the other hand, I noticed that my builds were a bit more noisy when the BF-3 was connected after them. Some white noise appeared at the top of the sweep, maybe it's the buffers in the BF-3, or something with the electrical environment. Putting the BF-3 before both Flintlocks solved it.

First 2 minutes it's one Flintlock alone, then I briefly compare with Boss BF-3. Finally, I conclude with 2 Flintlocks modulating the signal, demonstrating that 2 Flintlocks are working great together. That's why I'd humbly suggest to build an other one, as soon as you finish your first Flintlock unit : L'union fait la force.

Control settings in description.

https://soundcloud.com/ehlabas-ma-chaussure/flintlock-demowav

It looks much better, but there're still some suspicious pads :

Many thanks!
Will dig deep soon and compare.

At least one of the areas marked is just some non metallic speck of dirt. Will temovr anyway and recheck those points for any unwanted bridges.

[matopotato]

I wouldn't read anything into a few ohms difference in readings, even 0 to 13. That can be explained by how hard you're holding the probes on the pads while you're doing the readings. It doesn't have to mean anything about the actual board itself.
The important bit is that those points are joined by a low resistance, which means the track is generally intact and the solder joints are generally good!

Not sure I understand how the resistance varies without and power source connected. Only thinknI can think of is a half broken solder joint.

[antonis]

Not sure I understand how the resistance varies without and power source connected.

Flux residuals and probes tips pressing force..

[duck_arse]

I breadboarded a voltage divider trimpot setup, fed it 18V, took out 15V, then pulled IC7 and IC11. But there were only bypass. No effects signal, LED does not come on, but not sure I expected that. So can't say if waverings are there.
Should it be fed 9V the usual way as well? If some other point is subscribing to that and I took out those two ICs?

I will try to measure for any resistance between the various Vx points. Not sure if you meant VBs VRs or all of them. But might as well measure all.

no sir, voltage divider setup is no good. as soon as you try and pull current thru it to power the circuit, the voltage will drop and nothing happens. you want to feed in 18V so the 15V regulator can do its overhead thing and regulate properly. the point of the excercise was to move along the power supply, ruling out each part as you go. you feed 9V to the upvolts IC, but once you pull that out, you need to be feeding 18 external Volts to the point it was supplying. and then when you've pulled the regulator, your external volts have to do that work of suppling the 15V [ missing from pulled regulator].

VB was the only points I was interested in. anything/everything connected to VB. contrary to other's posted opinions, I think 13 ohms does matter. any VB pad-track-VB pad measure should be 0R and steady, with not even capacitor dischargings affecting the steadies. note - I've said pad to pad, NOT resistor leg to IC pin. and that might be an indicator -  measure from pad to pad on the copper side [don't tell me double sided, please!]. then measure from the solder on the pad copper-side to the component leg or IC pin components side. you were showing funnies with socket pins earlier, no? we want to disprove the board pads and traces, then the connections to the pads.


with all due respect of course.

[matopotato]

I breadboarded a voltage divider trimpot setup, fed it 18V, took out 15V, then pulled IC7 and IC11. But there were only bypass. No effects signal, LED does not come on, but not sure I expected that. So can't say if waverings are there.
Should it be fed 9V the usual way as well? If some other point is subscribing to that and I took out those two ICs?

I will try to measure for any resistance between the various Vx points. Not sure if you meant VBs VRs or all of them. But might as well measure all.

no sir, voltage divider setup is no good. as soon as you try and pull current thru it to power the circuit, the voltage will drop and nothing happens. you want to feed in 18V so the 15V regulator can do its overhead thing and regulate properly. the point of the excercise was to move along the power supply, ruling out each part as you go. you feed 9V to the upvolts IC, but once you pull that out, you need to be feeding 18 external Volts to the point it was supplying. and then when you've pulled the regulator, your external volts have to do that work of suppling the 15V [ missing from pulled regulator].

VB was the only points I was interested in. anything/everything connected to VB. contrary to other's posted opinions, I think 13 ohms does matter. any VB pad-track-VB pad measure should be 0R and steady, with not even capacitor dischargings affecting the steadies. note - I've said pad to pad, NOT resistor leg to IC pin. and that might be an indicator -  measure from pad to pad on the copper side [don't tell me double sided, please!]. then measure from the solder on the pad copper-side to the component leg or IC pin components side. you were showing funnies with socket pins earlier, no? we want to disprove the board pads and traces, then the connections to the pads.


with all due respect of course.

Again many thanks,
I understood the general idea.of following the rail but did not know my 15V design would fail.
I can retry 18 straight in to the 18V point, but 15V I do not know how to generate in my limited.home lab.
As for the VB measures, I will redo the wobbly ones on pad to pad level although the pads are either covered in solder on the one side, and the component side is not the same I guess if this is multilayer?
With my limited experience and understanding I believe that there are traces visible on either side so that would make it multilayer. I dare not write the d-word above since I really appreciate your help and support on this.
I understood the quest as to find the traces between VB points in the schema, and measure R closest road possible. I hope it did not mean to remove solder to reach each pad though.

Based on my own unscientific finding (R is wobbling in areas close to ICs where voltage is wobbling (?) and where there should be a wobble-free zone instead) to try and swap around IC1 and IC3 with say IC2 respectively.
Even if I did the wrong investigation (yes, I confess, I DID measure pin to leg. I will have to accept my punishment by my peers [although most in here are.on a much higher skill level than myself, so from higher tier peers...])
I still came across varying degrees of wobbling R values. And in an unpowered, waiteduntilCapswerehopefullydischarged circuit it still feels wrong. If me pressing so differently with them probes I'd had expected wobbling to varying degree all over the place. But the wobble location vs other VB locations was consistent.
I might take a plunge at some point and resolder/reflow R18 anyway despite the risk of doing the wrong or unnecessary thing.

[matopotato]

Ok, I am trying out a few things.
There is a wobbling R between R5-R18 compared with other points. Can't see I am pressing any different in different locations.
Then I went to verify where to put 18V straight in. If I understood this, the +9.18V is set and passes cathode of D7 to anode of D8, still stable, after the cathode of D8 there is voltage variations thereafter.
I would have thought the D8 being dodgy, but it might be some other component after that makes the swings carry over to D8 anyway. C33 is still on 4.42V at negative side, but swings around with D8 cathode in the 12V area´. Then on the other side of D9 it swings around 16V, which seems to be where it should be, only with steady voltage.

Then I notice with Range fully CCW there is no swing, also Speed has an effect on the swing, but not as clear.
So now finger is pointed towards IC4 being the one down in that corner. Here I already re-flowed the socket and made the IC legs find better contact.
Aha, I got an MC1458P spare to try, still swings.
But if I bring gentle pressure to the IC in its socket at pins 1 and 8, the swing stops...
Switch back to LT1458, and again it stops.
So now I am beginning to suspect the socket as such being strange. It certainly holds a looser grip than usual, so this is now suspect #2.
I then find a bit more elevated position of IC4 where I do not have to exert pressure where the swing stops.
So I plug it in with guitar and amp, and now there is no Flange anymore, as when I started the thread.
And voltage seems stable though. Then I fiddle a little bit with the IC4 LT1458 and find a place where the flanging is there (I think), but with the thumping, clicking at each cycle. But thankfully the voltage is stable on the 15 TP.

So now at least I have a way of getting rid of the variation in voltage. The resistance between the VBs at R8 and R18 was also stable and 0.2 when the IC4 positioned so that there was no swinging voltage.
I still kind of suspect the socket for IC4 to be an issue of sorts, but unlikely to cause the clicks.

[matopotato]

I would appreciate some sound samples

Here is a short demo.

I swapped the LM1458 for TL72 as suggested by Pacealot. Not sure if I notice any change, but it didn't hurt.
On the other hand, I noticed that my builds were a bit more noisy when the BF-3 was connected after them. Some white noise appeared at the top of the sweep, maybe it's the buffers in the BF-3, or something with the electrical environment. Putting the BF-3 before both Flintlocks solved it.

First 2 minutes it's one Flintlock alone, then I briefly compare with Boss BF-3. Finally, I conclude with 2 Flintlocks modulating the signal, demonstrating that 2 Flintlocks are working great together. That's why I'd humbly suggest to build an other one, as soon as you finish your first Flintlock unit : L'union fait la force.

Control settings in description.

https://soundcloud.com/ehlabas-ma-chaussure/flintlock-demowav

That sounds so much more flange than mine even when I try to think away the thumps.
Hope to get there.
Very nice thanks, makes me want to fight on.