Page 1 of 2

bandpass filter in a stompbox

Posted: 08 Apr 2018, 12:12
by snk
Hello
Hello
I have built a stompbox version of the Roland SH5 bandpass filter.

The assembly went fine, but no sound comes out from the unit :cry:
I suspect a wrong wiring (as i suck at this, and as i think i took care to triple check every soldering, and tested every resistor and the diodes before soldering, etc), as the unit was designed to be built for a modular and not as a stompbox.

Would anyone give a look at my wiring schematic, and tell me if it is correct, please ?
The first one is the "offboard/workbench" version, and the second one is the "final" version, using a volume pot and a 3PDT, which i am planning to do once the circuit is verified and known to work.
(fwiw, at the moment, the daughter board is not installed, i am using a 15V power supply)

Image

Image

Thank you !

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 14:44
by YellowBoy
Wiring looks fine. I'm more concerned about your power supply. This circuit calls for a bi-polar supply...hence the V+, GND and V-.

Is your 15V supply bi polar?
Is it DC or AC?

If it's unipolar then you need the daughterboard. If its AC then you need to rectify. Finally if it is indeed a bi-polar DC supply then you may very well have a problem with the build itself or an erroneous layout (but looks like it was verified by the designer)

HTH

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 16:44
by snk
Hi,
The circuit is designed for a symmetrical 15V power supply, but said to be working with +/-12V as well.
I have both power supplies : a DC 15V bipolar ( +/-) power supply (which i used for the testing, without the daughter board), and a DC12V uniploar power supply (in this case, i will connect the daughter board to get a bipolar PSU).

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 18:14
by YellowBoy
I missed this bit...so it was working before you boxed it? In that case the problem may be something silly then. You're wiring diagram works so maybe you made an error when you were wiring it into a box, something is now shorting against the box or a wire came loose/broke when soldering in the box.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 19:23
by snk
Hi,

No, i haven't boxed it yet : it's still out of the enclosure, and the 12V daughter board is not wired : for testing, i used a 15V power supply (as the circuit is expecting 15V, and as it is easier to spot an issue with only one single board).
My concern is that the circuit seems quite easy, there aren't a lot of components, i checked every resistor value before sodering them -as well as the diodes-, so i was somewhat hoping that it would be only a wire wired incorrectly... but it seems it's not that :hmmm:

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 20:09
by snk
Here are pictures of the build.
It doesn't seem too messy to me.
I noticed i forgot a 10K resistor, which was not shown as clearly as the others (it was half transparent). I added it to the circuit, but i still don' t hear anything coming through.
By looking again at the veroboard the author posted on his website, it seems there are some slight differences between his schematic and his build (

Image

Image

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 20:22
by YellowBoy
Cool. I'll have a look now. In the meantime can you post voltages from all the important areas? IC pins, v+, gnd, v- etc.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 21:31
by snk
Cool. I'll have a look now.
Thank you !
can you post voltages from all the important areas? IC pins, v+, gnd, v- etc.
Grrr ! I just broke the red arm of my DMM :evil:
I could only begin to make the measurements, so here they are for the first IC (TL072) :
from up to down, starting from the left row :
0
0
0
+15.15

right row (from up to down) :
+15.12
-13.5
+0.29
-1.08

I could only measure +15 on the upper right pin of C2, when the red arm of my DMM broke... I tried to solder an alligator clip instead, but i fear it's not that reliable anymore (i got weird jumpy values when i tried to measure IC2).

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 22:00
by YellowBoy
Sorry to hear! Always when you need them eh?

For future reference, IC pins are labelled as follows

1-8
2-7
3-6
4-5

...it'll make for better understanding if this convention is followed.

Is pin 4 really at +(positive) 15V? It should be at -(negative) 15V. I suspect your pins 5, 6 and 7 are off too. I've attached a a pic of your solder side (I've mirrored it to match the component side) There looks like a few solder bridges (marked in blue) and an extra cut (marked in yellow). Try looking at/fixing those (....and your multimeter lol) and let us know how you get on.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 22:15
by Nocentelli
Eagerly following this thread as it sounds like a great filter.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 22:31
by snk
Eagerly following this thread as it sounds like a great filter.
Yes, i do think so :mrgreen:
Welcome aboard !
I've attached a a pic of your solder side (I've mirrored it to match the component side) There looks like a few solder bridges (marked in blue) and an extra cut (marked in yellow). Try looking at/fixing those (....and your multimeter lol) and let us know how you get on.
Wonderful, thank you !
The hole shouldn't have been there indeed : it's fixed now.
One solder bridge was in fact a marking made by my pen for spotting the holes to drill.
I have cut the other pins which might have made a solder bridge.

Sadly, I still don't get a sound (and my DMM's red cable is broken, so i can't provide any further measures) :cry:

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 22:37
by microbailey
Hi snk.
I posted a reply - but it disappeared immediately :shock:

Anyway, looking at your wiring your pots are wired up wrong.
You've got "F1 end" wired to F2 pin3, and "F2 end" wired to F1 pin 3. So they're swapped.
That's not going to work!

Also note that F1 and F2 are two parts of the same dual-ganged pot, so they change in unison.
I hope you've used a dual-gang pot for this because the filter won't operate correctly if F1 and F2 change by different amounts.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 22:50
by snk
Hi, MicroBailey
Thank you !
Anyway, looking at your wiring your pots are wired up wrong.
You've got "F1 end" wired to F2 pin3, and "F2 end" wired to F1 pin 3. So they're swapped.
That's not going to work!
That was the first time i read a pot's pin called "end", so, as a pot has 3 pins, i thought "end" meant "last" (and therefore the 3d pin) :slap:
[edit] Ah, you mean that i have wired to F1 a wire which should have gone to F2 ? I am not sure, as from the schematic, it seems that it's supposed to be like that : from left to right : F2pin3, F1pin1, F1pin2, F2pin2, F2pin1, F1pin3... I think it's what i did (it surprised me when i was reading the layout, so i tried to stay focused when doing the wiring, eh eh), but i can try differently. Do you think it might be a mistake from the layout ?
Also note that F1 and F2 are two parts of the same dual-ganged pot, so they change in unison.
I hope you've used a dual-gang pot for this because the filter won't operate correctly if F1 and F2 change by different amounts.
Yes, i am aware about that : i have ordered a dual ganged pot, but it hasn't arrived yet. In the meantime, i was very curious to try it that way and hear how it sounds ;)

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:01
by snk
Ok, i soldered an alligator clip to the red cord of my DMM. I can't guarantee the readings are accurate, but hopefully it can help.

So here are the results for pins 4 & 8 (for IC1, the one on the left):
pin4 = -15.13
pin8 = +15.13

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:13
by YellowBoy
So far so good. Post measurements for all the pins on both ICs. It might give us an idea of where the problem is.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:15
by snk
I made new measurements of each IC (with the alligator-clip-repaired DMM, so accuracy is not 100% guaranteed. However, i double-checked, and it gave me each time the same values) :

IC1
pin1 : -0
pin2 : -0
pin3 : -0
pin4 : -15.13
pin5 : +0.68
pin6 : -1.11
pin7 : +11.34
pin8 : +15.13


IC2
pin1 : -13.65
pin2 : +3.25
pin3 : 0
pin4 : -15.13
pin5 : 0
pin6 : 0
pin7 : +2.74
pin8 : +15.13

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:27
by microbailey
snk wrote: Ah, you mean that i have wired to F1 a wire which should have gone to F2 ? I am not sure, as from the schematic, it seems that it's supposed to be like that : from left to right : F2pin3, F1pin1, F1pin2, F2pin2, F2pin1, F1pin3... I think it's what i did (it surprised me when i was reading the layout, so i tried to stay focused when doing the wiring, eh eh), but i can try differently. Do you think it might be a mistake from the layout ?
I was just looking at your diagram at the start of the post and it shows a wire coming from the stripboard called F1 end which is going to the pot which is called F2 (and vice versa). Based on that it looks swapped to me, but I haven't checked versus the SH-5 schematic.

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:33
by snk
Oh, i see : you're right, but i think it is only a mistake on my diagram. I don't think i made the same error on the actual build, but i will check once more for safety ;)

Re: bandpass filter in a stompbox

Posted: 08 Apr 2018, 23:35
by microbailey
Those voltage readings - IC2 pin 1 at -13v doesn't sound healthy.
Pin 1 is an op-amp output (if I'm understanding the IC numbers versus the SH-5 schematic correctly) and I would expect the output to be close to mid-point of the +/- voltage rails.
Check for either a short to something on either pin 1 or pins 2 or 3 in IC2.
Also if those pot connections are swapped it might do something crazy to the op-amp inputs.

Re: bandpass filter in a stompbox

Posted: 09 Apr 2018, 10:42
by snk
Hi,
I will investigate the Opamp's pins.

I have made new measurements, and I also swapped the TL072, to check if one opamp could be faulty : it gave me the same measurements.

IC1
-0
0
0
-15.13
+0.13
-1.33
+10.88
+15.13

IC2
-13.63
+1.64
-0
-15.13
-0
-0
+1.86
+15.13

I have corrected the diagram, and this is actually the way i have wired it.
Image