MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

ST Modular Karl filter
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author ST Modular Karl filter
hw408
Hi everybody
I am stuck on this build and I have the schematic, but not sure what is happening. I dont want to post any schematic because it was sent to me privately. Im wondering if anybody on muffs has built or is presently building the Karl filter by ST Modular.

On the chance you are familiar..


There is a LP_L input and a LP_R input.

Each one goes through a resistor (R1 for LP_L and R16 for LP_R), to its own 13700, into the + inputs.

With a scope, I am finding my signal at the input jack side of the first resistor (R1 for LP_L and R16 for LP_R), and no signal on the IC side of the those same resistor(s).

Someone said maybe there is a signal but it is just extremely tiny - im not sure.
But the same thing is happening on each channel, going into different ICs.

I reflowed every piece, the ICs have power, led lights, no smoke, diodes correct polarity. I checked continuity point to point on the LP_L path, and the board seems to do what the schematic says. The co figuration jumpers are set to "full low pass" (1,2,4 and 6 jumpered).

I bought enough parts to build a second one but i am getting frustrated.
I know nobody can guess without a schematic so I am basically venting/fishing.

Has anybody finished a Karl?
Thanks for reading!
cygmu
I don't have this module or the schematic, but for what it's worth...

A common way to set up the LM13700 input is to attenuate the input signal via a voltage divider, by about a factor of 500. A typical setup might be 100k input resistor and 220R resistor to ground, though the details vary between implementations of course. So the theory that your signal is still there but is very small -- tens of millivolts -- fits well with a typical schematic. You should be able to see this on the scope, though, if you set the scale appropriately.
guest
is there signal anywhere else on the board?

what are the voltages on pins1,16 of the LM13700s? they should be ~1V.
cygmu
guest wrote:

what are the voltages on pins1,16 of the LM13700s? they should be ~1V.


I don't have the schematic so it is hard to disagree with this but it seems odd to me -- I'd have expected the LM13700 to be powered from +/-12V so you'd get about -11V at the control pins. Is it powered from 12V and 0V in this circuit?
hw408
cygmu wrote:
I don't have this module or the schematic, but for what it's worth...

A common way to set up the LM13700 input is to attenuate the input signal via a voltage divider, by about a factor of 500. A typical setup might be 100k input resistor and 220R resistor to ground, though the details vary between implementations of course. So the theory that your signal is still there but is very small -- tens of millivolts -- fits well with a typical schematic. You should be able to see this on the scope, though, if you set the scale appropriately.


Hi Cygmu thanks for the reply.

That sounds similar to this:

100k input resistor (R1) from the jack
meets with both
a) pin3 (+ input) 13700
b) 470R (R2) to ground

When i turn the scope's v/div up 4 or 5 clicks (to max) i do see a tiny, blurry wave that follows my signal generator.

interesting discovery ty
hw408
guest wrote:
is there signal anywhere else on the board?

what are the voltages on pins1,16 of the LM13700s? they should be ~1V.


pin 1 = -10.5v
pin 16 = -10.5v

When i put my fluke probe on them i temporarily hear the signal from the output, but it is not affected by Karls filter knob.

power pins i believe are
pin 6 (-V)= -11.29v
pin 11 (+v)= 11.29v

I will poke around some more looking for that extremely tiny signal I am now aware of.

Edit:
The output of 13700 pin 5 has -2.3v which does not seem to change when i vary the amplitude of the input signal
cygmu
hw408 wrote:

That sounds similar to this:

100k input resistor (R1) from the jack
meets with both
a) pin3 (+ input) 13700
b) 470R (R2) to ground

When i turn the scope's v/div up 4 or 5 clicks (to max) i do see a tiny, blurry wave that follows my signal generator


If you know the scale that you are looking at you can figure out whether this is correct. The wave you see there should be about 1/200 of the amplitude of the input wave. If it is significantly smaller than that, there's an issue around that area. If it matches expectations then that section at least is working ok and we can look at the next stage.
guest
thanks cygmu! its supposed to be 1V off the bottom rail. the problem is with the bias current generator (CV to current protion that feeds pins1,16). putting the meter there allows a small amount of current to flow, and turns on the amplifier.

follow pins1,16 and see where they go. they most likely go through a resistor to a transistor. check that those transistors are the correct type and in the right orientation. what are the voltages on all 3 pins of the transistors?
hw408
guest wrote:
thanks cygmu! its supposed to be 1V off the bottom rail. the problem is with the bias current generator (CV to current protion that feeds pins1,16). putting the meter there allows a small amount of current to flow, and turns on the amplifier.

follow pins1,16 and see where they go. they most likely go through a resistor to a transistor. check that those transistors are the correct type and in the right orientation. what are the voltages on all 3 pins of the transistors?


Sure enough, pin 1 of 13700 goes through a resistor and first through Q1 collector and then Q2, both 3906 transistors joined at the emitter, and out Q2 collector.

Q1:
E= 0.56v
B= +/- 0.003v
C= -10.3v

Probing the Q1 collector lets some noise out the output, not affected by cutoff pot.

Q2:
E= 0.64v
B= +/- 0.003v
C= 0.001v

Both Base measurements bounce around 0 but not stable. I reflowed one solder ball on one of these.
guest
turn the cutoff knob full up, and measure the voltage on the base of Q2 again (should be 200mV or so). then measure the voltage on either side of the resistor that goes to Q1.
guest
also, check the voltage on either side of the resistor that goes to the collector of Q2, and make sure that resistor is the right value. if you have a clear photo of your board, that can sometimes help find the problem.
hw408
guest wrote:
turn the cutoff knob full up, and measure the voltage on the base of Q2 again (should be 200mV or so). then measure the voltage on either side of the resistor that goes to Q1.


Cutoff knob full CW

Side note - when i probe Base of Q1, i hear the effect of the filter.
Sound goes away when i remove the probe. I think i am getting two versions of audio, depending on how hard i push the probe, i wonder if if this is not a good connection.

Q2 with cutoff full CW
B= 0.114v

Q2 with cutoff full CCW
B= -0.110v

Resistor R32 that feeds Q1= -10.5v regardless of cutoff pot

I reflowed the Base of Q1 and now there is sound. Wow i think that was it, i have cutoff control and (left) resonance control too. This is on the Lowpass_L (left) input. Effected sound comes out the Left and Mix outputs.

tears of joy waah

Switching the input to the Right jack, there is no soun, but I will focus on the second 13700 with scope and see what happens. The Q1/Q2 seems to be a shared cutoff control.
At least i have a working left side LP circuit to compare the right side to.

we're not worthy you gave me hope cant thank you enough SlayerBadger!
guest
awesome, glad its half working. if you get stuck on the other half, report back with what youve found.
hw408
LP_L uses 13700 (U1) = working
LP_R uses 13700 (U3)

I have two signal generators one in each input.
Listening to Mix out, I hear the L input signal, good.
My input singals measured about the same, so i should be able to go pin to pin and compare U1 to U3.

Pins 6, 11 (power) match U1 to U3.
Pins 1, 16 (amp bias) match, range -10.5 to -2 w/cutoff low to high

Pins 7,8 (buffer input/output) no match.

U1:
pin7 = -0.4 to 1.3 (cutoff min to max)
pin8 = -1.9 to 0.165 (cutoff min to max)
pin5 output i see a wave on the scope good

U3:
pin7 = -2.2 (no change w/cutoff change)
pin8 = -10.8 (no change w/cutoff change)
pin5 output is stuck at -2.2

Ive been reflowing the bits around pin7/8.
I would guess the "cutoff control" section is trusted now, since it works for LP_L, does that sound right?
But my U3 buffer pins are stuck.
guest
sounds like u3 might be bad. pin8 should always be about 1V below pin7. are you measuring these voltages directly on the IC, or on a trace nearby? if there is a bad connection, then you wont see the right voltage if youre looking on a trace.
hw408
guest wrote:
sounds like u3 might be bad. pin8 should always be about 1V below pin7. are you measuring these voltages directly on the IC, or on a trace nearby? if there is a bad connection, then you wont see the right voltage if youre looking on a trace.


Ive been measuring ground to the IC legs.

i was poking around the second half of U3, to see if pin9/10 was similar to pin7/8.
close but not exact.

pin9 = -8.0 (vs pin8 -10.8)
pin10 = -2.2 (same as pin 7)

This is odd, when i touch the probe to pin 10, 12, or the IC side of R22 i hear the missing signal, but it isnt affected by the cutoff pot.

I could try replacing U3.
Ive been avoiding that until there was a reaon to take the risk.
guest
its possible that putting the probe on pin10 changes the value. do you have 2 probes, so you can look at pin10 and pin9 at the same time? maybe with your scope? there is a good chance U3 is bad, though.
hw408
guest wrote:
its possible that putting the probe on pin10 changes the value. do you have 2 probes, so you can look at pin10 and pin9 at the same time? maybe with your scope? there is a good chance U3 is bad, though.


Hi again
I replaced U3.
Previously Pin8 was stuck at -10.5 and Pin7 stuck at -2.3, I think 9 and 10 were the same.
After changing the IC (which was rough, tore a piece of a pad on pin5 but there was still a lot left to solder to so I went ahead.

Some of these seemed to no be stable very quickly so I took a note.

U3
13700 for LP_R channel

Cutoff Min Cutoff Max
pin 1 -10.300 -9.800
pin 2 0 -0.040 max creeping lower
pin 3 0 -0.002
pin 4 -0.010 -0.003
pin 5 -0.700 0.910 Min kept creeping slowly lower
pin 6 -11.2
pin 7 -0.700 0.900 Min kept creeping slowly lower
pin 8 -2.600 -0.245

pin 9 -0.020 0.340
pin 10 1.150 1.500 Max seemed to creep slowly lower
pin 11 11.2
pin 12 1.150 1.500 Max seemed to creep slowly lower
pin 13 -0.020 -0.004
pin 14 -0.010 -0.003
pin 15 -0.030 -0.060 Max crept higher?
pin 16 -10.300 -9.900

I see signal on the scope at R16 on the jack side, but nothing on he IC side Maybe an extreme blurry very there signal.

I'm not sure if any of these numbers indicate anything.
guest
getting chips out is a pain. sometimes i just clip all the legs and take them out one at a time. it would be nice if they were soceketed, then you could just swap the 13700s to see if that was the problem.

those voltages all seem ok. can you run your tests again, but this time put 5V DC at the jack input. you should see pins8,9 also go to 5V DC. be sure to have it setup for lowpass filter mode with the jumpers.
hw408
guest wrote:
getting chips out is a pain. sometimes i just clip all the legs and take them out one at a time. it would be nice if they were soceketed, then you could just swap the 13700s to see if that was the problem.

those voltages all seem ok. can you run your tests again, but this time put 5V DC at the jack input. you should see pins8,9 also go to 5V DC. be sure to have it setup for lowpass filter mode with the jumpers.


This time I took 5v from a bench power supply, through a patch cable, into the right side input jack which we are investigating.

No input in the left channel this time.

You mentioned pin 8/9 and comparing to my last numbers, they seem about the same.

Cutoff Min Cutoff Max
pin 1 -10.300 -9.800
pin 2 0.000 -0.007
pin 3 0.000 -0.002
pin 4 -0.013 -0.003
pin 5 -0.740 0.895
pin 6 -11.240
pin 7 -0.740 0.892
pin 8 -2.680 -0.245
pin 9 -0.028 0.349
pin 10 1.180 1.484
pin 11 11.260
pin 12 1.180 1.485
pin 13 -0.029 -0.004
pin 14 -0.013 -0.003
pin 15 0.000 -0.002
pin 16 -10.300 -9.890

I cant get these numbers to look nice in a table, blank spaces keep collapsing.
I didn't try to use the scope this time.
Thank you again for your time!
guest
it looks like the signal is not getting to the chip. turn off the power, and measure the resistance from pin3 to the input (should be 100k), and then from pin3 to ground (should be 470ohm). be sure to measure with the probe directly on the IC pin, and not a nearby trace. its possible there is a problem with one of the jumpers.
hw408
guest wrote:
it looks like the signal is not getting to the chip. turn off the power, and measure the resistance from pin3 to the input (should be 100k), and then from pin3 to ground (should be 470ohm). be sure to measure with the probe directly on the IC pin, and not a nearby trace. its possible there is a problem with one of the jumpers.


I am amazed to find exactly
100k resistance between pin 3 and the input jack
468R between pin 3 and ground

Here is a pic of the jumpers

edit:
adding reference from ST for jumper settings for LP mode.
I believe these are rotated 180° relatove to my picture (but that would be awesome if that was the problem)
guest
ok, so reapply the 5V (or if you have 10V handy, that would make it easier) to the input, and measure on either side of the 100k, and see what the voltages are. it should be 5V and 25mV.
hw408
guest wrote:
ok, so reapply the 5V (or if you have 10V handy, that would make it easier) to the input, and measure on either side of the 100k, and see what the voltages are. it should be 5V and 25mV.



Giving the Right input 10v now. From HP bench supply.
Measure at the jack pins = 10v.

Jack gnd pin to R16 - on one side I read that same 10v
R16 other side i read 0.046v

U3 Pin3 = 0.046v

I replaced R16 before this measurement because I was reading zero on the IC side of R16, and 10v on the other. Too weird.

So maybe this is actually an improvement?

I will move my power supply back and be ready to try another audio test if you say so.
cygmu
That’s exactly what it should read and evidently your work on R16 has helped matters so definitely worth investigating whether you’ve fixed it, I’d say.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group