MS-20 (LM-13700 version),Polivoks or EMS VCF on veroboard?

Sebo · Joined: Apr 27, 2007 Posts: 564 Location: Argentina

AC mixer means that can only accept AC signals, in practice the coupling is done by a high pass filter with a sub sonic cutoff frequency, so in general an AC mixer is considered capable of mix audio signals but not CV signals. Also an AC mixer will kill any DC that try to come in.
A DC mixer is suitable for any signal, audio and CV, and it will pass any DC.

I'm not sure to understand your last question, but if you will always use that path, your VCA could be AC or DC, don't really matters.

isak · Posted: Sun Jan 06, 2013 12:06 am Post subject:

Understud, thank you for that!
I'll build the mixer and let you and the guys know if it worked for the Jorg ms20.

cheers.

isak · Posted: Wed Jan 09, 2013 4:36 am Post subject:

Hi Sebo.

I build the mixer and added 10uF Tantalums to all inputs like you adviced me.
The + side to the input, the - side to the resistors.
It's sound better now Smile

You where right about the vol input!
I lower the vol input in the mixer and the square and saw sound much much better.

One thing is still bothering me, the self osc is comming to soon when turning the reso pot CW, is there any way to make it start in about 3 o'clock?

Thanks Sebo.
Isak E.

Sebo · Joined: Apr 27, 2007 Posts: 564 Location: Argentina

Hi isak:
Great you get it sounding right!!!
I'm on vacation now, far away from home, I have to see the circuit to give you an answer. When I get back I'll check it out.
Cheers.

isak · Posted: Wed Jan 09, 2013 8:24 am Post subject:

Hi Sebo.

enjoy your vacation man.
See you soon Smile

feggster · Joined: Sep 12, 2011 Posts: 52 Location: uk

hiya,
I built this and it works great (a few modifications to the layout to get it working) now I would like to add a 3 way switch to change LP HP BP
I have a slide switch (shown below) can anyone translate the wiring diagram for my switch...my brain hurts

cheers.

4p3t.jpg

Description:

Filesize:

45.58 KB

Viewed:

46553 Time(s)

ms20lpbphp_switch.jpg

Description:

Filesize:

51.82 KB

Viewed:

1223 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

gila_crisis · Joined: Feb 06, 2014 Posts: 3 Location: Switzerland

I built a MS20 VCF following this schematic I found on takeda's site:
http://www.aleph.co.jp/~takeda/radio/MS20clone.html

But I have a problem. I used a TLC2272 and a OPA2134 for the dual opamp, transistors are BC560c and I'm powering the circuit with a MAX1044, so that I can get +/-9VDC out of a normal 9V power supply.
I double checked everything and the voltages on the IC seams to be correct but it doesn't work right.
After a while the circuit is powered the sound starts to come out and for a while it works, but after a bit I noticed that with the freq fully turned down, the filter slowly opens! I measured the voltage on the circuit and I found out that on the collector of Q1a (as per takeda's schematic) the voltage slowly arise with the time, resulting in this "self opening" of the filter. Do you have any suggestion what the cause may be?

gila_crisis · Joined: Feb 06, 2014 Posts: 3 Location: Switzerland

I found the mistake.. I just forgotten to connect pin2 of the opamp with the 470k resistor going to -15VDC!
Now it works like a charme! I'll post the layout soon.

gila_crisis · Joined: Feb 06, 2014 Posts: 3 Location: Switzerland

here you'll find my layout for Takeda's MS20 VCF. I added a charge pump to use it as a stompbox with a standard 9VDC power supplies (also with -/+9V instead of +/-15V the filter works really well! but you can omit the charge pump part of course).

ms20_filter.png

Description:

Filesize:

145.02 KB

Viewed:

3787 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

jmcecil · Joined: May 11, 2013 Posts: 20 Location: Prattville, AL

Sorry to resurrect the oldie, but I'm trying to use protoboard for the first time and this seemed like a fairly easy project. I ran into an issue though. trying to read the diagrams on the first page I find a jumper coming off the -15 and I can't figure out what it is for. Can someone explain please?

Posted Image, might have been reduced in size. Click Image to view fullscreen.

Posted Image, might have been reduced in size. Click Image to view fullscreen.

Also, the schematics are no longer at the link on the initial post. Would anyone have a link?

Thanks much

-minus- · Joined: Oct 26, 2008 Posts: 787 Audio files: 13

I'd say that cut to the left of your long rectangle should not be there. If you have already cut this board, you will need to bridge that gap. You could use a piece of resistor leg and solder to the reverse side. Or you could swap R6 and the jumper, then bridge the gap with a three hole horizontal jumper to take the (-)V to pin 11 of the TL074.

jmcecil · Joined: May 11, 2013 Posts: 20 Location: Prattville, AL

roglok · Joined: Aug 28, 2010 Posts: 202 Location: uptown

jmcecil · Joined: May 11, 2013 Posts: 20 Location: Prattville, AL

Fahz0r · Joined: Jul 20, 2014 Posts: 1 Location: Netherlands

I've come across this layout (on page 1) a couple days ago and after fixing a couple things to it the filter is working fine and dandy! If anyone is still interested, here is the redraw.

(Apologies if I've missed this already being posted somewhere, and the spelling mistakes)

A little note just incase you were wondering: I've put in "R13.1" and "R13.2" simply because I liked this exact resistance on the resonance, since it's not too loud now but still you can crank it. You can also fit in a 10K trimpot there if you want.

And this is -12/+12v Very Happy

which works perfectly fine.

Edit: Added lopass/hipass switch

lm13700_MSfilter_Fix.jpg

Description:

Filesize:

367.21 KB

Viewed:

1895 Time(s)

This image has been reduced to fit the page. Click on it to enlarge.

Snaper · Joined: Feb 28, 2014 Posts: 217 Location: Hungary

The +/- jumpers should be switched on tl074.
Pin 4 to -, pin 12 to +

hoyager · Joined: Feb 16, 2010 Posts: 27 Location: New Zealand

Snaper · Joined: Feb 28, 2014 Posts: 217 Location: Hungary

hoyager · Joined: Feb 16, 2010 Posts: 27 Location: New Zealand

Cool, it seems people are starting out building stuff by using this, which I did, so thought it'd better not be misleading

Here's some mods I do on this circuit to keep the res under control, overdrive the input a bit and get the freq pot to behave with a full range.

On the last layout posted - lm13700_MSfilter_Fix.jpg

1. Make R7 2.2k and the link to ground on the opposite transistor, 2.2k as well (as per ms20). So both transistor bases go to ground via a 2.2k.

2. Make the CV resistor from the Freq (cutoff) pot 300k and make R3 and R4, 7.5k (Or don't cut under pins 1 and 16 of LM13700 and use 1 7.5k resistor to both pins).

2a. Use 47n in place of c1 to cut down on some cv bleedthrough (rounds corners of square wav if an lfo is used) - not necessary for manual cutoff control

3. R14 make 2.7k (or 5k trimpot) for self oscillation later in the travel of the res knob and make the res knob 10kb, put a 2.2k resistor across lug 2 + 3, and use an 8.2k resistor from the res3/out to the lug 3 of res pot (as per ms20)

3a. Use 3 1n4148 in series in each direction and 10k in the feedback loop for res opamp (between pins 1 and 2) (as per ms20). I use 1s1555 which are what they used in the original. I think have a different forward voltage curve (going higher than 1n4148 but at a lower rate....) Anyway I think they break up a bit nicer and sound a bit more garbly. This might be due to them being more uneven in forward voltage drop than modern manufactured 1n4148, but again I don't really know.

4. Use a polar 10uf cap for c5 (+ side at - input (pin 9 of TL074))

5. Low pass input resistor R18 make 7.5k for more overdrive and add a 47k resistor from the hi pass input (gnd end of C4) to gnd .This will tie the cap to ground when there is no input, eliminating some switching complexity, although this might require make before break for click-less switching)

6. Lofi opamp like LM348 works well too (along with LF347) can overdrive filter without clipping. I found the TL074 or 072 will shutdown in this situation as buffers if there is too much input (has to be quite a lot, but filter does sound good with quite a lot of input)

7. I use 2.2n's for C2 and C4, and compensate for high freq loss (only very high freq) with 2.2n and 470k in series in feedback loop of an output opamp. If you want a mellower filter just use 2.2n's.

Also in the this last layout, I think the diodes and resistors D1 D2 R13.1+2 are supposed to be 1 line down so they are across pins 1 and 2 of the opamp.

Andy

Snaper · Joined: Feb 28, 2014 Posts: 217 Location: Hungary

hoyager · Joined: Feb 16, 2010 Posts: 27 Location: New Zealand

No sorry, just as the last layout that was posted by someone else is plus these changes.

Shouldn't be too hard to follow, I tried to make it as clear as I could. You could recreate the existing layout with diylc (google it, its free) and add in the changes.

Andy

wackelpeter · Posted: Fri Sep 05, 2014 12:07 am Post subject:

Maybe this will be of interest for someone...

I've recently built 2 of These ms20 VCF's with the lm13700 but the second one didn't want to oscillate properly...

following Rene's schematics i've used the latter suggested design (mostly what Hoyager did) with 2,2n caps instead of 1n, and with the 8,2k and 2,2k resistors placed at the reso-pot.
also i used a 1,8k resistor and 5k trimmpot to the inverting Input of the Feedback stage of the opamp...
the resistor of the summing stage after the freq-pot is 150k and the resistor to gnd at the base of the trannie is 680 ohms instead of 1,8K.
the Output cap is 22uF instead of 0,47 and reso pot is 10k...
Also added is a pot for the Input Signal.

That said my first filter worked very well but the second wont oscillate or just with a minimal effect on Sound when reso is turned fully on...
checked the 2nd one a couple of times but couldn't find the mistake... then compared it with the first one and found that on the first one i accidentally reversed 2 of the Diodes in the Feedback path...
I used the design with 2 rows of 3 diodes...
when you revert one of them in each row you get a very cool sounding effect... just using link instead of 2 diodes in each row didn't work for that...

maybe some of you want to try this "trick" with the diodes it sounds really good...

Thelonious · Joined: Mar 03, 2015 Posts: 5 Location: France

Hi,
I project to make this MS20 filter clone, so I do a new vero layout (smaller, and without the mistakes). I'll post it very soon.
But I don't understand why there are these 100nF (C6 to C9) between +15 -15V and ground, and why put it 2 times ? They are not present in the schematic.
Thanks
It's an old post, hope someone follow it yet.

Snaper · Joined: Feb 28, 2014 Posts: 217 Location: Hungary

I think they are fot stabilizing the power. I usually add 2 10u and 2 100n.

Thelonious · Joined: Mar 03, 2015 Posts: 5 Location: France

Ok thanks snaper.
I finish my layout.
I made 2 versions, one "normal" and one with hoyager's mods.
But I have some questions.

For the "normal" version :
What is the value of C4.1 ?
Switch A1 and A2 are connect to ground, and A3 without connection ?
The DPDT is ON-ON (B1 is Lowpass input, and B3 is Highpass input) ?

For the hoyager version :
Don't understand this : "use an 8.2k resistor from the res3/out to the lug 3 of res pot (as per ms20) " ? (It's a bit confused)
And this : "add a 47k resistor from the hi pass input (gnd end of C4) to gnd ." ? (If I understand, he want to put a 47k in serie with C4 before to go to ground).

I can send you my layouts if you want.
Thanks