| Author |
Message |
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
 Posted: Mon Nov 26, 2007 9:15 pm Post subject:
Quad VCF for Casio SK samplers! |
  |
|
Hi Everyone,
Well, I've just completed a huge project, which has been on my mind for a while now, ever since finding this forum - to build a true polyphonic, 4-voice VCF module, with CV input and resonance, running off 9V DC.
It has taken me about 3 months of spare time to get this thing researched, thought out, breadboarded, built and optimised for the Casio SK sampler keyboards (I have an SK8). But it has finally come!!
Below are the results of my efforts.
But I wanted specially to post this message, in order to say a big THANK YOU to everyone in this DIY community; for both directly and indirectly helping me with this project by the way of advice, and links to circuits and information.
I especially want to thank Tim Escobedo, who designed the Q&D VCF, from which this quad VCF is wholly adapted from. Using it as a starting block got me half way there in one hit! Thanks to all the people here who put me on to it!!
I've written up a complete illustrated guide on how to make this quad VCF, which I call the "PHAT PHILTER BANK", and how to set it up for the Casio SK series keyboards. The guide is here, at the bottom of the page here:
http://www.jz-server.de/forum2/e107_plugins/forum/forum_viewtopic.php?559
There is also on that page an mp3 example of one of the SK presets with the filter applied. Makes the humble SK-8 sound like a real analog synth!!
You can also make it as a stand-alone unit, or adapt it to another type of keyboard, although another keyboard may or may not be able to take advantage of the full features that this VCF unit has - such as CV input. The SK samplers have separate internal envelope CV signals for their VCA's, which is particularly useful for driving this VCF. The guide shows you how to set it up in various ways.
Again, thank you all SO MUCH for sharing the knowledge that enabled this to happen!!
Cheers, Graham
| Description: |
| Component side of the PHAT PHILTER BANK |
|
| Filesize: |
185.93 KB |
| Viewed: |
882 Time(s) |
| This image has been reduced to fit the page. Click on it to enlarge. |
|
| Description: |
| Track side of the PHAT PHILTER BANK |
|
| Filesize: |
133.13 KB |
| Viewed: |
696 Time(s) |
| This image has been reduced to fit the page. Click on it to enlarge. |
|
| Description: |
|
| Filesize: |
128.06 KB |
| Viewed: |
606 Time(s) |
| This image has been reduced to fit the page. Click on it to enlarge. |
|
| Description: |
|
| Filesize: |
109.27 KB |
| Viewed: |
637 Time(s) |
| This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
|
Back to top
|
|
 |
Tim Servo
Joined: Jul 16, 2006
Posts: 924
Location: Silicon Valley
Audio files: 11
|
| Posted: Mon Nov 26, 2007 9:27 pm Post subject:
Quad VCF for Casio SK samplers! |
|
|
Neato Burrito!
I had no idea you could do that with an SK series sampler. You definitely get points for venturing into new territory. Only one request: can you possibly post an audio clip or two here so I don't have to go sign up for another Yahoo group? I have too much email to sort through as it is  Thanks!
Tim (having a neato burrito with salsa) Servo |
|
|
Back to top
|
|
|
Randaleem
Joined: May 17, 2007
Posts: 456
Location: Northern CA, USA
|
| Posted: Mon Nov 26, 2007 9:30 pm Post subject:
|
|
|
VERY! Nice!
Congratulations. Beautifully integrated into the synth!
Kind regards, Randal |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
|
|
Back to top
|
|
|
danielwarner
Joined: Dec 18, 2006
Posts: 65
Location: Sacramento
|
| Posted: Tue Nov 27, 2007 1:08 am Post subject:
|
|
|
This looks great! In the guide the SK-5 and 8 are specified... is there anything prohibiting use with an SK-1, or you simply didn't have one to try it out on? If so I may have to venture into new territory... when the opportunity arises
Cheers,
Daniel |
|
|
Back to top
|
|
|
v-un-v
Janitor
Joined: May 16, 2005
Posts: 8932
Location: Birmingham, England, UK
Audio files: 11
G2 patch files: 1
|
| Posted: Tue Nov 27, 2007 2:20 am Post subject:
|
|
|
Hey Graham!
Thanks for posting this (I've still got my SK5 upstairs  )
 to electro-music.com!
I used to be on Yahoo's SK list 6 or 7 years back, then for some reason I couldn't log back in. Are Albie (from Portland, Oregon) and Freya (London, England) still on there?? I miss them loads! You folks are more than welcome to come over here you know. There are no Yahoo-in-you-face adverts here and you can upload stuff easily 
_________________
ACHTUNG!
ALLES TURISTEN UND NONTEKNISCHEN LOOKENPEEPERS!
DAS KOMPUTERMASCHINE IST NICHT FÜR DER GEFINGERPOKEN UND MITTENGRABEN! ODERWISE IST EASY TO SCHNAPPEN DER SPRINGENWERK, BLOWENFUSEN UND POPPENCORKEN MIT SPITZENSPARKSEN.
IST NICHT FÜR GEWERKEN BEI DUMMKOPFEN. DER RUBBERNECKEN SIGHTSEEREN KEEPEN DAS COTTONPICKEN HÄNDER IN DAS POCKETS MUSS.
ZO RELAXEN UND WATSCHEN DER BLINKENLICHTEN. |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
| Posted: Tue Nov 27, 2007 3:20 pm Post subject:
|
|
|
Hi Daniel,
I don't have an SK-1 to try it out on, but yes, it will work on the SK-1, although the hookup points are slightly different to the SK5/8. Download the SK-1 schematics here:
http://www.jz-server.de/casiology/manuals_s.html
In the schematics it says pins 89-92(EH-EE) are the four envelope signal outputs. Tap them at the earliest convenient soldering point out from
the processor, that way you get the highest envelope CV voltage
possible so the filters open nice and wide.
Now all you need to do is locate the 4 individual audio lines. These are pins 93 (SH), 94 (SG), 95 (SF) and 96 (SE). Tap
them directly AFTER each of the 100k resistors on each audio line
directly out from the processor. Tapping them any further down the
line reduces the treble, which the filter needs with such low quality
samples.
Cheers and good luck!!
Hi v-un-v,
I checked the member's database and couldn't find either of the 2 names you mentioned. Perhaps yahoo has a login expiry if you don't login for a while. Reapply for the forum, and I'll approve it straight away!
Thanks for the kind offer about uploading stuff, I try not to fill forum servers with unnneccesary MB's without permission first, and it is a pain for people to have to join the yahoo group to access it, so thanks!!
Cheers, Graham
Last edited by gmeredith on Tue Nov 27, 2007 3:33 pm; edited 1 time in total |
|
|
Back to top
|
|
|
v-un-v
Janitor
Joined: May 16, 2005
Posts: 8932
Location: Birmingham, England, UK
Audio files: 11
G2 patch files: 1
|
| Posted: Tue Nov 27, 2007 3:27 pm Post subject:
|
|
|
Hi Graham!
It was 7 years ago!  
Thanks anyway
I'll probably stay here, but thanks for the invite anyway .
_________________
ACHTUNG!
ALLES TURISTEN UND NONTEKNISCHEN LOOKENPEEPERS!
DAS KOMPUTERMASCHINE IST NICHT FÜR DER GEFINGERPOKEN UND MITTENGRABEN! ODERWISE IST EASY TO SCHNAPPEN DER SPRINGENWERK, BLOWENFUSEN UND POPPENCORKEN MIT SPITZENSPARKSEN.
IST NICHT FÜR GEWERKEN BEI DUMMKOPFEN. DER RUBBERNECKEN SIGHTSEEREN KEEPEN DAS COTTONPICKEN HÄNDER IN DAS POCKETS MUSS.
ZO RELAXEN UND WATSCHEN DER BLINKENLICHTEN. |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Tue Nov 27, 2007 4:14 pm Post subject:
|
|
|
Nice work , Graham. Are the schematics up anywhere? I couldn't find them. Thanks!
Last edited by frijitz on Thu Nov 29, 2007 6:20 pm; edited 1 time in total |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
| Posted: Tue Nov 27, 2007 4:40 pm Post subject:
|
|
|
Hi frijitz,
I actually don't have any schematics for it that I've drawn myself, as the circuit is a simple adaptation of Tim Escobedo's Q&D VCF, for which the schematics are here:
http://filters.muziq.be/diy/lehrberg/qdvcf
My circuit is basically 4 Q&D VCF's hooked together, with a few extra resistors and capacitors on the inputs and outputs, to optimise it for the SK samplers.
Cheers, Graham |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Tue Nov 27, 2007 6:34 pm Post subject:
|
|
|
| gmeredith wrote: | | I actually don't have any schematics for it that I've drawn myself, as the circuit is a simple adaptation of Tim Escobedo's Q&D VCF ... |
Many thanks for the link! |
|
|
Back to top
|
|
|
germaniac
Joined: Aug 04, 2006
Posts: 200
Location: California
Audio files: 7
|
| Posted: Thu Nov 29, 2007 11:44 am Post subject:
|
|
|
Hey Graham,
Nice work!
| gmeredith wrote: |
In the schematics [for the SK-1] it says pins 89-92(EH-EE) are the four envelope signal outputs. Tap them at the earliest convenient soldering point out from the processor, that way you get the highest envelope CV voltage possible so the filters open nice and wide.
|
Question: in your opinion, would it be possible to sum these CVs (passively or actively) and then send them to an outboard VCF? I've got an SK-1 that might be nice with a CV out. . . .
Regards,
Joe |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Thu Nov 29, 2007 12:11 pm Post subject:
|
|
|
| gmeredith wrote: | I actually don't have any schematics for it that I've drawn myself, as the circuit is a simple adaptation of Tim Escobedo's Q&D VCF, for which the schematics are here:
http://filters.muziq.be/diy/lehrberg/qdvcf |
Hi again Graham --
I'm having trouble getting that circuit to work right. Any hints as to (1) what the input level should be (2) what kind of LED to use?
Since it is a twin-T network in a feedback path, it seems to me it should be more like a bandpass rather than a lopass. Does it really act like a lopass?
Actually I'm seeing what looks like a chaotic response, which is really getting me interested in this thing. But I'll put that on a separate thread.
Thanks!
Ian
Edit: BF corrected. Sorry!
Last edited by frijitz on Thu Nov 29, 2007 6:19 pm; edited 1 time in total |
|
|
Back to top
|
|
|
Tim Escobedo
Joined: Nov 27, 2007
Posts: 3
Location: California
|
| Posted: Thu Nov 29, 2007 5:31 pm Post subject:
|
|
|
Sounds good!
Frijitz, you're absolutely right. They're simply bridged-T/twin-T filters. Technically, they should act more as bandpass filters. Sonically, however, they can be good enough and difficult to really differentiate.
The circuit was really hacked for a DIY 9V powered guitar stompbox. For most of the sweep, there's considerable gain, so I recommend cutting the input level to give the least distortion throughout the sweep. This is for guitar levels. Synth levels even moreso. One interesting thing about using this type of filter with the non-inverting input is that it kind of fakes a lowpass sound as the filter sweeps down. Since the gain is high throughout most of the sweep, it drops down to near unity at the low end of the sweep.
"Unbalancing" the t-filter network has the effect of broadening out the width of the usual bandpass response. The result also fakes a lowpass sound sans resonance.
I used some generic red LEDs as sort of variable resistors. Hack quality, but good enough for relatively slow envelopes. |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Thu Nov 29, 2007 6:17 pm Post subject:
|
|
|
| Tim Escobedo wrote: | ....They're simply bridged-T/twin-T filters. Technically, they should act more as bandpass filters. Sonically, however, they can be good enough and difficult to really differentiate.
The circuit was really hacked for a DIY 9V powered guitar stompbox. For most of the sweep, there's considerable gain, so I recommend cutting the input level to give the least distortion throughout the sweep. This is for guitar levels. Synth levels even moreso. One interesting thing about using this type of filter with the non-inverting input is that it kind of fakes a lowpass sound as the filter sweeps down. Since the gain is high throughout most of the sweep, it drops down to near unity at the low end of the sweep.
"Unbalancing" the t-filter network has the effect of broadening out the width of the usual bandpass response. The result also fakes a lowpass sound sans resonance.
I used some generic red LEDs as sort of variable resistors. Hack quality, but good enough for relatively slow envelopes. |
Tim!
Many thanks for all that great info. That's a very interesting circuit. It strikes me as not totally obvious from just looking at the diagram what the effects of the asymmetry in the network and the added Q resistor would have on the perceived sound. I can see that it might sound somewhat like a lopass at the low end, now that you explain it.
I'm driving it from a synth tri vco and I have used either +12V or +/- 12V power. I am seeing some incredible chaotic behavior. I am going to start a separate thread on this, so as not to hijack Graham's cool project. Hopefully I can get it started tonight -- I already have an amazing recording of a sweep going through a period-doubling cascade to chaos.
Thanks again for checking in on this thread, I really appreciate your input.
Ian |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
| Posted: Thu Nov 29, 2007 8:23 pm Post subject:
|
|
|
Hi germaniac,
| Quote: | | - "would it be possible to sum these CVs (passively or actively) and then send them to an outboard VCF? I've got an SK-1 that might be nice with a CV out" |
It is possible to sum the CV lines. You will need to put a signal diode on each CV output you make, otherwise they will interfere with the casio's VCA internally (I don't think it will damage it, but it might make for some strange results). But be aware that the Casio's CV lines are not PITCH CV lines, as in other synths, such as analog synths like an SH101. They operate the VCA level in the casio, not an oscillator VCO pitch. So if you were wanting an SK with "CV outs" to control another synth, such as an SH101 - it will not work in the normal manner - all other analog synths with CV input use the CV for PITCH.
Hi Frijitz,
| Quote: | | - "I'm having trouble getting that circuit to work right. Any hints as to (1) what the input level should be (2) what kind of LED to use?" |
1. The input level should be about line level.
2. Regarding the LED, I just used a common red 3mm LED. If you are getting chaotic response, try reversing the orientation; you may have had it in backwards.
Also, until you get it running right, run it off a 9V battery, not off the internal rails of the synth, until things are working. Also, don't hook the CV line up to a synth CV, until you've got it working. It won't handle more than about 3V DC as a CV input the way I've set it up, because the Casio SK only puts out 2.5V DC for its CV. Most synths outupt up to 10V. Put a 100k trim pot in the CV input to handle more than 3V input
By the way, don't confuse the CV outputs of other synths with the Casio's CV. They do completely different things. The Casio's CV operates the VCA - not the note pitch. It's processor does the pitch. All other synths use a CV line for PITCH, not VCA or VCF.
Meaning, that if you hooked up the CV output from, say, a Roland synth such as the SH101, and connected it to the filter's CV input, the higher the note you play on the SH101, the more the filter will open, and the reverse as you play lower notes. It could appear random and chaotic if you have it hooked up in this manner, but were expecting results like I have created with the Casio. The Casio is a completely different beast, and this Phat Philter Bank was designed with it specifically in mind. Mind you, though, you can use this to your advantage to control the filter cutoff with a CV-outputting synth, shuch as an SH101 - if you play different notes on the SH, the filter will open and close higher and lower. Play a sequence on the SH101, and you could get the random filter cutoff effect that the SH101 has when you use its LFO to control it's own filter
| Quote: | | - "it seems to me it should be more like a bandpass rather than a lopass. Does it really act like a lopass? |
It seems to allow low frequencies at any cutoff position for me, but not treble when the cutoff is at a low frequency. If it is bandpass, the bandwidth may be quite wide, and so give the appearance of a low-pass.
Cheers, Graham
Last edited by gmeredith on Thu Nov 29, 2007 9:53 pm; edited 2 times in total |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
| Posted: Thu Nov 29, 2007 8:38 pm Post subject:
|
|
|
Hey, Tim!!
Thanks for the Q&D VCF - it made this filter possible! I've been searching far and wide for a simple DC single rail filter for ages until I found yours.
By the way - I downloaded a pdf from someplace of the Q&D VCF (not from your site - I can't find where from now) with a PCB screen layout and parts layout also, not just the schematics. I don't know if it was yours, or whether someone drew it up on your behalf, but the PCB has an error in the track layout - with the Vref 1M resistor placement. I corrected the track layout and put it up on the Yahoo CasioSK group forum.
Cheers, Graham |
|
|
Back to top
|
|
|
germaniac
Joined: Aug 04, 2006
Posts: 200
Location: California
Audio files: 7
|
| Posted: Thu Nov 29, 2007 9:27 pm Post subject:
|
|
|
| gmeredith wrote: |
It is possible to sum the CV lines. You will need to put a signal diode on each CV output you make, otherwise they will interfere with the casio's VCA internally (I don't think it will damage it, but it might make for some strange results). But be aware that the Casio's CV lines are not PITCH CV lines, as in other synths, such as analog synths like an SH101. They operate the VCA level in the casio, not an oscillator VCO pitch. So if you were wanting an SK with "CV outs" to control another synth, such as an SH101 - it will not work in the normal manner - all other analog synths with CV input use the CV for PITCH. |
Okay, understood on the diodes. Thanks.
Understood on the Casio CVs as well--I gather they are strictly speaking more like ADSR outs, since they reflect only the volume envelopes of the sounds. I'd guess true pitch CV from a Casio would probably need MIDI-to-CV hardware of some sort. . . .
Regards,
Joe |
|
|
Back to top
|
|
|
gmeredith
Joined: Jun 28, 2006
Posts: 82
Location: Tasmania, Australia
Audio files: 1
|
| Posted: Thu Nov 29, 2007 9:46 pm Post subject:
|
|
|
| Quote: | | I gather they are strictly speaking more like ADSR outs, since they reflect only the volume envelopes of the sounds. I'd guess true pitch CV from a Casio would probably need MIDI-to-CV hardware of some sort. . . . |
Yes, that's right. The CV is actually shaped by the processor internally, depending on the preset sound you select, and is simply spat out to the Casio's VCA as a pre-shaped ADSR signal. The VCA is the only analog part of the Casio. The sound source is sample ROM, ie. digital, so there is no pitch CV to speak of.
Regarding getting MIDI to CV out from a Casio, ther is a guy who designed a MIDI OUT circuit for his SK8, so he could use it as a portable master keyboard (neat ).
http://www.jz-server.de/forum2/e107_plugins/forum/forum_viewtopic.php?138
If you could get him to make you one, you could then get a MIDI to CV box for the MIDI out, which would then drive an analog synth or whatever takes a CV input that you wanted.
Cheers, graham |
|
|
Back to top
|
|
|
Tim Escobedo
Joined: Nov 27, 2007
Posts: 3
Location: California
|
| Posted: Thu Nov 29, 2007 9:56 pm Post subject:
|
|
|
| gmeredith wrote: |
By the way - I downloaded a pdf from someplace of the Q&D VCF (not from your site - I can't find where from now) with a PCB screen layout and parts layout also, not just the schematics. I don't know if it was yours, or whether someone drew it up on your behalf, but the PCB has an error in the track layout - with the Vref 1M resistor placement. I corrected the track layout and put it up on the Yahoo CasioSK group forum.
|
I'm glad it's worked out for you. It's probably the loneliest circuit I ever posted, and I can't recall ever getting anyone's feedback about it. Which is kind of a shame, because I found it to be pretty useful.
A couple months ago, someone told me they found my circuit PCB online. On investigation, I found quite a few PCB and perfboard layouts floating around that I haven't done. So be warned, I can't vouch for the accuracy of any layout found online. |
|
|
Back to top
|
|
|
snufkin
Joined: Apr 06, 2007
Posts: 57
Location: the world
|
| Posted: Sun Dec 02, 2007 2:59 pm Post subject:
|
|
|
this looks like an amazing project thanks Graham i will be doing this to both my sk samplers i think 
and Tim Escobedo you are a living legend I have built 4 of your "snippets" each one a marvel of usability economy
pwm pedal i love you so |
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007
Posts: 1536
Location: nyc
Audio files: 41
|
|
|
Back to top
|
|
|
slacker
Joined: Nov 18, 2007
Posts: 301
Location: England
Audio files: 11
G2 patch files: 1
|
|
|
Back to top
|
|
|
|
Forum index » DIY Hardware and Software
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
