Author |
Message |
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Tue Aug 21, 2007 8:09 pm Post subject:
|
|
|
I reiterate: Holy Cow!! This is great news! What really amazes me is how you fit that all onto one board. Your perfboarding skills are amazing.
Thanks for the current draw number - that's right in line with what I recorded, forgot, and Fonik reminded me of.
The load thing might indeed be the CD4013. IIRC, the debouncing on that is a bit different, and the debounce cap serves also to provide a narrow pulse. Now, one thing that might isolate it is if the load reacts to an external load pulse the same way or not. The external load pulse and the manual load pulse are ORed together (IIRC, the schematic isn't in front of me), so if a wide load pulse does the same thing, I'd look at either the CD4013 or the CD40106 that provides the reset one-shot.
I'm eager to hear what you think of the thing (if it was worth the task of wiring that thing up).
Fan-freakin-tastic!
Scott _________________ My Site |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Tue Aug 21, 2007 8:20 pm Post subject:
|
|
|
Hey - sometime, once you get it all going, check out how fast you can clock the thing before it doesn't react to the clock. Bill suggested that to me yesterday, and I hadn't really tried it. Mine stopped clocking at around 44 kHz or so, which surprised the heck out of me - I didn't think it would clock nearly that fast.
If yours does the same thing, there's another cool use - the gates and triggers go away long before you reach that frequency, but the voltage outputs become audio outputs, the timbre of which is controlled by the pots. The frequency is also controlled by if you're in the 8X2 or 16X1 mode, or have gate bus 1 reloading at a certain point. If you can clock it that fast, run noise or an audio signal into the random input and listen to that as you sweep the clock freq up and down - it can get pretty wild!
Speaking of which, does the gate bus 1 load cause the same kind of problems the manual load does? That might be another clue.
Take care,
Scott _________________ My Site |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
Posted: Tue Aug 21, 2007 10:45 pm Post subject:
Making Klee Noise |
|
|
Hi Scott, I got the Klee hooked up to synth an played for a little while...WOW very cool! It makes a normal sequencer seem rather boring.
It's going to take me a while to get out of the normal sequencer mode, and get my brain wrapped around everything that is going on in the Klee. With just a few minutes of playing I have gotten some pretty cool sounds already. I need to go back through the Klee threads and pay closer attention to some of your patches in order to get a better feel of it's capabilities.
Here's are some early observations:
1.) The load switch works pretty much every other time(not always), and with an external load pulse it works every other time consistently. I'm wondering if I mis-wired the load section. I really haven't analyzed what you are doing there completely , so I need to think about that section.
2.) As bits are added to pattern the pitch can quickly go out of hearing range, so much so that you have to tune down the VCO. I was wondering what you thought of fixing the Max Voltage to the CD4066 switches to say 5 volts and then doing the scaling in the output circuit instead.
This would do two things: 1st) Allow positive and negative with the octave scale. 2nd) I noticed with the Octave and Semi scale switches both set to minimum there is no CV at all, which makes sense because this is the only supply to the CD4066 bus. Scaling on the output with a fixed MaxV would guarantee that you would always have some CV going into the CD4066 switches. I'm not sure if I am thinking clearly here but it seems like since the shift register bits are all summed together your are going to get multiplication of of the scale voltage, not just the individual pot voltage settings. With this in mind, I think that you would only get true octave scaling when there was only one bit set. By Scaling in the output amp section, the multiplication of the scale voltage would not occur. Am I making any sense here? Another thought occurred that by scaling in the output you could do the A, B, and A+B channels independently. Maybe all three channels could be driven by the existing Octave and Semi rotary switches and you could add a center off toggle switch to each channel that would allow you to go +/- an octave separately.
Overall I am extremely pleased and after I have gotten some sleep and my eyes have cleared up from the solder smoke I want to really dive into some serious Klee patching and sound generation. Thanks again for your design, help, compliments, and encouragement!
Tim |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Wed Aug 22, 2007 7:19 am Post subject:
|
|
|
Quote: | 2.) As bits are added to pattern the pitch can quickly go out of hearing range, so much so that you have to tune down the VCO. I was wondering what you thought of fixing the Max Voltage to the CD4066 switches to say 5 volts and then doing the scaling in the output circuit instead. |
That idea was actually approached, but if you think about it, it would still hit the rails pretty easily before attenuating the output. If you set it to some value that would prevent it from hitting the rails before attenuation, then you're limiting the max voltage swing you can get with low bit density patterns.
The original Klee had around 14 or so volts on the pots, and, though there were only four of them, it was very easy to make the thing swing ultras-sonic and infra-sonic as well. So each pot had to be tuned carefully, and one had to pay attention to bit density, which wasn't so easy, because the only way to program it was through the random input (one couldn't load patterns), and there were only four LEDs - one for each tap of the CD4006 shift register. That fact and the pots being interactive led me to describe it as akin to molding loose clay. That's the reference Romeo Fahl used to coin the term "Klee", because of the clay reference combined with the abstract style of Paul Klee's art.
When I started messing around with the Model 2, which has sixteen pots, I started thinking about how to limit the voltage on each pot so that things would be easier to program. The thought struck me to use tuned intervals as the max voltage, and these intervals would have to be fairly small, with sixteen pots. I also still wanted to preserve the essence of the original Klee - though it was a bitch to program, it still produced some very interesting patterns by the mere fact it was so volatile. That's why I prefer to leave at least one larger range there (I think 4 or 5 volts is good).
Anyway, I could either limit that voltage before or after the mix. Before the mix would eliminate the need for multiple attenuators. After the mix, one would either need three stepped attenuators to easily reproduce a pattern, or potentiometers, which would make tuning a bit more fiddly.
The solution was to go for a mix of small intervals, and a large one or two (and a variable one, if one goes for that option, which I recommend). The large range preserves the original Klee functionality, and also allows one to use the Klee as a normal one step sequencer with a decent range as well. In your case, you went for the semitone and octave switch option of the model 2, right?
Anyhoo, the small intervals make the thing much easier to get interesting and repeatable voltage patterns out of it. One method is to program a pattern in the 1 volt range, and then try the lower intervals - often the voltage pattern is transformed in some very cool ways.
Quote: | This would do two things: 1st) Allow positive and negative with the octave scale. 2nd) I noticed with the Octave and Semi scale switches both set to minimum there is no CV at all, which makes sense because this is the only supply to the CD4066 bus. Scaling on the output with a fixed MaxV would guarantee that you would always have some CV going into the CD4066 switches.
I'm not sure if I am thinking clearly here but it seems like since the shift register bits are all summed together your are going to get multiplication of of the scale voltage, not just the individual pot voltage settings. |
No, you should only get the pot voltages. If a pot is set to produce zero volts, it will contribute 0V to the pattern. The bit selects a CD4066 to pass the selected max voltage. This max voltage from the CD4066 goes to the top of the pot, the center tap of the pot contributes the voltage to the voltage mix. If the pot is turned all the way down, nothing is contributed, and if it is turned all the way up, it contributes the max voltage. Any setting in between full CW and full CCW of the pot will contribute that voltage (between the ranges of 0V and the max voltage)
So, if you have a bit pattern containing 7 bits, but only one pot set to contribute, say 1 volt, throughout the pattern a V/Oct VCO will jump up an octave seven times. If you have two pots set for 1V, when the bits select both pots, you will get two octaves, when the bits select only one pot, you will get one octave, and when they are not selecting either of the pots, you will be at the base frequency of the oscillator. So, with two pots set in this instance, you should get a pattern containing a combination of three octave notes.
If the second pot is set halfway, it will contribute only a half an octave, so now the voltage pattern should be a sequence that spans four notes - the base note, half an octave up, an octave up, and an octave and a half up.
Quote: | With this in mind, I think that you would only get true octave scaling when there was only one bit set. By Scaling in the output amp section, the multiplication of the scale voltage would not occur. Am I making any sense here? Another thought occurred that by scaling in the output you could do the A, B, and A+B channels independently. Maybe all three channels could be driven by the existing Octave and Semi rotary switches and you could add a center off toggle switch to each channel that would allow you to go +/- an octave separately. |
Well, the scale voltages shouldn't be multiplying - only the voltages tuned by the pots should be contributed to the mix. Is this the case, or am I misunderstanding your question? In other words, I rarely have anything go ultrasonic unless I have something greater than 1V range, and a high number of bits and a fairly large number of pots cranked up.
Quote: | Overall I am extremely pleased and after I have gotten some sleep and my eyes have cleared up from the solder smoke I want to really dive into some serious Klee patching and sound generation. Thanks again for your design, help, compliments, and encouragement! |
Tanks for the encouragement! Here's a PDF of the operating tips I sent to the (other) proto-testers. As I told them, it's not a bible or anything, but it's how I approach programming the Klee. I'm certain there are other methods, but this will get one started. There's also a "Know the Klee" doc - if you haven't seen it, it's right here:
http://electro-music.com/forum/download.php?id=8646
It explains this probably a little better than my pawings of this post.
In this tips doc, you'll see a mention of "golden pot settings". I'll often come across a set of pot settings that just never stop producing very cool voltage patterns. I've got such a set right now on the breadboard - I'll write the settings down and send them to you and see if that helps any.
Take care,
Scott
Description: |
|
Download |
Filename: |
Klee_Ops_Tips.pdf |
Filesize: |
9.22 KB |
Downloaded: |
3863 Time(s) |
_________________ My Site |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
Posted: Wed Aug 22, 2007 9:08 am Post subject:
|
|
|
Hey Scott,
Thanks for the Op Tips PDF. That will help greatly I'm sure. I think I was way too tired last night and should not have posted my ramblings. I need to work with the Klee a lot more to really get the feel of what I was seeing. Without a doubt I can see that the potential is incredible. A lot of my experience was probably operator error! BTW I am using Option 3 for the scaling.
Regards,
Tim |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Wed Aug 22, 2007 6:41 pm Post subject:
|
|
|
Okee-dokee, Tim. Here are the pot settings I have on the breadboard. The last several samples I've posted used them.
1 = full on, 0.5 means set halfway, 0 is all the way off.
Range: 1V (1 Octave).
Pots:
1. 0
2. 0
3. 0
4. .5
5. 0
6. 1
7. 0
8. 0
9. 0
10. 0
11. 1
12. 0
13. 1
14. 1
15. .5
16. .5
With this setup, I like to run one voice with output A and the other voice with output B. I generally use A+B to control the cutoff of one of the filters with these settings, and one of the other outputs to control the cutoff of the other filter.
Start off with a pattern that has between 5 to 7 bits. Every so often, flip between 8X2 and 16X1 (or vice versa) just to randomly juggle the bits.
For the Gate Bus, I'll experiment - generally I will have no more than 1 or 2 switches set for Bus 1 and one or two set for Bus 3 - the rest are in the center Bus 2 position.
I like to patch triggers and gates from separate bus outputs for a single EG (and do it for both voices). Like, for instance, the gate from bus 3 and the trigger from Bus 1 will control one EG, and the other EG will be set up with some other combination. Lately I've been patching the Voice A VCO's sine straight into the mix, or I will run the sawtooth through another filter and control that with a third EG. I then mix it in and out (as I do with all the voices at different times). Play with the EG settings extensively, too. Do the merge switch thing - trigger an EG only with triggers from a merged/unmerged bus.
I'll tune the oscillators up and down, usually with short, stabby EG settings til a relationship catches my attention. Generally I'll tune the A voice higher and the B voice lower, but not all the time.
Main thing is, just experiment to your heart's content. Don't be afraid the twist a pot here and there while you have a pattern going. Or shuffle bits around with the load function, too. When you have a good groove going, start playing with the range switches without touching anything else. Crank a filter's cutoff intermittently. Modulate the filters with LFO's - nice interactions with the pattern often occur.
Especially if you run the voices out to a delay, man, you'll find yourself hearing the birds chirp at dawn before you know it. Your significant other will eventually become confused with your unexplained absences in the la-bohr-atory....
Take care,
Scott _________________ My Site |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
Posted: Wed Aug 22, 2007 9:17 pm Post subject:
|
|
|
Wow, thanks Scott for the details in your last post.
My mistake on the first test run last night was to approach it like a normal seq. I set all of the pots so that it walked up the scale which sounded normal with one bit set. Then I started adding bits and it got out control fast as you can imagine. I also was gating only on the main so it got to be pretty brutal for me and even more so for the cat!
Anyway, tonight was a totally different story. I started out following your Ops Tips (few switches, few pots with low pot settings) and setup two voices with different gates and CV buses. Amazing, it's now starting to remind me of a Scott Klee sample. I think the cat even liked it this time!
I'm not ready for prime time posting of MP3 samples yet, but soon. I think if I had built the Klee 1 first and gotten the feel on that, then the migration to Klee 2 would have been more natural. Man, there are a lot of switch and knob combinations on the Klee 2! As you said, one small tweak of a knob or switch can really change the whole character of a sequence.
I am certainly going to play with the values in your last post and put it through it's paces. I quickly found out tonight that I need more patch cords now.
Regards,
Tim |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Thu Aug 23, 2007 9:23 am Post subject:
|
|
|
Woo hoo!! Fantastic. You're the first and only person I know of to have built a model 2, perfboarded no less. My hat's off to ya.
Take care,
Scott _________________ My Site |
|
Back to top
|
|
|
mono-poly
Joined: Jul 07, 2004 Posts: 937 Location: Rotterdam, Netherlands
Audio files: 2
|
Posted: Thu Aug 23, 2007 12:53 pm Post subject:
|
|
|
Now you should post some demo's ofcourse |
|
Back to top
|
|
|
Luka
Joined: Jun 29, 2007 Posts: 1003 Location: Melb.
|
Posted: Thu Aug 23, 2007 4:28 pm Post subject:
|
|
|
Tim could you post a picture of your completed perf board layout design
it would be most awesome if you could
congrats on the build success |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
Posted: Thu Aug 23, 2007 5:19 pm Post subject:
|
|
|
Hi Luka, I posted some close-up pictures of the top and bottom of the perf board. Did you see those? I didn't do any formal drawings of the layout. I just built it out section by section, according to Scott's schematics. In fact, I was sweating it towards the end wondering if it all was going to fit. What kind of information are you looking for?
Regards,
Tim |
|
Back to top
|
|
|
Luka
Joined: Jun 29, 2007 Posts: 1003 Location: Melb.
|
Posted: Thu Aug 23, 2007 7:37 pm Post subject:
|
|
|
yeah i notice those
just wondering what changes you made as you said it wasn't 100% functional at hat stage
reason being i have only got pcb on order and now realise i would like 2 so im trying to get some good solid perf board layouts before i do it myself |
|
Back to top
|
|
|
Randaleem
Joined: May 17, 2007 Posts: 456 Location: Northern CA, USA
|
Posted: Thu Aug 23, 2007 10:42 pm Post subject:
|
|
|
Luka wrote: | i would like 2 so im trying to get some good solid perf board layouts before i do it myself |
Hi Luka,
Uncle Krunkus did stripboard layouts for the Klee 2 project before he laid out the commercial em Klee PCB's. They're available in the main Klee thread. They work, but there have been a few minor changes since they were laid out. So they're a Great starting point if you don't mind stripboard!
Best regards,
Randal |
|
Back to top
|
|
|
Luka
Joined: Jun 29, 2007 Posts: 1003 Location: Melb.
|
Posted: Thu Aug 23, 2007 10:49 pm Post subject:
|
|
|
cheers randal
i guess i probably have seen them then and forgot
been going nut all over the interwebs for the last month researching my modular and starting to go nuts ;P
ill check it out |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Aug 24, 2007 8:04 am Post subject:
|
|
|
Dude!
You're Kleeing!
Very nice sample - thanks for posting that. It didn't take you long for you to get into the groove, my first samples were usually just a collection of unrelated blonks and thwips, you're already getting some nice lines together. For me it's fantastic to hear a Klee sample I'd never heard before a zillion times before I recorded it. If I had a Klee trophy, I'd send it to you.
Take care,
Scott _________________ My Site |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Aug 24, 2007 8:43 am Post subject:
|
|
|
Hey Luka,
The file is probably still missing after the crash - it needs a bit of correction to the schematic. I may upload it with the note and fix it later, as I'm pretty tied up with getting the proto done.
Also, I can add your name to the seconds list of the Klee - if someone for some reason or other doesn't purchase their reserved boards, then I turn to this list.
It's also possible there could be a second run if there's enough interest.
Cheers,
Scott _________________ My Site |
|
Back to top
|
|
|
synthmonger
Joined: Nov 16, 2006 Posts: 578 Location: flada
Audio files: 3
|
Posted: Fri Aug 24, 2007 10:31 am Post subject:
|
|
|
Me = jealous of your perfboard. It's just "Superawesomegreatradicalkillerbosssick"!
Would it be possible to make a track without the use of reverb? |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Aug 24, 2007 11:03 am Post subject:
|
|
|
Quote: | Would it be possible to make a track without the use of reverb? |
For me, it nearly is but most of the early Klee demos were done with no effects, so the Klee wouldn't be confused with the effect. Most of them are probably gone out of the Klee thread, but my stunted Klee page has them here:
http://mypeoplepc.com/members/scottnoanh/birthofasynth/id24.html
They're more or less in chronological order, and none of the early samples used any effects. That was a limitation I placed on demonstrating the functionality, but generally I'm a reverb/delay nut of the first order, meself, when it comes to most synth stuff.
Cheerios,
Scott _________________ My Site |
|
Back to top
|
|
|
mono-poly
Joined: Jul 07, 2004 Posts: 937 Location: Rotterdam, Netherlands
Audio files: 2
|
Posted: Fri Aug 24, 2007 11:12 am Post subject:
|
|
|
I like your demo man!
Thanks for making it and putting it online! |
|
Back to top
|
|
|
synthmonger
Joined: Nov 16, 2006 Posts: 578 Location: flada
Audio files: 3
|
Posted: Fri Aug 24, 2007 11:58 am Post subject:
|
|
|
Thanks Scott but I've already downloaded all those Klee mp3's on your site. Your beat making skillz are fresh enough to have a folder in my music collection
I'm a reverb/delay nut too, but for me it's easier to hear the sync differences without any effects. That's what I'm mainly interested in. |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Aug 24, 2007 2:06 pm Post subject:
|
|
|
Ah, yeawp. Sync differences are easiest to hear with really short attack/decay, too. Or best, when using drum voices.
I'm going to have to find something around here and fashion into a Klee plaque for Tim. I'm still just thoroughly amazed.... _________________ My Site |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
Posted: Fri Aug 24, 2007 4:52 pm Post subject:
|
|
|
Thanks Scott, and everybody for the compliments.
Synthmonger, I'm sure I will be posting more samples in the near future with and without reverb. For my first sample last night I was just happy to get something pleasing to the ear, recorded, converted to MP3 and then posted.
After working with the Klee concept more and learning Scott's finesse, I will concentrate more on the production side. I'm still having trouble with the intermittent load function but I am having to much fun to stop and troubleshoot it right now.
Scott, another thing that I thought would be nice is to replace the pattern toggle switches with push buttons that set and reset as you push them. And in addition to the pushbuttons possibly add an option that would automatically load the new pattern when you touch any of the buttons. It would keep the existing pattern the same except change that 1 bit. To do this right will require a PIC and if I get around to it I would like to add another row of sixteen LEDs that would indicate the existing pattern and be updated immediately after pressing a pushbutton. This would make it really nice for adding or subtracting a bit here and there while the sequencer continues to run. I'm not going to tackle this anytime soon!
One more thing, I used 3mm red LEDs from Circuit Specialists that were $1.50 for 100 and they are plenty bright with the low current of the Klee. The part # is: BAG-RED3MM. I tried the yellow and green ones and they are useable too but not quite as bright...this might save you a little money over buying special low current LEDs.
Regards,
Tim |
|
Back to top
|
|
|
THeff
Joined: Sep 01, 2006 Posts: 229 Location: Florida
Audio files: 33
|
|
Back to top
|
|
|
|