Electro-Music Klee Sequencer Build and Applications Thread

[blue hell]

Just an idea.

Was toying a bit with linear feedback shift registers and the thought occurred to me that when you'd take a couple of XNOR ports connected to the Klee's internal step signals and feeding back into the random input the Klee could also serve as a pseudo random generator.

It could also be XOR ports, but with XNOR it would start on an all zero's internal pattern, with XOR's a preload might be needed more often. When driving it at audio rate frequencies it could even serve as a noise generator then.

Anyway, thought it might be an interesting experiment for some

[State Machine]

[State Machine]

Guys, I just wanted to reiterate for Scott the importance of using good quality switches. It is months later after my building effort and yesterday two more pattern switches just gave up while I was using it. Thats a total of about 9 switches giving up on me. I used "pulls" from other equipment to save a buck but it has bitten me big time.

Take his advice and mine and use NEW, DECENT quality switches for you panel as these will be flipped many many times in the course of Klee usage

Bill

[a.b.o.z.]

huh.
I'm a slow klee builder.
got another question.
for now I'm going for 12V operation with 10V gate\trigger levels.
in some future I'm planing on expanding of the system, adding real modules not soundlabs and moving to 15V power supply but stay with 10V gate levels.
question is: will I have to pull those resistors out (1k, R63-70 on digital board) and replace them with 1k5? Is there another way to switch to 15V operation?

And another one: It is about .1% resistors. I put regular 1% resistors in there but i picked up 25 of them with same value or with .1% drift. So i got 99.5 and some are 99.6
When i soldered them to the board and try to measure some of them turn out on digital meter to be like 91.6 or even lower? Why is this? I'm confused. Is this ok to be like this?

Thnx.

[factus10]

[a.b.o.z.]

It's aliveeeeeeeeeeee!!!!!!!!!!!!!
very nice

My second real project and only one bug...
When clock is NOT active and I press manual step button Klee freezes... huh?
That action also freezes soundlabs LFO even when LFO is not connected to Klee. Actually whole soundlab goes crazy..humming and sounding ill. Like there's something wrong with PSU.. Klee and Soundlab share same power supply...
To get everything back to normal I have to plug out mains and re-apply power.
That is single problem.
Everything else is working a ok.
what gives?

[Scott Stites]

Congrats on the build!

As for the problem: here are a couple of investigative suggestions:

1. With the clock stopped, monitor the positive rail - see if it drops when you press manual step (it shouldn't).

2. If possible, disconnect the Soundlab temporarily from the power supply and try the function to see if there's any difference.

Let me know what you find, and we'll go from there.

Take care,
Scott

[a.b.o.z.]

thnx scott!

Just like you said it... it drops down to 4.7V but I can't disconnect soundlab from PSU now, because it is "part" of it... but if it's needed to troubleshoot it truly I'll do it..

cheers


Ivan

[Scott Stites]

Thanks for the info. The manual step function works by connecting your positive rail to ground through R28, which is a 470K resistor. This should only pull around 32 microamps through the resistor, assuming your rail is 15V. R28 is acting as a pull-up resistor. It keeps a positive voltage applied to C12 and the input pins 1 and 2 of U4, which holds the output pin 3 low while the step switch is not pressed.

When you press the step switch, R28 goes low - this pulls the input pins of U4 low and the output pin high so that it creates a high, which causes the Klee to step forward one step (it also serves to hold the master gate output high as long as the switch is pressed). The charge on C12 is also dumped to ground. When you release the switch, R28 goes goes high and charges C12 back up so that any release bounces don't cause retriggering. When C12 is sufficiently charged, the output pin 3 of U4 goes low, allowing another step function to be performed the next time the switch is pressed.

Anyway, 32 microamps should not affect the rail voltage. Double-check R28 to make sure it is indeed 470K.

If you have an oscilloscope, check to make sure your rail doesn't actually begin oscillating for whatever reason - that would certainly cause a DMM reading to indicate something other than what you expect, and could cause mayhem in the Klee and Soundlab.

In the meantime - what type of power supply are you using, and what are the rail voltage and current ratings on it?

Good luck and let me know what you find - whether it's something or nothing.

Take care,
Scott

[a.b.o.z.]

ok I found out that my R28 is 220K.
Im using MFOS bi-polar 12V PSU.
I'll change this fella tomorrow and let you know result.

thnx for debuging


Ivan

[Scott Stites]

Whoa whoa whoa! Mistake on my part - it should be 220K - that's the right value. I must have had 470K on the brain. Let me think about this some more.......

[Scott Stites]

Here's something to try:

Take a test lead (alligator clip, etc,) and clip it onto the positive post of the step switch. Clip the other end of the test lead to a 1K resistor. Now, with no clock to the Klee, tap the free end of the 1K resistor to an available piece of ground. Don't push the step switch - we're just imitating pressing the switch with some extra resistance in the path. My panel is grounded, so I just tapped the end of my resistor there. Each time I tapped the free, unclipped end of the resistor to ground, the Klee advanced one step.

Now, with yours, if you experience the same garbled behavior, grab a higher value resistor and repeat the process. Eventually you will either reach a point where the Klee actually steps and doesn't make your PS sizzle, or it'll just do nothing. Either way, let me know what you find.

BTW, is this the MFOS wall wart supply?

Sorry again about that resistor value....

Cheerio,
Scott

[a.b.o.z.]

ok, I didn't replace it yet. Will chek this later today...
and psu is this one: http://musicfromouterspace.com/analogsynth/powersupply.html

[State Machine]

Wow, guys, pretty tough to swallow that the extra 54 uA is sucking the 12V rail down to 4 or so volts Please try this also. With no power applied to your Klee, hang an ohm meter across the 12V and ground then push and hold the STEP pushbutton and see where that resistance goes.

Bill

[Scott Stites]

Agreed. I'm not sure why this is happening, but it is reminscent of something I ran into when I was first designing the clock and load section. With a particular configuration of values, I ran into rather the same set of squirrely symptoms that Ivan is experiencing (though I do not recall the rail dropping). This, oddly enough, had to do with how close the switch ground feed was to the main ground input of the breadboard. If I moved the ground point of the switch away from the main feed, the problem went away. Having said that, I did change the values/configuration after that, and never had the problem again. During proto, that section got even more beefing up. Still, I thought it would be interesting to see if grounding the switch at some other point made any difference, in the off chance this is what Ivan is seeing.

Cheerio,
Scott

[Scott Stites]

Any luck, ABOZ?

[a.b.o.z.]

didnt find the the time..jet. soon

[Scott Stites]

Okee doke

[blue hell]

Time jets I know what time flies are

[Photon]

Well I finally have completed enough synthesizer that it makes sense to start building my klee. This morning I started cutting metal and the 2 control panels are looking good. They're a bit scratched up (recycled aluminum-rescued from the bin) but it will be covered by a label so no big deal.

This Klee incarnation will be a wedge-shaped desktop console, the control interface split into 2 panels. The programming elements will be on a 15˚ downward sloping panel and all other controls on the rear horizontal panel with all jacks in rear as to not obscure controls.


here's a photo of the panels and a jpeg of the graphic layout.

more soon as progress permits.

peter

---

klee-graphics.jpg
Description:
Filesize:	317.56 KB
Viewed:	194 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

---

klee-panels.jpg
Description:
Filesize:	114.67 KB
Viewed:	181 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[loss1234]

looks like a real nice design!

i must ask you and everyone else, how do you cut the slider holes? a router?

a drill? a special punch?

i am going to be starting my klee this spring and i am not sure if i should go with pots or sliders.

thanks

[Photon]

[blue hell]

[Photon]

I have access to a milling machine at work.

However, if you look back though this thread you'll see that others have been able to do a very good job at home using hand tools. It will just take much longer.
p.

[blue hell]