Electro-Music Klee Sequencer Build and Applications Thread

mark_olson · Joined: Oct 26, 2006 Posts: 177 Location: Lawrence, Kansas

You all probably think Scott is trying to be funny but he's not - everything he says about the Wichita experience is true. Wink

Sound · Joined: Jun 06, 2006 Posts: 842 Audio files: 1

Wiring my Klee, I have noticed, that I put an extra led in the frontpanel: The LOAD led. Surprised

I don't know why I did it, Surprised

Well, my question is: It will work connecting the led to pin 10 of the U3 CD4071 through a 6,8K resistor? (schematics digital board page 3 of 4)

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.

Scott Stites · Posted: Sat Nov 29, 2008 9:13 am Post subject:

I know the reason why you did it - because it makes sense. Not putting a Load LED in has long been one of those things that makes me want to kick myself. Actually, not putting in an External Load LED is the thing that makes me self-flagellate. It would be really useful to be able to see the rate of the re-load signal if one is doing that.

Putting the LED signal at pin 10 of U3 would probably not cause it to light at all - the external load and load pattern switch signals have been narrowed to an extremely narrow pulse at that point - and as soon as that point sees a gate bus load signal, *poof* it goes away as soon as the pattern loads, so that would be too quick, too.

However, if you put the tap point at the output of pin 1 of U1 (LM358), it would light every time the Klee received an external load signal - it would stay on as long as that signal was high. It wouldn't light with a manual load, or a gate bus 1 load, but an LED wouldn't be all that useful on those situations - with a manual load, you know you're pressing the key, and a gate bus 1 load is never going to illuminate for longer than a zillionth of a second without adding some stuff to slow it down. Putting it at pin 1 of U1 would give you the most useful info of all the load operations though, and that is it would tell you when the Klee was externally told to load, which is the visual info I flying-tiger kick myself daily for not putting in.

I'm thinking going straight through a resistor to the LED from pin 1 of U1 would not foul up the works. 6K8 might do it, or even 5K6, since the output at that point is somewhat lower than rail voltage. Haven't tried it, but now that you mention it, I think it would. That IC can supply enough current at that point. I say give it a whirl - I'd be very interested to know if it works (which I think it will).

Take care,
Scott

Uncle Krunkus · Posted: Sat Nov 29, 2008 3:30 pm Post subject:

Didn't I tell you Scott?!?
I said "If I've told you once I've told you a thousand times, why have a master gate bus LED and no external load LED? It doesn't make sense! Chewbacca,......... Endore,...... It doesn't make sense!!!!"
I did tell you that didn't I?
I'm sure I did.
Didn't I?
Maybe it was,..... ummmm,.......

Scott Stites · Posted: Sat Nov 29, 2008 10:07 pm Post subject:

Hmmm....I remember you mentioning that the clock and gate bus 1 LED had essentially the same info; I didn't want to get rid of either because I use them both. I don't remember you mentioning anything about an external load LED, but that doesn't mean you didn't, I just don't recall it.

Cheerios,
Scott

Uncle Krunkus · Posted: Sat Nov 29, 2008 11:28 pm Post subject:

Of course I didn't say anything about an external load LED! Rolling Eyes

Scott Stites · Posted: Sun Nov 30, 2008 7:12 pm Post subject:

Sound · Joined: Jun 06, 2006 Posts: 842 Audio files: 1

It works!!!

Some mistakes like the clock switch ON (ON) was reversed, the bus 3 is the 1 and the one is the 3.

Scott, the External load led is working from pin 1 of U1 through 5,6K resistor. The bright is the same that the other LED.

There is an inconvenience, is that due I strap wiring the front panel with non insulated wire, there is any short when I put the front panel in the case because the wires that connect the PCB with the front panel push the strap wires and some led don't illuminate. So I will demount it looking for a solution.

After that I will keep reading step by step, the chapter 9 "electro-music Klee Sequencer Bring-Up Procedure" of "Building the electro-music Klee Sequencer" documentation as I did with the rest of the chapters.

This documentation is very complete, clear and educative, and with it, a beginner like me is achieving build a Klee.

So, T H A N K Y O U ! ! !

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

Folding the case, with some clamps, woods and a hammer. Aluminum 1mm thick:

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

The Klee:

Uncle Krunkus · Posted: Mon Dec 01, 2008 3:28 am Post subject:

Beautiful job there Sound. Very Happy

Scott Stites · Posted: Tue Dec 02, 2008 9:58 am Post subject:

No kidding, that is one smart panel and wiring job! How's the cal going?

vtl5c3 · Posted: Tue Dec 02, 2008 10:22 am Post subject:

Can someone point me to the post that describes the A-B/B-A mod?

Scott Stites · Posted: Tue Dec 02, 2008 11:16 am Post subject:

That one landed in the panel thread. I think this link will get you there (it's towards the bottom of the page):

http://electro-music.com/forum/topic-21954-250.html

vtl5c3 · Posted: Tue Dec 02, 2008 11:50 am Post subject:

Ahh... Thank you, sir!

Looks like a good mod. Too bad my panel space is all used up already.

EdisonRex · Posted: Tue Dec 02, 2008 1:57 pm Post subject:

I finally got around to (ie I got my confidence up enough to) getting the aluminium rack panel, drilling it, putting the graphics on, and mounting the components. It looks great, but I'm not going to post a pic until I have some backside work to show.

So next up is the wiring. I'm getting together some coloured wire (all I have is black and I don't think it's going to work). Although would ribbon cable work better? I have a lot of that to use and I was thinking it might be easier to manage.

Then it's just sorting out the PSU (I have the parts but need to put it together)... I might be a lot more motivated after wiring the mess.

Who has done ribbon cable instead of hookup wire? Pros/cons?

Uncle Krunkus · Posted: Tue Dec 02, 2008 2:30 pm Post subject:

I did ribbon cable on just about everything.
I used rainbow stuff where there were different types of signals/connections on the one molex, and standard grey IDE type ribbon for the buses. You only need to know which side is which on one of those, as they are sequential. It all worked out nice and neat. Mosc used the pic on the shopfront if you want to check it out. It's in one of the threads too, but I'm not sure where.

EdisonRex · Posted: Tue Dec 02, 2008 2:36 pm Post subject:

Hmm, thanks for that. I like the ribbon cable idea, so I'll go find your pics.

blue hell · Posted: Tue Dec 02, 2008 2:41 pm Post subject:

EdisonRex · Posted: Tue Dec 02, 2008 3:13 pm Post subject:

Oh, yay! Thanks Jan. This is very helpful!

Sound · Joined: Jun 06, 2006 Posts: 842 Audio files: 1

Scott Stites · Posted: Thu Dec 04, 2008 8:52 am Post subject:

First of all, I'm surprised anyone in this world can still go to a local store and buy an MC14034!

I've never tried an MC14034, but I generally use MC1 and CD parts interchangeably with no problem. I'll peruse the datasheets to see if there's any diff, but I think if you've gotten this far with them, they're working.

Gate Bus 1 Load on Bus 2 is an odd thing. The function can surprise you because a bit that's high somewhere else on Gate Bus 1 may cause a reload when you're not expecting it. Does this ever happen if you do not have any of the gate bus switches set to gate bus 1?

Load Switch: This switch should ground out R27 (220K) which sends a pulse through R19, which is squared off by U2 (input pin 5, output pin 6). Actually, this is the "kinder" of the two momentary switches, because debouncing on it is not critical - in other words, if the CD4034s get two extremely fast pulses because of bounce, they just load the same pattern twice, which, on a human scale, is just once.

On my Klee, when I first turn it on, it has to be clocked at least once before a pattern can be loaded. This is due to the initial on state of my particular brand of CD4013 (U5). Once I've done that, it loads reliably. Is this what you're encountering, or is it deeper than that?

If so, we need to figure out if the load system is working properly. The best way to do that is to see if you can reload reliably using an external signal on the external load port. Do you have a pulse LFO you can connect to that port and see if the Klee is re-loading with each pulse? If so, clock the Klee at a fairly high rate in 16X1 mode and set the reload signal to a freq that's around half of the clock signal. The Klee should reload repeatedly and consistently before it can get through the entire pattern. It that works as expected, we can focus more on the two anomolous anomalies that are anomalating your Klee.

Cheers,
Scott

Sound · Joined: Jun 06, 2006 Posts: 842 Audio files: 1

Sound · Joined: Jun 06, 2006 Posts: 842 Audio files: 1

Well,

I have tested the PCBs following step by step the chapter 8" Final Assambly of the Klee" pag91of Building the Klee. All was OK.

Due the bus 2 Is ON when the buses 1 and 3 are OFF, I thought that maybe there is a short within, Logical gate bus2 section after pin 4 of U13 to the front panel (pag 4/4 digital board schematics and pag 5/9 front panel schematics) and the Load in section, before pin10 U3 to the front panel. (pag 2/4 digital schematics and pag 3,6/9 frontpanel schematics)

I have tested all the points of these sections with each other point of the other, with the multimeter and I didn't notice any short.

After I changed all the IC involved in the Load section and the Logic gate bus 2 section.

I don't understand.... well any advice will be welcome. Very Happy

Ah, I remember that I didn't do a good job when I was soldering the components, I think that I overheated all them. Maybe I damaged anyone?

Scott Stites · Posted: Thu Dec 04, 2008 1:08 pm Post subject:

Here's a quick test I'd like you to perform:

Set the Klee to 16X1, and make it a 1 bit pattern. Set all the gate bus switches to bus 3. Observe your gate bus LEDs - only the gate bus 3 LED should flash.

If you've confirmed only gate bus 3 LED flashes, set gate bus 1 load on. Then clock the pattern and see if the load happens on a particular step or steps. If it doesn't load at all or immediately loads, let me know.

Then, set all the gate bus switches to gate bus 2 and try the same procedure, first observing if only Gate Bus 2 LED flashes, then do the gate bus 1 load switch on thing again. Again, let me know what you find out.

Of course, make sure the gate bus merge switches are not set to "merge" - if they're set to merge for a particular gate bus, the gate bus LED all of these switches are set to will stay on constantly.

D5, D6 and U1 (pins 1 and 2) of the Digital board form the logical gate bus 2 function. My thinking is, if one of these diodes is kaput, or backwards, you're going to get some weird action going on it could feed a signal back into gate bus 1 or gate bus 3.

Let me know!

Cheers,
Scott

Scott Stites · Posted: Thu Dec 04, 2008 1:20 pm Post subject:

D5 in particular. If that's shorted, gate bus 3 can feed back to gate bus 1. D5 blocks that from happening.

Edit: shorted or backwards, that is.