Author |
Message |
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Tue Sep 15, 2009 5:42 am Post subject:
|
|
|
Yes, dan, I am driving the chip with a level-shifted VCO. This is because we want to be able to vary the clock rather than use a fixed clock or one that is only adjustable by potentiometer. You'll see the VCO input gets buffered, then it drives the control input of the LM13700 VCO itself, followed by a level shifter and then into the OX1 pin. The inverter chain that would normally drive the oscillator just does some buffering in this case. It's not documented in the datasheet, but I think it will work OK.
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
Posted: Tue Sep 15, 2009 6:53 am Post subject:
|
|
|
hmmm
as of now I can NOT get this to work. It could be my fault.
first problem is the 13700 is bleeding really badly into the BBD chip.
this may be because it is on +/-12 and is a very strong signal.
(even when it is not connected to the MN3207 I can still hear it)
the other problem is that the 13700 does NOT seem to be making the clock chip
create a clock.
I could be doing something very wrong.
I will keep at it, but so far, My MN3201 makes a clock when I clock it with a resistor and a cap but when I try it with the OTA, i get nothing.
when you tested this, did you run it on +/- 12 or +/- 15?
thanks _________________ -------------------------------------------- check out various dan music at: http://www.myspace.com/lossnyc
http://www.myspace.com/snazelle
http://www.soundclick.com/lossnyc.htm http://www.indie911.com/dan-snazelle |
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
Posted: Tue Sep 15, 2009 7:40 am Post subject:
|
|
|
ok last comment
I am now able to get a better clock out of the MN3102
i switched the 33k on the I in to a lower R
HOWEVER there is still far too much bleed at this point
i cant seem to find a way to get rid of it....for now
However this could be a problem with BREADBOARDING and not something that would ever be an issue on a PCB>
INVENTOR...please let me know what kind of results you have had
thanks so much
dan _________________ -------------------------------------------- check out various dan music at: http://www.myspace.com/lossnyc
http://www.myspace.com/snazelle
http://www.soundclick.com/lossnyc.htm http://www.indie911.com/dan-snazelle |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Tue Sep 15, 2009 3:29 pm Post subject:
|
|
|
Dan,
Are you feeding the LM13700 directly into the MN3102 chip? It can't take that. You should be powering the LM13700 on +/- 12 or +/- 15 Volts and the two BBD chips on +5 from the regulator. I'm guessing that you did not include the level shifter perhaps, the little circuit with the 2n2222 transistor?
Can you describe in greater detail what you mean by "bleeding", I don't quite understand.
I have not prototyped this, I'll be ordering parts soon to build it.
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
Posted: Tue Sep 15, 2009 4:23 pm Post subject:
|
|
|
well when i say bleeding i mean that the vco is bleeding into the output.
it is driving the chip but BBDs are so sensitive that i think having a clock with such high rails on the same breadboard is causing problems
maybe i did something wrong.
but i have built quite a few other circuits this week successfully.
it is possible that driving the lm1700 off lower rails would help
when you get a chance, just try the vco, mn3102 and mn3207 and see if you have any problems
thanks so much ( i really do want to get this working!!!) _________________ -------------------------------------------- check out various dan music at: http://www.myspace.com/lossnyc
http://www.myspace.com/snazelle
http://www.soundclick.com/lossnyc.htm http://www.indie911.com/dan-snazelle |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Tue Sep 15, 2009 4:33 pm Post subject:
|
|
|
What do you mean by "bleeding"? _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Tue Sep 15, 2009 4:41 pm Post subject:
|
|
|
OK, this is a debugging issue and is not to be unexpected. Someone mentioned the concern before. The solution might be to put a cap from the input of the LM13700 VCD to ground, right after the BBD stage. I'm not sure what size of cap would do. This will add a second filter that will hopefully quench the clock noise. Hope that helps, let's work out the details together, be patient!
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007 Posts: 1536 Location: nyc
Audio files: 41
|
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Wed Sep 16, 2009 3:53 am Post subject:
|
|
|
Well, I just ordered the parts for the prototype KS board this morning, so I should be able to begin prototyping next week.
About the clock noise, it occurs to me that we should also put a small cap in parallel with the feedback resistor of the loop summing amp. That, and a cap on the input of the VCF should help the problem, though I do not know if that will solve it.
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Thu Sep 17, 2009 4:10 am Post subject:
|
|
|
Upon revisiting Dan's observations, it occurs to me that "bleeding" is indeed the correct term for what might be going on here. Dan mentioned that removing bypass caps actually helps the problem, and that it might be an issue with breadboarding limitations.
That seems to be the case. The datasheet for the LM13700 does not describe the frequency of oscillation, but the cap is so small that I gotta think it's in the 100 kHz region or thereabouts. That won't fly too well on a breadboard with all the parasitics and stuff, so here's what I would do if I were Dan:
Slap a bigger cap on the VCO...
Slow that puppy down, all the way down to 10 kHz, which is the minimum frequency that the part supports, creating the maximum delay. Sure you won't be able to fully test it across the frequency range, but hey what do you want, it's just a breadboard.
Just this morning's observations, true or false...
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Tue Sep 22, 2009 3:40 pm Post subject:
|
|
|
Update: I got my DigiKey parts yesterday, still waiting on SmallBear parts but they take a little longer. I'm patient. So soon I will be able to breadboard the Karplus-Strong board.
Some recent ideas:
1. The datasheet for the mn3207 shows an active filter after the delay line that removes this clocking noise apparently. I could just unbuffer the stim input, feeding it directly into the KS summing amp, then use that opamp for the active filter. That might be the best thing to do to properly quench the clock noise.
2. Since the KS and BS boards will be the same size, one could stack up a sandwich of one of each, BS on top and KS below. These "bricks" would each create one voice of a sort of eChucK-style rearrangeable synth, in which you place the bricks on a table and wire them up with jumpers. The boards are also made to be panelized for use in a more conventional synth cabinet.
3. We seem to have thought up a third companion board to the KS and BS boards. It's an SS board, for Song Sequencer. If we make this one, it will perform song-length macroscopic sequencing of the various voices. See the Song Sequencer thread under DIY/Developer's Corner for details.
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
|
Back to top
|
|
|
wmonk
Joined: Sep 15, 2008 Posts: 528 Location: Enschede, the Netherlands
Audio files: 15
|
Posted: Mon Oct 05, 2009 12:41 pm Post subject:
|
|
|
How's the project going? Did you prototype it?
I bought a few delay chips today. So I'll test it out next weekend.
Quote: | 1. The datasheet for the mn3207 shows an active filter after the delay line that removes this clocking noise apparently. I could just unbuffer the stim input, feeding it directly into the KS summing amp, then use that opamp for the active filter. That might be the best thing to do to properly quench the clock noise. |
I looked up the datasheet, but I didn't see an active filter. Could be that the datasheet was written in japanese, but there was no schematic. Do you have a link? |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
|
Back to top
|
|
|
wmonk
Joined: Sep 15, 2008 Posts: 528 Location: Enschede, the Netherlands
Audio files: 15
|
Posted: Mon Oct 05, 2009 3:16 pm Post subject:
|
|
|
Well, that's a LM13700, not the MN3207 you mentioned
An OTA based filter is nice, but I'm not sure if it's ok to use it to remove the clock signal. What is the frequency range of the clock signal? If high enough, there is no problem, as the audible frequencies will pass through. |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
|
Back to top
|
|
|
wmonk
Joined: Sep 15, 2008 Posts: 528 Location: Enschede, the Netherlands
Audio files: 15
|
Posted: Mon Oct 05, 2009 3:51 pm Post subject:
|
|
|
Nobody's going to hear it? I think the overtones will make a difference in sound. But if you set up the this filter at around 25kHz, there is no big problem. You use another filter with a much lower frequency offset in the signal, so there aren't many high frequencies left.
I'll check out the datasheet. Thanks |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Mon Oct 19, 2009 12:42 pm Post subject:
|
|
|
Today I got around to doing some prototyping on the KS board. I hooked up the power, made the 5V regulator circuit work, wired up the oscillator and it's bias control opamp, then - no sound! Blue Hell (Jan) helped me diagnose the problem. The feedback resistor was the wrong value (should be 10k, not 20k), and also I had the positive and negative inputs reversed. I should be more careful when I create schematics. Oh well, at least it works now.
The oscillator cap is way too large, as expected since the datasheet had little or no description of the circuit. It only oscillates from 200 Hz to 2kHz, so the cap needs to be way smaller. Also the output is a triangle wave, which makes sense in retrospect, and that's OK because the level shifter and clock chip will square it up and fix the duty cycle.
Overall good progress was made and I'll work on the rest of the circuit soon.
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
wmonk
Joined: Sep 15, 2008 Posts: 528 Location: Enschede, the Netherlands
Audio files: 15
|
Posted: Mon Nov 09, 2009 12:51 am Post subject:
|
|
|
How's the prototyping going? Got some extra delay chips last saturday. Now I've to get another breadboard to test things out. Any updates on the schematic? |
|
Back to top
|
|
|
Inventor
Stream Operator
Joined: Oct 13, 2007 Posts: 6221 Location: near Austin, Tx, USA
Audio files: 267
|
Posted: Mon Nov 09, 2009 2:00 am Post subject:
|
|
|
wmonk wrote: | How's the prototyping going? Got some extra delay chips last saturday. Now I've to get another breadboard to test things out. Any updates on the schematic? |
wmonk, the prototyping got put on hold because of the EM09 festival, but I do have some updates for everyone, which I will discuss here.
I was able to prototype the VCO. It didn't work so I discovered that pins 3 and 4 were incorrectly reversed. That is, 3 should be 4 and 4 should be 3. In other words, I got the inputs to the OTA backwards. The circuit will not oscillate if the pins are not correctly connected.
Also the capacitor of the VCO, as expected, is way too large. The datasheet did not provide any frequency formula or text description of the circuit, so I knew it would require adjustment. I believe it needs to be 100 or 1000 times smaller, something like that - I forget.
At EM09 I spoke with Rob Hordijt (sp?) who happens to be an analog guru and quite familiar with Karlus Strong. He drew me a schematic of this freaky way of making the loop such that the tonal quality of the circuit is superior. I will try to draw out his schematic soon, and I'll be pm-ing him for details as we go.
Last night in the chat I spoke with ScottG, and sort of evangelized him onto the project. It turns out he is a big KS fan from his FPGA music creation experiments (which sound great BTW). ScottG will likely take over the breadboarding and I will continue with the schematic capture and layout.
In addition to helping out with the breadboarding, ScottG showed me a web page which describes using a comb filter at the output of the KS loop. Here is the page:
http://www.fpga.synth.net/pmwiki/pmwiki.php?n=FPGASynth.DigitalWaveguide
Scroll down to the illustration in the section marked Phase 2 for a block diagram. It has separate controls for pitch, timbre, and get this - pickup position! How cool is that?
So in summary, we have significantly improved the quality of the design, but at the same time this puts us a step back in schedule. Hopefully with ScottG doing some prototyping we ill move forward more quickly. Well, that's the update. Cheers!
Les _________________ "Let's make noise for peace." - Kijjaz |
|
Back to top
|
|
|
|