The Prism Laser Guitar is done... time for Mark II!

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

And another, from Coolaudio:

http://www.coolaudio.com/files/public/V3207_DATASHEET.pdf

This is a 1024 step BBD with a range of 2.56ms - 51.2ms.

Here is the full list of Coolaudio semiconductors: http://www.coolaudio.com/products.html

Of course there's a catch - a minimum order of USD $2000. So - who wants to be a distributor? *sigh*

Inventor · Posted: Thu Jul 23, 2009 5:00 pm Post subject:

no problem, smallbear has an equivalent Shanghai Belling part in quantity one for a few dollars each. same delay IIRC.

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

I can also get the MN3209 on eBay for $5 each. They're probably Chinese clones, but as long as they work it's fine with me.

Now, before I even consider continuing with this, do you know if I can dynamically change the delay time while it's processing a waveform? Can I push the sampling frequency a little higher and lower than specs?

ian-s · Posted: Thu Jul 23, 2009 7:03 pm Post subject:

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Hey Inventor, what formula are you using to calculate the output frequency range of the various BBDs? The formula listed here:

http://ccrma.stanford.edu/realsimple/faust_strings/Exciting_Digitar_String.html

Gets different numbers, though I'm not sure if I'm using the right value for N.

Inventor · Posted: Sat Jul 25, 2009 7:49 pm Post subject:

jeff-o, it is Fo = fs / N, where N is the number of stages divided by two.

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Inventor · Posted: Sat Jul 25, 2009 10:46 pm Post subject:

Yes, it's a bit confusing what you can really do with the part when the datasheet is so confuzzled.

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

I'm wondering now if the current crop of microcontrollers might be up to the task of implementing the KS algorithm. The programming involved is way above my skill level, but I know quite a few programmers who might be interested in helping out.

I may pursue a microcontroller solution in tandem with an all-analog solution...

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Hey Inventor,

Have you had a change to mess around with any K-S filter stuff? I haven't had much time due to family stuff (happy things!) so I've basically been surfing the net trying to find sources for MN3209 and MN3204 ICs. According to my calculations they have the right delay lengths for an ordinary guitar and an ordinary bass guitar, respectively. Since I may only be able to fit four "strings" on the next guitar, I have to consider making a dedicated bass.

I've also been looking into dedicated microcontrollers and DSPs that could do everything digitally, but I know that I don't have the programming chops to use one of them just yet. I really must brush up on my C skilz!

Inventor · Posted: Fri Jul 31, 2009 12:24 am Post subject:

Hi jeff-o,

I have been sick this week. I missed my radio show and slept all week - yucky. But I feel better now. I was able to set up my electronics bench and get the chips ready for the project. So that's at least a start. Once I get through my show, I'll have a look at it.

I would like to use this K-S circuit on one of my next projects, so I'm eager to give it a try. Right now I'm just thinking about it and how I'm going to test it. For example, I have a scope but it's kind of broken, and a meter. So how am I going to view the signals?

Maybe i'll build the LM386 amplifier with speaker first and *listen* to the signals...

Les

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Yeah, it seems very much like a "try it and hear if it sounds good" sort of thing. The scope will come in handy later for fine-tuning and debugging.

Inventor · Posted: Fri Jul 31, 2009 2:53 am Post subject:

You know, I've given this some thought, and I feel like I can implement the KS algorithm one piece at a time using ChucK. I can build code that tests the circuit using the line-in jack to probe two channels and the headphones out jack to drive two channels.

This is all very much in accord with my eChucK project, so I think I'll do it that way. In fact, I can just hook up a delay line and test it and go from there. I'm glad i've chosen to take some time and think about it and let things sink in a little bit.

I'll keep you posted on progress.

Les

yerpa58 · Posted: Fri Jul 31, 2009 11:30 am Post subject:

Using rare delay line ICs for Karplus-Strong seems like doing it the hard way. Any cheap microcontroller with some internal RAM can do the job all by itself. Here is an example, although this algorithm is hard to tune. I think your VCO method may be more useful in the long run.

http://www.reprolabs.com/fs7.htm

Inventor · Posted: Fri Jul 31, 2009 11:57 am Post subject:

The trouble is, Any cheap microcontroller requires a programmer for which I don't have any budget. It's a good solution for folks who want to invest in a programmer though. When I looked at dsPIC chips, I found that they were not available in low pin-count with a DAC.

There are lots of advantages to a microcontroller approach but I want to try out an analog approach first for my particular application.

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Also, I've found an inexpensive source for MN-series chips. They ship from Singapore but seem legit so far. If I don't get burned by someone on the other side of the Earth then I'll be sure to recommend them. Wink

Inventor · Posted: Wed Aug 19, 2009 5:12 pm Post subject:

jeffo, for an update on the hardware of the project, see this thread:

http://www.electro-music.com/forum/topic-36165.html

I don't have it working yet but it's only been one day, so give me time.

Les

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Inventor · Posted: Wed Aug 19, 2009 5:40 pm Post subject:

kewl man, keep us posted! Smile

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Well, "sometime soon" hasn't come yet.

Inventor, have you had any luck?

Inventor · Posted: Mon Nov 02, 2009 12:19 am Post subject:

hey jeff-o, i posted to my thread in eChucK forum, so go there for some news. I just got done prototyping the oscillator which works at low speed. Need to test it at high speed, but guess what?

I met Rob.

No shit, Howard's friend Rob and he's a true wizard at analog electroonics. Karplus-Strong ain't nothing new to him and he knew all about it. He told me some secret details that rock. I will be including them on the next version of the circuit. The man is so incredible with his knowledge and intelligence that I had him sign the back of my paper check that he drew a schematic on.

So kick ass man! you just wait, there will be a ks board for you to use in the future. Best to ya man, Love and all that stuff. Be good.

Les

jeff-o · Joined: Jul 19, 2009 Posts: 97 Location: Ontario, Canada

Whoa, cool! I sent a few emails to Rob I think, but he didn't draw a schematic for me! Well if you've got something in the works then I'll just have to sit tight and see what you've come up with. I don't see myself having time to work on my next guitar until the new year now -- I've committed myself to building a bunch of Christmas presents that will eat up the remaining scraps of free time.

Heading over to eChuck now...