First post...

[shobley]

Hi,

I came across this site from Boomdog's site. I managed to put together his budget XY scanner and was amazed when I got it to work.

I've been tinkering around with a laser harp for the last couple of years. The only surviving footage is here:

http://www.youtube.com/watch?v=c2NcoWqzMLE

It's really simple - using an array of proportionally silvered mirrors to split the 100mw laser roughly into 8 parts.

I now feel I'm ready to move up to the next level and create a scanning harp. So I ordered a DT-40pro, DAC and software from Jian.

I've put together a MIDIBOX PIC to handle the TTL -> MIDI conversion using the laser blanking signal from the DAC. Eventually I want to replace the DAC with my own hardware - this is why I joined, I'm hoping that there's someone out there that can help me.

I'm hoping this is doable using a PIC or something since the pattern I want to create is very simple - a single axis fan of 8 beams.

Any advice at this early stage is greatly appreciated.

Steve

P.S. If you've never seen a scanning laser harp in action check this out:

http://www.youtube.com/watch?v=o-B_RvvdpRw

[buffo]

Hi Steve;

Welcome to PhotonLexicon. I can't view your laser harp videos at the moment (I'm at work), but I've seen a few of them in the past. It sounds like an interesting project.

I'm not the one to ask about programming a PIC to generate a signal for your scanner to create the fan pattern you mentioned. We've got other members here that should be able to help you though.

I'm curious as to why you feel you need to generate 8 static beams in the first place though. Wouldn't you get the same result if you just sent a sine wave to the galvo and let it scan a solid fan? I wouldn't think that the photo detectors (the "string sensors", if you will) would react any different to a continuously scanning beam vs one that paused at each detector location for a split second. All you're looking for is the detection of the beam being blocked by a finger, right? So you need - what, maybe 1/100th of a second response time? You can scan at an order of magnitude faster than that and design your "trigger" to only respond to events that last longer than 1/10oth of a second...

Or is it more of an artistic decision? (You want the strings to appear as single entities and not as a wide fan?)

Adam

[shobley]

Yes it's purely an artistic decision.

But...

The presence of a blanking signal makes it possible for the harp to work with only one sensor. As the beam moves from position 1..8 the sensor looks to see if there's any reflected light at that position and uses this information to work out if the string has been "plucked".

You no longer need the upper frame and the beams can go off into space...

[buffo]

Ah ha......

I hadn't thought about that. Clever! Ok - I see where you're headed.

Unfortunately, I can't help you with the PIC programming. But I'm sure someone here can.

Adam

[shobley]

Yes it is clever. I wish I could claim to have invented it

Here's a page that explains it very well:

http://www.laserharp.org/html/frameless.html

[stanwax]

Hi There Shobley
good to have you aboard for starters and even better to have a 'new' interest line for discussion on here - assuming you are the first laser harp builder to join.
Where in the world are you - im just being nosey really - and hoping that one day when we have helped you in your quest you might bring your instrument along to one of our US or UK laser meets. that would be a mega attraction!
Its a few years since I did any PIC stuff but to be honest it should be a real doddle to do with a PIC and i certainly wouldnt mind helping you out as this is such an interesting project.

PM me if you wanna talk serious

Rob

[shobley]

I'm originally from the UK (Leicester) - but I moved to the US in 1998. Been in Indianapolis, IN ever since.

I have a pic programmer and software (somewhere in the basement) I bought it a while ago along with a couple of sets of Midibox hardware (www.midibox.org). I've only just assembled them but the MIOS operating system makes scanning pins, and sending out MIDI signals really easy.

When creating the fan of beams is it better to use a back and forth continous sweep or a staircase wave, so that when the beam is 'unblanked' the mirror is stationary?

BTW I have a Lasever 532 module with TTL and analog blanking inputs.

Steve

[stanwax]

The scan need not be blanked when you are scanning except when the beam returns from the far end to the begining which we will assume it will do much more quickly than when it is ceating the beams. While on its forward track the laser may not need to be blanked as long as you pause long enough at each waypoint the quick jumps to those points will hardly register. It really depends on laser power, how clear or not the air is and the ratio of time of jumps to pauses.

Rob

[buffo]

A frameless laser harp will need to know where the scanner is at all times. That's how it figures out which "string" is being played... When it detects reflected light, it checks to see where the scanner is. I believe it would be better to have the galvo moving back and fourth constantly in this case. (This will give you the maximum possible response time.) Unfortunately, it also means your detection circuit needs to be very fast, and you must have very accurate position information on the galvo or you'll be off on the note. If you do it this way though, you don't need to blank the laser, and you can detect a string pluck no matter which direction the beam is scanning.

Adam

[stanwax]

you can extract a position signal from the staircase wave to tell you whch string is being plucked. Lets slow it down and say it pauses on each string for half a second then moves on, even just measuring how many half seconds + jump time have elapsed since the begining of the scan will tell you where the beam is that is being plucked

Rob