Laser Scanning Display Project

[Georg]

Dear ElectroNick,
can you read it!!

Just the topline is little bit too late compared to the next line. Surprisingly the speed was not so critical! The movie was made with 34.840ms rotation time.
I had to mirror the letters, probably this depends on the turnning direction of the motor...(Now by posting came into my mind: would have been easier to change the wires  :))

[ElectroNick]

Woo-hoo!!! It (kinda) works!!!

Now I get a renewed appreciation of how important a good quality mirror is for this project. Before that I could not make heads to tails and wasn't even sure if the lines are transposed or some are simply missing or what. There seems to be a line missing in the middle, so there may be some issue with timing, we'll have to address that. Also interesting is the top line's shift. Looks like something holds up only the first pass of the program, and not the other ones. Have to  look closer at the code for what might be causing this.

So, what's the inclination difference between the mirrors, 0.25° or less?

How are you driving the laser diode, by the way?

Thanks so much for moving the project alone, I have a feeling I might have lost interest by now if not for your help! I'll try to put some more time into it, just need to finish up another little project this weekend.

Cheers!

[Georg]

It is nice when you see some laser text ...
I made some changes in the code.
- caused by the interupt function, the first row is later. Therefore I postponed all the following rows by 3xTimepixel
- the line "if(currentRow > 7) { currentRow = 0; }" results in a short value for the mirrordelay. Therefore the next row has added to the old row (you can see 8 instead of 4 character rows)

I have added 2 pics, made shortly after each other.
One is better than the normal impression! In the other one I marked my headache issues (usually it looks like that).
Current values: rotationtime 54ms +-0.2ms Pixeltime 26

For test I have increased the characters to 5 "T"s.
I still have 2 headache issues:
- in row 5 the dots are left from the others, when I change the image it comes closer.. No idea why?! rowoffset=7, actualrow=5 -> currentrow=4? Nothing special in this row?? Why should the row appear earlier?
- the later characters have black holes in the letter. I have no real idea for a reason or what to test?! It would mean that the laser is not switched for this characters...

To your questions:
Mirrors: I am sorry but I cannot find the drawing for my mirrorholder! I think it was .25degree between the mirrors. I search in my room, maybe it appears somewhere... maybe I did send you the drawing already?
Laser: I bought a pencil laser and removed the housing...

[ElectroNick]

Hi Georg,
I'm on a road today, can only look at it tomorrow. But in the meantime, can you post your modified code here? I'd like to take a look. The "black holes" in some characters - are you using serial communication via USB? See if you can either increase the rate or simply disable the serial - I've found it too slow and had to disable it. Basically, looks like some parts of the code are either too slow or run too many loops or something like that.

One row not in line with others - I had a lot of that with my bad mirrors. However, the fact that this row is in the middle, can also mean something software-related. Have to see the software again - haven't looked at in in months.

[Georg]

Hi ElectroNick,
the code is already attached to the last post (graupner_2.zip).
I can check end of the week for the serial stuff... it slows it down. The ESC starts not by itself, I have to play with the start values until it works. Therefore the serial input is easy.
Maybe I move it into the setup - and free the loop from all serial code.
Maybe you find something to improve in the code ...
Bye

[Georg]

Hi ElectroNick,

It flickers - but in reality not as much as in the clip. Maybe the motorspeed is not stable enough - or each rotation is not triggered exactly enough by the reflection light sensor...
The distance between two lines in the middle of the characters is too big - that is a mistake in the angles of the mirrors! In the former clips some points were missing: finally I find out, that this was a problem of the polish of the mirrors... (all mechanical !!)
The code is attached!
As next a will try a strong green laser. Hopefully that one is fast enough! With the green laser I want to project the text to the wall of a house - will be a huge character size!
Thanks again for your code! It makes it possible ...

[ElectroNick]

Wow, this is so cool! This is a great improvement over what we've had so far!

I don't actually know if you can make it stay absolutely still but it appears that there may be an unnecessary code executing once per 8 turns  (because all of the characters are moving at the same time - just my conjecture at this point). Else it may be skipping an entire 8-turns cycle.

Sorry I was super busy lately and could not spend any time with the project, I am so glad you are making progress on your own!

   

[Georg]

Hi ElectroNick,
You are right maybe the flickering comes from the software. I do not understand what you mean with the unnecessary code once per 8 turns? Can you explain this a little bit more? (do you think of my changed code or the original one?)

I started now the green laser. But the green laser of my laser pointer is too slow...Usually the green laser is very,very bright, but turning it on/off very fast it gets dark, partly does not switch on at all.
I think with lasers you have a lot of experience. Should I just buy another laser pointer and hoping that one is faster? what do you propose to do?
Bye

[ElectroNick]

I think it's not the laser that's slow, it's the driver that controls it. I had the same experience trying to use my own TTL-controlled laser driver: it just could not switch fast enough using its TTL input (which was originally designed for ON-OFF function, not fast modulation with periods measures in microseconds). I ended up driving a laser diode like a simple LED - no driver at all, just a current limiting resistor. I can't recommend it for long term use and this is just another hurdle to overcome before the project is finished. But driving the laser diode directly from a voltage source (not constant current driver) appears to be the fastest way to switch it.

As far as the actual laser diode, those in laser pointers are usually 5mW. You can find a rather inexpensive 250mW red laser diode these days that were designed for DVD-RW drives. It's 50 times more powerful, so it should be much brighter than a laser pointer. Green laser diodes are just a bit too expensive at this point, I would avoid them until the rest of the project's parts are worked out. But green lasers do have one important advantage: human eye is more sensitive to green, so we'll have the same brightness with less powerful laser. Problem is: it's cheaper to buy a couple of 250mW red than one 50mW green, so I would stick with more powerful red for now.

[Georg]

I had just luck with my laser pointer driver, that it is fast enough! (sometime luck is helpful
Thanks for your hint:
I found a laser diode with a lens from a dvd/cd drive. On the diode is written SI362EK6C (no search result in google).
I have no idea how to connect it?! It has 3 legs.
What do you suggest as resistor?
And what voltage should I set on my powersupply ? 5V?
I would drive it over a NPN-transistor to supply enough current.ok?

[ElectroNick]

Hi Georg,

If this is not from a DVD-RW writer , it may actually be an infrared, not a red laser diode. Both red and IR are 2.4V-2.5V devices, so you would need at the very least a current limiting resistor to drop your voltage from 5V to 2.4V  . Either that or a proper constant current driver, of course. Like I said before, I tried to use my own: http://elabz.com/ttl-controlled-laser-diode-driver/ but it would not switch fast enough, so I simply went for a current limiting resistor. While I'm powering this project from a powerful and stable bench 5V supply, it should work. If the incoming voltage fluctuates, you will most definitely need a constant current driver. Without knowing the power of the laser diode, it's hard to say what value the current limiting resistor should be. These usually start lasing at 25-30mA, so you might want to drop the extra 2.6V by using a 2.6/0.030 ~= 100Ω resistor and then go down in Ω if this is a more powerful laser diode.
 

[Georg]

Hi ElectroNick,
now it is clear I will order one ...
PS; do I need a lense for the laser diode?

[ElectroNick]

Yes, you do need lens and you do need a housing. Good thing they are usually sold together
Check out this post: http://elabz.com/laser-cutting-diy/ for some info on laser diode housings . Basically, look up AixiZ or O-like (brands) on eBay and plenty of choices in $10 range should come up.
 

[Georg]

I am waiting for the new laser ... (in reality it looks more stable)

[ElectroNick]

Wow, Georg, I am very touched! 

Is the message hard-coded in or is it coming over the USB from PC?

Cheers!