lissajous Pattern laser light and sound project

[large_ghostman]

For a bit of background see this thread https://www.electro-tech-online.com/threads/thoughts-on-new-project-idea.147085/

In a nutshell I wasnt going to do my Art exam or project, then had a brainwave on how to make it less boring and throw a load of electronics into it! It also fits my remit for the next relaxation project (yes I solder just for fun).

So I decided to go with Lissajous laser light and sound show, while it has been done to death I wanted to add a couple of twists. So first off for those landing from planet google, this is not a intentional tutorial but more of a project progress thread.

First off the easy bit...........

SIMPLE PATTERNS

The thinking so far (and it could change) is too have 3 lasers (different colours), these will point into mirrors probably made from old CD's. that I mount on old PC fans or similar. The fans are going to be mounted on servos for both the X and Y axis, this is simply to alter the angle of the mirrors and dosnt have to fast. I could use old CD case stepper motors with the screw thread for this but not sure yet.

The sound will be stereo and fed into band pass filters, the filters will be user adjustable via pots. The speed of the fans and to some extent the position of the fans will be controlled via the signal from the filters (using ADC from micro). The idea for this bit is not to be completely in sync but to be tethered a bit to the music but allow the user to adjust some the filters and therefore pattern. So we should end up with something a little bit like this, but controlled mainly by music.

The next bit is harder.

Taking it a step further the mirrors will be stationary and the pattern controlled by vibrating the mirrors, again the other thread has some chat on how this might be done but at the moment I am flexible. What I am aiming for is patterns done to music like the second video (below), but instead of an OScope I will use the laser. It Sounds complicated but I dont think it will be, I have messed with Matlab and the patterns being mathematical or not too bad to do. The next video shows the effect I am after.

This one shows some other things that can be done, having a play for a short while with software it dosnt seem to hard to do. I am not worrying about how to move the laser's yet as this can be done a million ways.

I wasnt going to bother with the Art exam but having read the brief for it I realized I could make it cool, and have it fit what I wanted to do for the next project (just for fun). I have two weeks to play and decide if I am going to do this exam (its a resit, as i didnt show for the first exam), if I go ahead then I will have another 5-6 weeks (ish) to do the project. If I get as far as the patterns on the OSCOPE I will go with that, but the aim is to do this with lasers..

QUESTIONS I AM UNSURE OF.

Some of these will seem silly, but some of them have never cropped up for me before. I want to try some the patterns on the scope, I have 3 function generators and will be using the scope in X,Y mode. All my FG's are either single channel or a mix of high frequency/low frequency outputs. I dont have one that will out put two different signals at the same time (one driving X One driving Y).

1) Can I safely tie the grounds from my FG's together and feed one signal into X channel and the other into Y Channel?

3) If I do tie my grounds do I then also tie it to the scope ground? I am concerned with ground loops and have never tried two FG's at the same time

Or do I simple tie FG1 to X channel including the ground from FG1 to Ground on scope X channel? And do the same with FG2 to the Y channel?

Or Leave all the grounds untied?

At the moment I am not worrying about the other aspects, I just need a test set up first to try the patterns out as this will dictate what path I go down.
I will post each stage of the experiment and build in this thread

 

[JimB]

1) Can I safely tie the grounds from my FG's together and feed one signal into X channel and the other into Y Channel?

3) If I do tie my grounds do I then also tie it to the scope ground? I am concerned with ground loops and have never tried two FG's at the same time

Or do I simple tie FG1 to X channel including the ground from FG1 to Ground on scope X channel? And do the same with FG2 to the Y channel?

Or Leave all the grounds untied?

You seem to be all messed up in your head there LG!

Excuse me while I waffle...

A signal needs two "wires" to make a circuit.
The inputs to the scope are (usually) BNC connectors. The centre pin of the connector is one wire, the body of the connector is the other wire.
The bodies of the BNC connectors are (usually) connected to ground, the chassis of the oscilloscope.

Same with a function generator.

Just use BNC to BNC leads and connect one function generator to the scope Y input, and the other function generator to the X input.

Does this clear your thinking?

JimB

 

[large_ghostman]

You seem to be all messed up in your head there LG!

Excuse me while I waffle...

A signal needs two "wires" to make a circuit.
The inputs to the scope are (usually) BNC connectors. The centre pin of the connector is one wire, the body of the connector is the other wire.
The bodies of the BNC connectors are (usually) connected to ground, the chassis of the oscilloscope.

Same with a function generator.

Just use BNC to BNC leads and connect one function generator to the scope Y input, and the other function generator to the X input.

Does this clear your thinking?

JimB

Answers it perfectly Thanks, Incase your wondering... Your answer matches my last option lol, only I didnt put it well . So you have also answered what I was hinting at, when I add generators then the grounds are meant to tied. What messes things up is a nasty experience with 2 ATX PSU's in parallel, so now I am paranoid .

I was asking because in my head I have jumped forward, I will be intercepting the signal before it goes to the scope. But yes I was mixing it up and it came out all wrong , handy you speak fluent LG lol.

I have the individual bits pretty much sorted, the hard part will be getting it all stitched together and working. I will need to build a PSU for this project, I have a old PSU board from a HP logic analyzer, its the 350XXC. the PSU is toast but the main toroid transformer on it is fine. Its a beast of a transformer and I will get the spec later, but seeing as the main PSU board is huge and the instrument it came from power hungry with alot of different voltages, I figure it should make a good start point for a PSU with negative rail for the op amps I will need.

I have the signal side sorted, I will use something like FL Studio for producing the sound files. It has many options and is basically a software channel mixer package (and more), so making the signals will be straight forward. Not sure what I will use to play the file, maybe a laptop or even The R PI (but I am sure thats mono!! so will check).
I might even use one of those £3 crappy MP3 players as I only need to output a stereo signal and then feed this into the main board that controls the laser and filters. So later I will start posting the scrap etc I have to pinch parts from, for the complex pics I might try electro magnetic solenoids and some kind of home built laser mounting.

Main bits I need to do after I have tested a few things out are (not a complete list).

1) PSU.........Probably toroid transformer based

2) Laser driver circuit

3) Filters

4) Laser mounting and movement

5) Waveform's

Loads of other bits but will post a longer list later, this is just a list for me to start thinking about what parts I have and what I can use.

 

[large_ghostman]

I might try these for the laser movement for the more complex patterns **broken link removed**.
They are in the UK and cheap enough to try, if they dont work they useful enough for the spares bins , I have a feeling I might need to up the spring strength so it returns faster

 

[large_ghostman]

So I have been looking into this and there seems to be alot of information on laser forums about laser projectors. But they approach it from a different angle, they have a pattern in mind and want to project it. I might the pattern controlled by the sound but still look like a pattern, they use mainly USB sound cards and DAC corrector etc etc. To me this seems a bit complicated! The videos above show you can make all kinds of distinct shapes via sound, but the sound isnt what you want to hear most the time!

So looking at it I still think the way you do it on a O Scope is the way to go, I have a few ideas I want to try out before I explain what I have in mind. I need to use proteus and 55 timers first (I can never get micros to work in proteus).

But I think I can do this using the OScope technique in a different way, just in case though, my back up plan is the simple mirror and speed of rotation trick as a fall back for the project.

 

[JimB]

I might try these for the laser movement for the more complex patterns **broken link removed**.
They are in the UK and cheap enough to try, if they dont work they useful enough for the spares bins , I have a feeling I might need to up the spring strength so it returns faster

A word of caution...
The stronger the spring, the more force it will need to move it.
The faster you want to move it, the more force it will need to move it.
The more force you need, the bigger the current in the coils and the bigger the driver transistors.
etc.

An idea...
I think there are two parts to the project:
1/ Generating two electrical signals from the "music".
2/ Using the electrical signals to make the display.

For your initial testing of part1, use an oscilloscope.
Then if you run out of time/money/?? you have a working system.
Make the outputs of the "signal generator" a voltage which sits at 0v when there is no signal, and say +1v to move the spot to the right or up the screen, and -1v to move the spot to the left of the screen or down the screen.

Then, you have a known interface signal for input to the thing which moves the laser.
And if you have run out of time/money/?? you can say that you have a well defined interface for future expansion. (Got to be worth a few marks in the exam?).

JimB

 

[audioguru]

Have you ever connected the left channel of stereo audio to the vertical and the right channel to the horizontal of a 'scope to view a moving tangle of thread?

 

[large_ghostman]

Have you ever connected the left channel of stereo audio to the vertical and the right channel to the horizontal of a 'scope to view a moving tangle of thread?

Yes last night, but there is software that can output a cleaner signal, also you get to control what signal goes to which side etc. As video 2 and 3 shows it can be done with a scope, the 2nd and 3rd vids are using sounds specifically to make the desired picture. On the normal stereo head phone I do get a signal that isnt very clean, but using the spdf channels on the sound card I get a much cleaner signal.

The other video is done with altering the speed of rotation for the mirrors, the idea I had was to link the speed with changes in the sound signal. The second option dosnt require a specific picture, so it dosnt matter what is produced picture wise it will be a pattern that changes with the music, the idea of filters is just so you can control when the speed is change etc.

Some valid points Jim, I am trying something a bit different at the moment and so far so good. The other fall back option is trying on a PC OScope, if that works then I have the option of using the school projector and feed the laptop pic of the scope to that.

Its all trial and error at the moment, but I am trying to keep in mind how its done on a scope and trying to translate that to a laser. I have tried a self wound voice coil and that was ok but I ran out of wire, so I have ordered more and will see what I can do. The other project I have completed that keeps bugging me is the electro magnetic levitation I did. It used pushed pull on a H bridge and 2 hall effect sensors (linear not on/off), if I can find a way to move a bar or something this way then I would have quick control. I have 2 weeks to play before I have to decide what to go with, even the simple spinning mirrors tied to sound altering the mirror RPM will get me good marks. Considering I wasnt going to sit the exam its all good.

Thanks for the comments it gives me something to chew on.

 

[audioguru]

High frequencies cause a complicated image, using only low frequencies makes a cleaner image. Have you tried gluing a small mirror on the cone of a speaker then the vibrating mirror deflects the laser beam?

 

[large_ghostman]

High frequencies cause a complicated image, using only low frequencies makes a cleaner image. Have you tried gluing a small mirror on the cone of a speaker then the vibrating mirror deflects the laser beam?

Not yet but I will do , did you watch the videos?

The only problem I see with a mirror and speaker is you have more than a single axis of movement unlike you do on a scope. But I would need to try it or its just speculation on my part.

 

[4pyros]

Wow LG you are going deep.
I also have been looking into lasers as of recently.
You can put resisters across silenoides to increase respostime.
But you need servos to drive mirrors to alter the beam angel to make patterns.
I am sure there is a way to make a lazer act like a scope.
I think you need to think more about defection of a mirror controlled from a linear voltage.
Did I make sense?
But yes you will still need two axes of movement for the mirrors.

 

[large_ghostman]

Wow LG you are going deep.
I also have been looking into lasers as of recently.
You can put resisters across silenoides to increase respostime.
But you need servos to drive mirrors to alter the beam angel to make patterns.
I am sure there is a way to make a lazer act like a scope.
I think you need to think more about defection of a mirror controlled from a linear voltage.
Did I make sense?
But yes you will still need two axes of movement for the mirrors.

Yes you made perfect sense, I know mirrors is how its currently done but I cant help but think there is a better way. The video with the mushroom keeps nagging at me, I am positive there is something I am missing that is the key. I keep thinking of that video and how an O Scope works, its a subtle difference to how its currently done with mirrors but I think maybe the key might be the third axis or rather the fact that a scope beam moves at a linear rate left to right, then there is vertical deflection and horizontal deflection (not explained well).

So whats nagging at me is a motor thats fixed and switches fast left to right (would be hard but not impossible to blank on the return to left), then move the actual laser stage mounting X,Y with a coil (speaker cone whatever)

 

[large_ghostman]

Game over at the moment as the crap laser I had has died, I have a better one but not sure where it is. I will order another laser I found some cheap ones in UK.

I know it seems deep but my last project with the hexapod taught me alot about myself. It was a really complex project, far more so than it looks at first. Normally what holds me back is me thinking too much, then I get self doubt.

A good example was the question on function generators, I focused on the wrong thing and in the end jumbled it all in my head. Last time I found if I take it in micro chunks and let myself work in the way I do instead of trying to control how I work, I get more done.

I know I work in a fairly strange way, but if I try and work in a more normal way I get lost and doubt creeps in. I am positive the laser thing can be done and done differently to the way its done now, I also know I am not seeing the key yet. But I am not working the way i should work, I am working in a more controlled way than my mind works normally. So I will sleep on it tonight and try and tackle it the same way I did the last project, I do appreciate all the comments and suggestions, it helps me pick my way through stuff

 

[alec_t]

If you want to complicate things you could perhaps use three orthogonal mirrors each driven by a speaker cone to deflect a laser beam in three dimensions in a volume of mist (created by an ultrasonic transducer), to project 'holograms'.

 

[large_ghostman]

If you want to complicate things you could perhaps use three orthogonal mirrors each driven by a speaker cone to deflect a laser beam in three dimensions in a volume of mist (created by an ultrasonic transducer), to project 'holograms'.

Thats already done and while cool is getting away from the brief. Its actually an art project, the idea is to interpret a different media into art. So I chose sound to light, the idea isnt to get say a square just drawn as thats already done, if you watch the mushroom thats is what I am after but in laser rather than a scope screen. The shapes needs to be drawn by the sound, the mushroom etc is drawn by overlaying different frequencies using the sine and cosine of the wave and pulsing in square waves on top. Because the shape has to be drawn by the sound is why I have the spinning mirror back up plan, while real shapes would be cool they are not needed.

Mirrors have been tried to do this effect, if you read all the stuff the guy who did the mushroom video has done he says he has tried laser's and they dont work well. But he used the current mirror method like in laser projectors, I think that is why he had artifacts and such in the pic produced, the sound files work on the scope but translating to laser isnt as straight forward as I thought. The only thing I havnt explored is the constant left to right right that dosnt change on a scope. I am becoming convinced that mirrors make things worse, although its counter intuitive I think the mirrors are what is messing the picture up.

In a way my idea is more simple than mirrors, its a mechanical O Scope of sorts that moves the light beam rather use mirrors which is like a Scope have the beam fixed and the screen moved. I got close while testing but I am pretty sure whats missing is the time base affect you have with a scope. Moving the laser by vibration using transducer is what I mention in post one. The movement is good but too small. I have another laser coming so have a couple of days to see if I can sort a better mounting out. The worse bit for me is the sound mixing, many of the frequencies are fairly high and hurt my ears!!

 

[Les Jones]

Hi LG,
Here are some pictures of the idea I suggested the other day of using an old disk drive head actuator to move the mirror.You would need two mounted at right angles to each other. The mirror on the second one would need to be larger than the first as the beam would be moving along it's axis of rotation. Mounting them as close together as possible would reduce the size of the mirror needed on the second actuator. The mirror on this one is stuck onto a flat surface that I milled on the actuator casting with super glue. The piece of mirror is from an old scanner. I think part of a platter removed from the drive would also work well as a mirror. This morning I Googled "disk head driver mirror laser" to see if any one else has had the same idea and found quite a number of hits. This is one of them https://www.spikenzielabs.com/SpikenzieLabs/Laser_Display.html



**broken link removed**

**broken link removed**

Les.

 

[large_ghostman]

Hi Les, I tried this out already. The mirror needs to be a proper laser one or the artifacts are really bad, even then it dosnt look right. But My next test will use two hard drives, just not like that . I have to get a new laser before I test it but I think the answer is staring me in the face (literally at the moment), I will try and get the mounting done over the next couple of days and you can see what you think.

Those real laser mirrors are pretty strange, unless you get really close they look normal. But if you put them upto your eye you can see straight through them although its kind of smokey. I cant remember the proper name of them and its really bugging me lol. The problem with normal mirrors is it seems to spread the beam at certain angles, I might use HDD for the spinning mirror section of the project.

 

[large_ghostman]

I am wrong!!

You do only need X and Y, I keep forgetting the scope is in XY mode. so the problem has to be mirrors, but at least it simplifies things. Currently they use galvos to change the mirrors, so basically the same as what we have said. I think I might have a way to move the laser itself rather than a mirror. On the kickstarter website they guy did he explains about the laser trials failing. But he dosnt seem to know why

 

[Les Jones]

I think the beam spread is due to a normal mirror having the silvering on the back of the glass. You will get two reflections. One from the surface of the glass and the other from the silver on the back of the glass. The piece of mirror that came from a scanner could have been used both ways as there was no paint over the silvering as on normal mirrors. I expected the silvering to be soft and scratch easily if you try to clean it. It did mark easily so it must be some harder metal than silver vacume deposited. Using a disk platter as a mirror may solve this problem. When I was looking to see if anyone else had used this idea I found out that some disk platters are ceramic. All the ones that I have dismantled are hard aluminium alloy with a magnetic coating.

Les.

 

[large_ghostman]

The only problem with disk drive plates as mirrors is the hub in the center, some patterns you have to have the beam central and there is the hub in the way, same problem with CD's.
I am stuffed at the moment until

A) I find my other laser (need a serious bench tidy)

B) the new one gets here

I also have my MKII project that has hit a hump, so that is taking a little time away at the moment. For some reason the mems boards isnt working, so I need to get that sorted first as its the more important project