Help - Search - Members - Calendar
Full Version: Ultrasonic Focusing
Lumenlab > Audio Video Sciences > Advanced Projector Builder > DIY Electronics Projects
seesoe
i have always been looking into ways to make my PJ work cooler by doing electroinc mods/addon's to it
had a very nice idea to mount a ultrasonic detector on the front of PJ. and that will auto focus the lenses.
the lens's would be hooked to axies and with servo's a controler board would read the data from the ultrasoinc detector and adjust the servo''s, useing a simple program that will follow the rules, for when ex. the PJ is 5" away from wall it will know how to adjust everything.

i my self might not be able to do this, im still learning with pcb and electronic design

but i think this idea is real great, and i thought i might share it, with some thought it would be very handy


does anyone think they could over take something like this?

btw i got my idea when i was browseing through this http://www.hobbyengineering.com/H1145.html
flashedarling
That's pretty cool. How do commercial projectors autofocus typically? I'd be willing to mess around with that although what it makes me want to do is make myself a little robot projector that rolled around and adjusted itself manually. A R2D2 projector.... Hmmmmm...
pagercam
QUOTE (flashedarling @ Oct 31 2006, 06:50 PM) *
That's pretty cool. How do commercial projectors autofocus typically? I'd be willing to mess around with that although what it makes me want to do is make myself a little robot projector that rolled around and adjusted itself manually. A R2D2 projector.... Hmmmmm...

I don't think that any projector auto focus. Cameras on the other hand do auto focus and I beleive that they use a maximum contrast = best focus type algorithm, blury edges reduce contrast so maximizing contrast will find best focus.
seesoe
o i c
but still there must be a neat way to do it for PJ's?
pagercam
QUOTE (seesoe @ Oct 31 2006, 07:38 PM) *
o i c
but still there must be a neat way to do it for PJ's?


I can't see a way to do it without some sort of feedback, so you would need atleast a simple camera. If you use the ultra sonic ranging and then manually focus you could buld a table of focus position vs distance that should work fine, focus can generally tollerate a fair amont of slop.
seesoe
there could be a type of system where u input how far away from the screen the have the PJ and it will know this far means it needs all the lens X spaced apart

this could be another way instead of using the ultrasonic detector to find how far it really is
kunteper
QUOTE (seesoe @ Nov 1 2006, 05:45 AM) *
there could be a type of system where u input how far away from the screen the have the PJ and it will know this far means it needs all the lens X spaced apart

this could be another way instead of using the ultrasonic detector to find how far it really is


you can find a lot of different range finder circuits on the net mainly used for robotic projects.

It is a pretty good idea actually ... I have couple of robotics books with such circuits, I'll take a look and see if there is something that be of use ...
srinity
As someone who as worked with ultrasonics on robots in the past, you need to be very careful about using them (for this application). Their field of view is very broad. So, if you had something closer to the ultrasonic then the wall, for instance a A/V tower or CD tower or something, then the ultrasonic sensor would return the distance to that, not the wall. For robotics, this is great, as you want to find out if something is close, no matter where it is.

However, for a projector, you will need to "focus" the ultrasonic somehow, build a sound tube that only allows the ultrasonic to "ping" along one tightly colimated line.
kunteper
you might wanna check this out. A simple ultrasonic range finder from 25cm to 8metres .... pretty neat
Timm-McCoy
while this is entirely possible...
you're talking about using essentially a microcontroller to make this project as easy as possible.
so the actual focusing will all be in the software.
it's completely possible tho.
i'm thinking about designing a circuit for DIY PJ's that will not only keep the LCD turned off until the lamp fully warms up, but will also restrict you from trying to turn the lamp back on after you turn it off, until after the restrike time. it will of course also handle cooling, using temp. sensor(s) and PWM of a few 120mm case fans.

so yeah once i get like a crap load more time (already busy with other elec. projects for school and stuff). i'll have to throw this baby together.

if it really does work well, i might just have to make and sell this cool controller thing.
what kinds of features would you guys like to see?
EdZ
A lot of non-digital 'point and shoot' cameras use ultrasonic sensors for rangefinding and focusing (as do some digital ones). As Sirinity said, their field of 'view' (cone of sound?) is wide, which is why your pictures sometimes come out with odd focuses.

You could try hacking one of those el-cheapo 'laser' rangefinders )they're not actually measuring the distance with a laser, they use ultrasound. But as they're already focusing it and already have some control electronics in place it might save a bit of work).

This assumes you can find a decent variable-focus triplet.
adaminc
I know a lot of polaroid camera's use ultrasonic range finders for focusing
Metzelmaennchen
A friend of mine uses a NEC PJ, this one can auto adjust the focus!!!
I think it was the MT1075 ??? But I'm not quite sure...
seesoe
QUOTE (EdZ @ Jan 4 2007, 08:11 PM) *
A lot of non-digital 'point and shoot' cameras use ultrasonic sensors for rangefinding and focusing (as do some digital ones). As Sirinity said, their field of 'view' (cone of sound?) is wide, which is why your pictures sometimes come out with odd focuses.

You could try hacking one of those el-cheapo 'laser' rangefinders )they're not actually measuring the distance with a laser, they use ultrasound. But as they're already focusing it and already have some control electronics in place it might save a bit of work).

This assumes you can find a decent variable-focus triplet.

you know with the laser that might be faster and easier i saw i page about a simple laser range finder, it sends the laser out then wait for it to bounce back and it calculates how long it took to come back then it knows how far it is. i seen a digital version of this on hackaday i think it was i forget
NinHowFritz
Lasers would work, and they do sell ultrasonic transducers with different 'viewing angles'. You would want the narrowist.


With the ultrasonic, you could get a more accurate reading because sound is quite a bit slower than light.

However you do it, PIC microcontrollers seem popular for simple things like this that would require a computer.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.