Thermal Imaging Camera - possible mating with a film camera?

[nick mulder]

Howdo,

Take a quick look at this

There is a full manual Dead Link Removed also..

Its a proper Thermal Imaging unit - not just night vision/predator cam - and I've found a 'cheap' one (compared to what it originally sold for that is) and I'm particularly interested if I could mate this up with a 16mm, 35mm or video camera ... (I'm talking cine and not stills but the mod would be the same in either case) - Looking at page 10 of the manual I cant see any ground glass where the image would be focused, would I instead simply put a short focal length lens on a camera and poke it up near the finder ? (sort of approximating the camera to a viewers eye) ...

I'm reasonably capable of machining it if need be.

Anybody with experience in this sort of thing - optics/video-taps/mods/odd-ball projects appreciated

nick

 

[glbeas]

Since it has an eyepiece theres probably optics present to focus the virtual image for the eye, so it should work in principle to eyepiece projection used with astrophotography. A few simple experiments should reveal the best point to position the cameras film plane.

 

[nick mulder]

hmm,

I'm afraid the correct film plane will be inside the camera itself ... Is this a possibility ?

What if it is too big or too small for the format ? How could I focus it for the correct magnification ?

I know nobody can give specifics on this camera - but are there any general rules ?

 

[glbeas]

I don't think so. The eyepiece should project a virtual image at about the point of your eyes lens, so you should be able to use a camera mounted somewhere outside the eyepiece. Easy way to tell is to hold a 35mm up to the eyepiece and see if you can catch an image. Theres two ways to do the eyepiece projection, one without the lens using just the eyepice to focus the image on the film. A sheet of paper held in front of the eyepiece should be able to catch an image if this will work. The other way is with a lens on the body with the focal point of the eyepiece set about the front of the lens. Here you can confirm the action by looking through the camera. Play around with it, or wth a pair of binoculars if you don't have the unit yet, to see how this works.

 

[nick mulder]

righto ! I'll give it a go with a night vision scope I have here ... (should have thought of that earlier) thanks for the tips

 

[Dan Henderson]

we make good use of thermal imaging cameras in the fire service. Some of the latest models have outputs so that the images can be viewed and/or recorded outside of the fire area.

I've been around these things for 10 years now, and they still fascinate me.

 

[Tom Hoskinson]

We use Thermal Imaging Cameras for non-destructive evaluation of electronic equipment. The cameras we use are designed to record the results on both film and videotape.

 

[Helen B]

I guess that you already know about the frame rate mismatch, but just in case:

That Probeye doesn't have a standard video out for a good reason: it is a clever mechanical scanner (fully worthy of the John Logie Baird 1928 Videodisk Award*) with only six receptors and six LEDs that does not operate at the standard frame rate or field interlace of NTSC or PAL. The image is formed by only six LEDs.

If you were using a video or movie camera you could have some phasing problems if it isn't continuously variable speed (film) or shutter (video). Because it is mechanical and the source scanning and interlacing are physically linked the scan rate does not have to be fixed - it can vary - I don't think that it is crystal controlled because it doesn't have to be. The rotational speed of the mirror should be stable enough, however, to be able to sync up a variable speed or shutter motion picture camera and thus minimise rolling lines.

I hope that makes sense.

Best,
Helen
* Baird's 1928 video recorder used a mechanical link between the scanning mechanism and the disk recorder rotational drive to ensure that the two were perfectly locked in sync.

 

[AgX]

phasing

Helen,

The longer I look at the sketch of that optical system the more confused I get. If both the Probeeye and the cine-camera were at the same frame rate, must they even be in phase?
What image would one get if the camera gets constantly (exactly same frame rates) images from two mirrors at a time?

 

[Helen B]

The idea would be to capture one full scan per frame of film - ie one complete rotation of all ten mirrors on one frame. You would have to have a 180° shutter, and the frame rate would have to be such that exactly one full Probeye frame was captured while the shutter was open. There would be one Probeye frame lost while the cine camera shutter was closed. The Probeye frame rate is about 20 fps, so the cine camera would have to run at exactly half that - about 10 fps.

It would not be all that important where the cine camera shutter opened during the mirror rotation, as long as it was matched on every frame - hence having to have an adjustable speed cine camera. I didn't mean to give the impression in my previous post that both the speed and phase would have to be matched - only the speed.

You could also get a stable picture if you synchronised at 20 fps, but with a 180° shutter you would only record half the scan lines - the same half on every frame. You would be recording the lines from five of the mirrors (ie 30 lines) and the 30 lines from the other five mirrors would be lost while the shutter was closed.

Best,
Helen

 

[htmlguru4242]

That is a pretty cool device; an optomechanical thermal camera.

I wonder how hard it would be to build one of those ...

 

[nick mulder]

sorry, I thought I had subscribed to this thread - but I see that there has been discussion going on the meantime.

Well, My experiments with my nightvision scope through the finder on my Bolex proved to be of dubious quality so I passed on the probeye idea and am using my $$$ to buy other more useful items (more milling bits'n'bobs) and time into making xtal sync drives and motion-control gadgets instead...

Still its an interesting conversation...

thanks!
Nick