CCD chip in mineral oil...

On Monday, May 12, 2014 1:38:03 PM UTC-7, snipped-for-privacy@yahoo.com wrote: [about mineral-oil fill of a camera, not focusing well]

y cheaping-out is really part of the proof-of-concept notion. While I'd lo ve to make some money off this, if JQP can look at it and say "I could do t hat with $100 and an afternoon", so much the better...

So, forget all the exotic fluids. Most clear liquids have refractive index nowhere near air, a off-the-shelf refractive system is impractical. Get yourself a mirror-focusing system, instead, and stick with the mineral oil. Use a Newtonian telescope light path, instead of the lens.

On Tuesday, May 13, 2014 1:52:17 PM UTC-4, DecadentLinuxUserNumeroUno wrote :

als.

g.

Thanks, that makes a lot more sense, but I'm probably still missing why. I don't think the RI is significantly different from mineral oil and if it's for the "spontaneous combustion" issue, there shouldn't be enough oxygen in the system to support it anyway.

my cheaping-out is really part of the proof-of-concept notion. While I'd love to make some money off this, if JQP can look at it and say "I could do that with $100 and an afternoon", so much the better...

Now there's some thinking. For a one-off it would almost certainly will cos t more than the rest of my project combined, but it completely bypasses the problem and if someone mass-produced them, we're probably looking at 10 ce nts per unit in manufacturing costs to work with this specific camera.

Hmmm... I have an old, cheap mirror based telephoto lens laying around that I wouldn't mind sacrificing It might fave internal lenses, but that might not pose a problem. Worth a look.

Avoiding the mineral oil is to avoid amusement when pO2 in any trapped bubbles exceeds 1atm (together with any adiabatic heating).

You need something more like a Cassegrain microscope objective for this. eg.

or

This is getting into cost no object territory.

That's no real problem: a small bubble won't heat, it'll dissolve into the oil (and at low O2 concentration, any oxidation will just make some of the hydrocarbon into carbohydrate (alcohol). Purging, vacuum backfill, bleeding (like in a brake system) are pretty effective at oxygen removal.

Well, sure, if you want something commercial-off-the-shelf for a solution. But, you can make a pretty good spherical mirror with a couple of clean aluminum plates, a big ball bearing, and a 75 ton hydraulic press. It's a lot easier to afford solutions when you don't search catalogs: compare the cost of a couple of tin cans and a length of string, with a private telephone branch exchange!

Yeah, a few thousand dollars for an objective is out of my budget for now, but I don't think I need to resort to 'smithing my own mirror. I'm sure someone still makes those silly parabolic cigarette lighters for a few bucks (might need some polishing).

guessing this group will have the right people... I think I have more scientific knowledge than your average Joe, but this is beyond me at the moment.

done it, then there will be video and a detailed description on my web site), I'm running a small computer, submerged in mineral oil. That part is a well known technique and works fine with cheap materials.

camera, also submerged. Since I can't have any air pockets in this system, it would be helpful/simplify things if the mineral oil could be in direct contact with the CCD chip surface. This does not seem to work so far. I get a uniform, gray image. The camera works again once it "dries" out. What I'm trying to figure out is why (and what to do about it,of course).

stabilizer because it was handy. I haven't had much luck finding information on whether this substance is electrically conductive. Apparently the question doesn't come up much. :) I see a few references from reliable sources (not putting much stock in the hair-brained claims from supplement mongers) that it is an important Vitamin for "nerve conductivity", but none specify whether they mean "electrical" or "chemical" conductivity. My, el cheap-o mutitester shows the mineral oil as infinite resistance (it goes at least into the megaohms), but I assume the tiniest bit of conductivity would be enough to foul such a sensitive chip? Or is there some other factor at play that I'm not considering? At first I was thinking maybe a thermal problem but after several days soaking at room temperature, it still behaves the same...

submersion of electronics, but while I'm waiting, I though I'd pick some brains...

considering coating the chip with epoxy, under the lens, but as this will be irreversible, I was hoping to get some input on whether this has a chance of working before I sacrifice my (admittedly cheap) camera chip. Some distortion from the epoxy is acceptable, but I think I can mostly overcome that by sandwiching the epoxy between the chip and a portion of a microscope slide (it's not for anything critical and failure would only cost me a few hundred dollars and a week to order new parts).

other reasons why mineral oil might affect a CCD chip, or other chip coatings if epoxy sounds like it might be a problem. Thanks.

Sounds like an index of refraction issue to me. If you immerse the bare chip in mineral oil how do you get any kind of focus? BTW d-alpha tocopherol is a decent but not excellent insulator.

?-)

Not that i am sure how it is useful at visible and near IR wavelengths check out both cubic zirconia and titanium dioxide.

damn those are big tubes- "Technicians had emptied the tank for cleaning and had replaced a few hundred photomultiplier tubes (PMT), each 22 inches in diameter. "

Right, it was Kamiokande. But the two sort of rhyme, at least in English. ;)

Cheers

Phil Hobbs

I see you are busy making something :) What is the wildest optical gizmo you could conceive, money no object, but barring spece stations, etc? wildest = earth shaking. ?

When I start getting into serious pressure, it will likely be increasing ov er a period of hours and then decreasing over a similar period, so I don't really expect that to be a problem.

And yes, if this actually goes anywhere, bare "chips" should be an option. Half the fun at this point will be doing something that anyone with $100 ca n recreate, so I've also considered tracking down some acid to strip the ep oxy from the chips on the $40 linux computer I hope to use, but that might not be possible without doing the same with the circuit board, which is a m ulti-layer board and probably won't handle being "naked". It might also be considered somewhat dangerous without an investment in ventilation (and VER Y careful handling), and possibly result in byproducts that can't be easily (legally) disposed of. Yes, I might, theoretically be able to hand wire al l the chips together, but that might be more fragile than what I have no.

Boy, that reminds me of a story... :)

Den onsdag den 21. maj 2014 00.43.29 UTC+2 skrev snipped-for-privacy@yahoo.com:

ed

ar

over a period of hours and then decreasing over a similar period, so I don' t really expect that to be a problem.

. Half the fun at this point will be doing something that anyone with $100 can recreate, so I've also considered tracking down some acid to strip the epoxy from the chips on the $40 linux computer I hope to use, but that migh t not be possible without doing the same with the circuit board, which is a multi-layer board and probably won't handle being "naked". It might also b e considered somewhat dangerous without an investment in ventilation (and V ERY careful handling), and possibly result in byproducts that can't be easi ly (legally) disposed of. Yes, I might, theoretically be able to hand wire all the chips together, but that might be more fragile than what I have no.

what kind of pressure are you aiming for?

-Lasse

Roughly 16,000 psi...eventually...

Afterthought has occurred: there are lens systems that have no air in the imaging path, that use graded-index lenses. These are employed in (for instance) endoscopes. So, a standard camera might be available, or at least an off-the-shelf lens system, that DOES work just as well in a liquid medium, because it doesn't rely on reflection at curved glass/air interfaces.

with the one little caveat being the imaging area: it's about 1.8mm in diameter.

If both foci are right at the end faces of the GRIN, sure. Otherwise it does rely on refraction at the facets.

Cheers

Phil Hobbs

If I read the sales literature correctly, one face of the GRIN does butt up to the sensor. The claim, though, is that some of these imaging lenses can focus at infinity (which is the same focus underwater as in air). Plane wave into lens, focusing to a point at the image surface; that should happen regardless of the medium.

I don't mean that your suggestion wasn't an intelligent one--it certainly was.

The ray bending function of the GRIN will work fine in an oil medium, since it occurs inside the glass. There is a strong refraction at the facets, however, which will be quite different in oil vs. air. That will change the effective focal length of the GRIN rod by a lot, and will in addition change the aberrations of the system.

GRIN optics have come a very great distance in the last 20 years or so--AFAIK all new endoscopes are GRINs and not cascades of lenses like the older ones. I have no doubt that it's possible to design a GRIN system to work well in oil, but I do doubt that a stock one will do a good job for the OP.

Cheers

Phil Hobbs

I'm making assumptions, of course: mainly, that the front and rear faces of the GRIN are planar. If that assumption holds, the effective angular acceptance will depend on the index of the medium, i.e. the millimeters-from-center spot location will scale differently with the radians-from-axis incident light direction.

The GRIN that focuses at infinity in air, though, will likewise focus at infinity in oil or water. That is implied in the incident-plane-wave boundary condition at the front face. So, yes, the 'focal length' that identifies the effective magnification changes, but it doesn't lose infinity focus because of the medium.

It isn't a thin-lens system with an easy equivalent to lens location. The light rays follow curved paths, after all. I suspect GRIN systems have aberrations of unfamiliar types. How could one color-correct? Always use monochromatic illumination?