Jump to content
UltravioletPhotography

UV with Raspberry Pi & Smartphone


Andrea B.

Recommended Posts

Wow. I have always wondered how the bayer layer was chemically removed. Now converting to a very sensitive UV sensor seems easy I have a canon T3 sensor sitting in plastic case for over a year trying to figure this out! ( did not want to use the scratch method!) I'm a bit worried about dipping the sensor in acetone though because it might melt some plastic parts. Might just stick to the alcohol wash. Thanks for posting it. I would of probabaly never found it and companies like MaxMax would never tell me because then they would lose revenue. Haha secret is out!

 

Whilst the filter can be removed by careful scratching (a range of tools, such as metal tweezers or a pointed wooden object, can be used for this process), a far more uniform finish is achievable chemically; we adopted the latter approach using a five step procedure. We first submerged the sensors in photoresist remover (Posistrip® EKC830TM) and heated (70–100 ◦C) until the filter was entirely removed; this generally takes 10–30 minutes, depending on the level of applied heating and agitation. A second bath of this remover was then applied to optimise the cleaning procedure. Following this, the photoresist remover was washed from the sensor by a three stage cleaning procedure, using successive baths of n-butyl acetate, acetone and isopropyl alcohol. All the above chemicals were applied undiluted. For each of these steps, the sensor was submerged on the order of minutes, Sensors 2016, 16, 1649 3 of 8 to ensure a thorough cleaning. Careful and rigorous application of this methodology can result in uniform and clean sensors which exhibit greatly enhanced UV sensitivity relative to non-de-Bayered units. For example, a clear-sky image taken with a partially de-Bayered unit, captured through a 310 nm filter and with a shutter speed of 400 ms, exhibits an increase of ≈ 600%
Link to comment
Actually I'm not sure that MaxMax uses a chemical method?
Link to comment

MaxMax in a discussion about the manufacture of monochrome sensors makes several remarks about the used fabrication gear they picked up after it was no longer useable for new fab:

 

We have (at the moment) 11 spectrometers which can measure from 200 to 1800nm. Spectrometers are built to be optimized for certain spectral ranges so one might be for 200 - 400nm, and another for 900-1800nm, etc. Just in spectrometers, new they would have cost about $70,000. We have an ion milling machine that new would cost around $1,000,000 that we bought for $5,000 (it needed some work but most of the parts where there).

 

Fortunately for us, [the decommissioning of fab equipment] results in a somewhat small line of business of converting color sensors to monochrome by removing about 5 microns of the microlenses and CFA from the surface using some of the decommisioned semi fab equipment that has moved down the food chain.

Link to comment
I never saw the tape method, but after reading that webpage it only works on the very very old 20d. and I need live view. I'll report back once I find out how much the chemical costs.
Link to comment

https://www.edmundoptics.com/p/6mm-dia-x-9mm-fl-uv-ar-coated-uv-plano-convex-lens/6719/

this looks like the lens they recommend in the article. Might try with cheap raspberry pi camera first

 

https://www.edmundoptics.com/p/10nm-cwl-125mm-dia-hard-coated-od-4-10nm-bandpass-filter/33098/

 

I think this is the UV filter they used above, but doesn't it also need an IR cut off filter?

I thought most bandpass filters transmitted in the IR range? the article does not address this? seems like a major flaw in the research.

Link to comment

We did talk about this here:

https://www.ultravioletphotography.com/content/index.php/topic/3268-chemical-removal-of-bayer-filters-from-sensor/

 

The chemicals are cheap, but very dangerous. Do this only outside and have sand ready to drop on the area if fire gets out of control. The flash point is close to the heating temperature.

 

Andrea,

LDP, MaxMax uses a chemical method. I chatted with Dan about converting an Em5mk2 a year ago. The chemistry he uses requires the older Aluminum wire sensors. At that time he could not convert the newer copper wired sensors.

Link to comment
I'm guessing the chemicals destroyed the gold wires? epoxy coating them first is probabaly a good idea, lol.
Link to comment

If gold was used we would be good. Gold doesn't oxidize and is a good conductor. Thats why the USA shuttle used gold.

 

Sadly our cameras use the better conductor copper, which does oxidize and becomes much less conductive. Aluminum oxides might be easier to clean or may not alter the transmission as much. But aluminum is not as good a conductor as copper.

The best is pure silver.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...