Converting non-visible light to visible spectrum for photography

[js2020]

TL;DR

I would like to capture light from a wider spectrum than just visible light

I am trying to capture corona discharge on metal conductors. I am using a Canon DSLR and it works pretty good. I think there may be light from outside of the visible spectrum being emitted during these discharge events so I would like to see if it's possible to capture this light and actually see it using a DSLR. I know that I can get a high dollar corona camera, but I would like to know if there's a DIY way to convert my DSLR into one. it doesn't have to be perfect, just better than what I currently have.

 

[DaveE]

Optics that can pass UV light are expensive. So you probably don't have that. The sensor in the camera probably could respond to near IR, but they have undoubtedly included filters to block those frequencies. That camera was designed for people who DON'T want to have IR or UV mess up their photos, so it may be difficult to re-engineer it. If you have a source of just the IR or UV you want to extend to, you could experiment with removing the lens and illuminating only the sensor to see if it is feasible.

I have heard of people that have removed IR filters from web cams to see near IR though.

Before digital cameras you used to be able to buy IR sensitive film for use in normal cameras, they didn't have to filter out near IR because normal film wouldn't "see" it.

 

[Rive]

UV photography has some DIY sources: you can check around a bit but it requires a lucky camera and some skills.

 

[hybarzilay]

Hi,
You can capture corona discharge at night with your canon DSLR and get beautiful pictures. I'll be happy to see them myself. But if you want to see corona during daytime, you must have special optics, a filter to block the visible spectral range, and an image intensifier to enhance the UV radiation that is in the solar blind zone. There is a movie clip at ofil's youtube channel, called DayCor technology. See if you can implement it in your DSLR.

 

[js2020]

DaveE,
Thanks for the info. I'll see if I can find some DIY where they remove the filters. However, I will likely just get a different camera or deal with what I have. I'd hate to break it and be left without anything.

Rive,
Thanks for the suggestion. That's interesting. I'll look into some of the cameras listed on their site or see if mine has that option.

hybarzilay,
Yes I actually have a Canon EOS T7 (ES 2000D) that I've been using. It captures the visible light from flashover pretty well, as I'm doing this in a high voltage test chamber. I can also capture some purple corona too. I haven't been able to find anything on the specific wavelengths I can capture with it so I was hoping to find a way to do it guaranteed. I'll see what I can get from the clip you mentioned and keep looking into digital camera. I will probably look for a higher ISO camera to make the images of corona more crisp.

Here's a picture of corona on a connector. You can ignore the green. I have a PD checker on the floor with a tiny green LED that I didn't notice until I looked at the picture. Hey I guess it helped to illuminate the room just enough to make out what I have though...by the way, I have my HV lead connected to the wire and connector I'm testing. Their connection is in oil which is what's in the beaker. The purple on the glass is reflection from the oil.

 

[DaveE]

I will likely just get a different camera or deal with what I have.

yes, I agree, that camera body is too nice to screw-up.

 

[Baluncore]

I would like to capture light from a wider spectrum than just visible light
I am trying to capture corona discharge on metal conductors.

You could use an image intensifier/converter.
But why image the ionisation and convert that to optical?

Lightning rapidly heats and expands air, causing thunder. A corona also heats air, so you can expect to hear an ultrasonic crackling sound. Since the ultrasonic wavelength is short, and the pulses are very short in time, the corona breakdown could be imaged with an array of ultrasonic microphones and a computer to do the correlation.

The pressure wave radiated by the ionisation would have a dipole pattern. From that, with a 3D array, you might get an image of the (forked?) path taken by the breakdown.

 

[js2020]

Really what I was hoping to see if I would be able to image the same spectrum I can detect with a SiPM. Since the SiPM can detect light beyond the visible spectrum, I wanted to see if I could capture the same with the camera. I'm not sure what my emission spectrum looks like at this point so I'll need to find that out anyway. I'll look into acoustic detection for this as well. At the moment, I'm interested in this for locating problem locations in a larger medium/high voltage assembly. There are a lot of circuit boards with small chips, as well as wires and interconnects. This is a useful solution for locating problem areas, instead of troubleshooting the larger assembly, but I would like to gather as feasible to ensure what I'm looking at is correct.

 

[Baluncore]

For years, insulator faults on power poles have been identified with ultrasonic microphones at the focus of small dish antennas.
But if the fault is visible, heat is sufficient. An IR sensitive image intensifier is all that is required.
IR cameras that plug into a mobile phone for display are being used by power companies.

I have used a clear vinyl tube between my ear and guided by hand to search out electrostatic discharge on electronic circuit boards. The CVT is both an insulator and an audio wave guide. The sound gets louder as you move the open end of the tube closer to the fault.

Mounting an ultrasonic microphone on each PCB could automate the detection of a failure. A microcontroller would monitor and report sound level, raising an alarm if there is an increase above a programmable threshold.

 

[js2020]

I have a few thermal cameras so I'm curious to see if anything shows up that way.

I've never heard of using a CVT. That's another interesting approach. I won't be able to use it in my application as I do not have access to the DUT; it's in a closed HV test chamber. Multiple ultrasonic microphones and and rebuilding the location information may be one way. An ultrasonic microphone for each PCB would be a neat project too. I'll try an ultrasonic microphone for simple detection and put the per board method in my memory bank for future exploration.

 

[hybarzilay]

DaveE,
Thanks for the info. I'll see if I can find some DIY where they remove the filters. However, I will likely just get a different camera or deal with what I have. I'd hate to break it and be left without anything.

Rive,
Thanks for the suggestion. That's interesting. I'll look into some of the cameras listed on their site or see if mine has that option.

hybarzilay,
Yes I actually have a Canon EOS T7 (ES 2000D) that I've been using. It captures the visible light from flashover pretty well, as I'm doing this in a high voltage test chamber. I can also capture some purple corona too. I haven't been able to find anything on the specific wavelengths I can capture with it so I was hoping to find a way to do it guaranteed. I'll see what I can get from the clip you mentioned and keep looking into digital camera. I will probably look for a higher ISO camera to make the images of corona more crisp.

Here's a picture of corona on a connector. You can ignore the green. I have a PD checker on the floor with a tiny green LED that I didn't notice until I looked at the picture. Hey I guess it helped to illuminate the room just enough to make out what I have though...by the way, I have my HV lead connected to the wire and connector I'm testing. Their connection is in oil which is what's in the beaker. The purple on the glass is reflection from the oil.

 

[hybarzilay]

Dear JS,
It took EPRI 3 years to develop a daytime bi-spectral camera that enables seeing and recording UV photons of corona and the emitting sources. Since 2000 the cameras that EPRI developed evolved into HD compact handheld bi-spectral. I am not trying to discourage you, but I would like to offer you help.
[Personal contact information deleted by the Mentors]
best regards
hannah