Public Lab Research note


Canon A2400 IR conversion

by jbest | November 06, 2013 18:15 | 57 views | 9 comments | #9740 | 57 views | 9 comments | #9740 06 Nov 18:15

Canon A2400 IR conversion

The excellent tutorials for the various Canon PowerShot conversions inspired me to attempt a conversion of a Canon A2400. I chose this camera thinking that it would not be that much different than the others and because it is small and light and affordable at $80. Because I thought it wouldn't be much different, I wasn't as prepared to document the process but I quickly found out that it was in fact quite different than what I've seen and read about before. I was able to successfully complete the conversion with a functional camera, but I'll warn everyone up front: this might not be an ideal option until we can develop more community input on the best way to disassemble and reassemble it. The primary problem was that the lens and CCD assembly had to be removed and the ribbon cables detached. Hopefully the following notes and photos will help others improve on the process.

Process

Removing case

In other examples of IR conversion, some models could be modified with the front half of the case in place, but I had to remove it in this process.

There are six external screws to remove - two on the bottom, right, and left. One is under the USB cover.

IMG_2856.jpg
Left screws.

IMG_2859.jpg
Right screws (one already removed, one under USB cover)

IMG_2858.jpg
Bottom screws

With all the screws removed, remove back case (black) by carefully separating at the bottom and hinging at the top. You may need a fingernail or screwdriver to separate the top part. The back case has two rubber covers for the USB port and battery compartment (for external power supply). As you remove the front cover, note how their "tails" are looped over plastic pins on the front cover. Make sure you replace them correctly on reassembly.

IMG_2860.jpg
Back cover removed with rubber covers still attached.

Getting to the CCD

At this point I was expecting to just remove the back cover so I was removing every screw that made sense. The next series of photos will show most of these steps, but I'm not sure I captured all the details.

IMG_2866.jpg
The LCD panel and another screw on the bottom. This screw was covered by the case. I did not remove it at this point, but eventually it does need to come out.

IMG_2867.jpg
The LCD panel lifts from the left side to gain access to the screws underneath. These screws hold the CCD and lens assembly to the frame.

IMG_2868.jpg
Ribbon cable attaching the LCD panel to the body. This is one of the few ribbon cables I didn't separate.

A thin metal plate sits in the LCD tray/frame. Remove this and remove all the screws in this area except the three recessed screws in the middle that have glue on them. These hold the CCD in place and will be removed later. At this point I still had the front cover loosely dangling but attached but it was connected by a little adhesive that held the shutter release wires to the front case.

IMG_2871.jpg
Camera frame and metal plate

IMG_2872.jpg
Front case removed from body.

IMG_2873.jpg
Front of camera body

At various points of the disassembly and reassembly, you'll need to lift edges of the button and sensor unit on the top of the camera and re-secure it. There are little plastic tabs that catch on the frame underneath. There is just one screw under a ribbon that will need to be removed in a later step.

IMG_2875.jpg
Top of camera with buttons, sensors, and small battery.

At this point I'm still trying to figure out how to get inside this to get to the CCD. With so many screws out, the only thing holding the various parts together were the ribbon cables.

IMG_2876.jpg
Flash unit with capacitor still attached. Eventually I accidentally separated the cable, but it is probably necessary in order to access the CCD.

IMG_2880.jpg
Screw under ribbon cable on top of camera. I can't remember if this has to come out, but I did remove it at this point.

IMG_2882.jpg
With most of the screws removed and a little pulling and shuffling, it all falls apart. Note the separated CCD ribbon cable.

At this point, the other tutorials cover the details of removing and replacing the IR filter so I won't cover that here because it didn't seem to be different than other Canon camera models. The three screws on the CCD have glue but could be removed fairly easily.

IMG_2883.jpg
Detail of the CCD and lens unit

IMG_2884.jpg
CCD removed with rubber gasket still in place.

IMG_2886.jpg
IR filter removed. I put the blue filter in at this point but didn't take a picture.

IMG_2887.jpg
CCD back in place.

Eventually the second ribbon cable of the lens unit slipped out while I was trying to reassemble. You will probably need to re-insert these a number of times so it's important to get comfortable with this. You can't just shove them back in. You have to lift the light brown plastic clamp that is on the back side of the ribbon jack, then slide the ribbon in, and lower the clamp. In this picture the CCD ribbon is out and the clamp for the CCD ribbon is up, ready for the ribbon to go in. The other ribbon is in with the clamp in the down position.

That was the last picture I took. I probably made a number of failed attempts to reassemble because the ribbons kept slipping out as I tried to fit everything back together. At this point I don't have enough experience to provide any more pointers on how to do this better, but it will take patience. All the various parts seemed to not quite fit well until I started to put the screws in and it all snugged together and became solid and like new. I do suggest that you start reassembly by ensuring that the CCD and lens unit cables are in place then carefully replace the screws into the metal tray on the back that hold the CCD unit to the frame. Once that is secure, you don't have to worry about those ribbon cables coming loose.

Questions and next steps

Hopefully these notes will serve as a starting point for someone else to find a better, more reliable way to make this conversion. We hope to conduct some NIR aerial surveys using this camera paired with an identical, but unmodified camera. If any problems or new information arises from that in relation to this conversion, I'll update this note.

Note: Originally I thought that my conversion "broke" the flash but today when I was trying it out, it worked so I've edited the note to reflect that. When I first thought the flash didn't work, I might have had the flash disabled, or maybe the capacitor had lost all charge when I was disassembling and didn't have time to charge when I immediately assembled and tested. In any case, the flash is working now, but the challenges of the detaching ribbon cables still remain.


9 Comments

Oh, man -- Cindy Regalado and I converted a few of these with a group of students at the Waag Society about a month ago and they were a real pain! only 2 of 3 made it.

Impressive documentation, thanks! I'd tell newbies to avoid this camera, to be honest, compared to the ease of an A495. I'm adding it to the camera conversion page too: http://publiclab.org/wiki/infragram-convertible-cameras#Acceptable

Reply to this comment...


Wow, quick response! I was going to also add this to the "acceptable" category so I'm glad you already did that. I too would not recommend this to anyone doing this for the first time (though I fell into that group) or who aren't willing to risk destroying a camera. Had I known how risky it was, I might have chosen a different camera, but since it worked, I think it will be great for our use. I'll be testing the battery life on some small surveys, and if it isn't great, I'll be putting together an external power supply for dual cameras and will possibly trigger the shutter with an Arduino or RasPi (easier to sync from what I understand) instead of CHDK since we are looking into using barometric and temp sensors and will have other hardware up there anyway.

Reply to this comment...


Thanks for suffering through this. I just converted two A2200s, and they are quite straightforward compared to your experience. As far as I can see, there is not much difference between the A2200 and A2400 that would influence aerial or infrared photography: same sensor size and Digic 4 processor, and similar battery. The A2400 has 16 MP instead of 14 MP, but I consider that to be a liability. The A2400 has 5x instead of 4x zoom, but that is mostly irrelevant (the widest focal length is the same). So the choice between the two is clear if you want to convert one.

I am curious about your plan to fly an unmodified and infrablue camera and sync the shutters. If you fly a pure infrared camera (Wratten 87 instead of infrablue) you will have a more pure NIR channel for making NDVI or NRG, although you would have to align the photos from the two cameras. Another way to get a pure NIR channel is to use a red filter (Wratten 25A) instead of blue. But I do not want to suggest that you take that camera apart again!

Reply to this comment...


The only reason that I would want to fly both at the same is to kill two birds with one stone and get both an NIR and visible light image (in "true" colors) without flying two surveys. I'm not so concerned about precise camera alignment because at this point I don't have a need to combine the two images. For that matter, precise syncing isn't important yet either, I just wanted to experiment with external triggering. Originally I had planned on two unmodified, aligned, synced cameras to make stereoscopic images but since then I've learned a lot and found that tools like VisualFSM and others can give me the results I need without stereo cameras. I'm interested in trying other filters like the Wratten 25A you mentioned (and mentioned at http://publiclab.org/notes/nedhorning/10-30-2013/red-vs-blue-filters-for-ndvi ) and I'm willing to open the camera up again to see if that might fit our needs better. I'm still deep in learning phase and trying to figure out how to apply the techniques that Public Lab has pioneered so I'll be experimenting more and asking a lot of questions in the near future I'm sure!

Reply to this comment...


Just word of caution if someone is planning to convert his Canon IXUS 132 in to NIR or NGR.

I own a Canon IXUS 132 and done the infrablue convention a few days ago. It has similar internal constructions and layout as the Canon A2500 and I have encountered similar issues with dismantling of the camera and installing my #2007 infrablue filter.

Special care must be taken when you reassemble the camera not to pinch or damage the electrical ribbons. Unfortunately I did not come across this useful webpage before I did my convention so it took me about 4 hours to complete this exercise by trial and error.

After doing the White Balance adjustments I started to take some infrablue/NIR photos. I realised that all my photos were blurry and out of focus. I tried playing around with the Automatic /manual focus adjustments but still can not get to focus properly. The only time it focus and produce a clear picture is when I zoom closer to an object about 10 to 20m away.

Can someone explain this or advise me on this..? Jaco

Is this a question? Click here to post it to the Questions page.

Reply to this comment...


Hi, Jaco - perhaps you should check out this set of issues, which others have figured out: http://publiclab.org/wiki/infragram-convertible-cameras#Focus

Hope that helps!

Reply to this comment...


Jaco,

The fact that your camera now focuses only in certain ranges suggests that the sensor is the wrong distance from the lens. At the factory, the screws through the sensor plate were adjusted to hold the sensor plate a particular distance from the lens and perpendicular to its axis. You have to recreate that position when you replace the sensor plate. If there were springs under the sensor plate, this can be even more important. It is possible that tightening or loosening those last screws will fix the problem.

Chris

Reply to this comment...


Sorry I want to rectify my mistake above when I was referring to Canon A2500, it should be Canon A2400 which has similar interior layout as the Canon IXUS 132.

Reply to this comment...


Chris & Warren thanks very much for the valuable advice. Now it’s back to operation "delicate" by opening this camera to play around with the sensor plate screws tightness. Jaco

Reply to this comment...


Login to comment.