I like DIY. And this exercise was something I wanted to turn my hand to.
I bought a 'broken' 1000D on eBay. Except for a mainboard with old firmware and the reason it's not working all the parts are in excellent condition as spares.
The overall concept is to avoid changes that cannot be reversed.
Nice work and idea. Was this just a technical exercise or have you got the camera functional enough to see how it goes ?
I'm looking at a similar idea using an old Film SLR body and my existing DSLR lenses but adding the sensor from a web cam or using my NexImage system. If I can also get cooling it would be a bonus and as you say 'something I wanted to turn my hand to. '
Good ol' DIY is alive and well for sure.
Cheers
Brent
Last edited by ZeroID; 08-12-2011 at 06:29 AM.
Reason: Spelling .... !!
Hi Brent. Getting close to installation in the camera and adding the Peltier and heat sink. If all goes to plan, I'll have it up and running just before Christmas. Temperature sensor and control to add - plus writing the software for the Arduino.
It's mainly the technical challenge. Taking what others have done and making what I think are improvements - that remains to be seen. Lots to consider when you get into it.
The whole thing could be simpler and less work except for the cutting and bending around the camera framework and internals.
Your idea might be a lot easier. Localizing the cooling and reducing the mass to be cooled, plus good cooling of the back of the Peltier and sealing to prevent condensation around the sensor and electronics seem to be the main considerations so far.
Two more images. Sensor electronics are protected with a layer of BluTak. Green actually. The Finger is insulated with double sided tape and duct tape -tight fit in places. Temp sensor, shutter remote and USB will be fitted prior to fitting the back cover, alp.g with additional insulation.
Two more yet again. Assembled and camera working. No function disabled, except for auto cleaning. Manual cleaning only.
Need a less bulkier remote jack. Space is adequate but tight. USB is available for remote or tethered shooting.
Next, attach TMP36 temp sensor. No calibration necessary. Range -40C to +55C. Voltage change across diode with temp change. Analog from temp sensor switches a logic level MOSFET to control current to Peltier device. No PWM to Peltier or heat sink fan. Temp control +/- 0.1C, but it remains to be seen whether this is possible in practice. I will post a a readout when it's up and running. here
And I've picked up a KM film SLR camera for $20 on TradeMe. Lens included. Time to get the fine tools and the magnifying glass out and start modifying.
Looking forward to seeing the results Brent. The expensive bits for me were the power supply and Peltier.
Two more pics, yet again. Temp sensor as close to the sensor as I could get it without cutting bits of here and there.
Modified remote shutter jack. Miniature is good here. I didn't use shrink tube and coated all the electrical stuff in silicon.
The attachment plate for the heat sink Peltier is fixed in place with double sided tape. Not the soft foam type. Interesting to see what effect cold has.
Camera was not as good as expected. You get that when you buy unseen. May be able to use it but doesn't accept my lenses (camera is too old) and the one lens that came with it has fungus and is wobbly. Shame because it was once a nice lens.
Oh well, $20 down the sink ...
Funny thing was there was a couple of new batteries in the bottom of the camera bag. Loaded them in and off she went just like a good thing.
Just can't keep them down can you ...?
I'll keep looking ...
Next is the wiring loom for Peltier, current switching FET and fan power. A stabilizing bracket is needed between camera and heat sink to prevent flexing of the cold finger
The heat sink is mounted slightly rearward to balance the weight of the lens.
Progressing - added two more images. A stabilizing bracket is needed to prevent flexing of the cold finger with the weight of the heat sink. Bracket will attach to tripod mount thread and the heat sink mounting plate. As mentioned, some wiring to complete.
The heat sink may seem like overkill. However, I selected it to have a significant margin over the sum of power dissipation requirements - that is, Peltier plus the power required to drive the module.
DeltaT is determined by hot side dissipation. I'll see how it performs.
Well done Rowland - looks very similar to my Central DS mod.
I'm afraid I don't possess your tech skills or nerve to try something like this.
Look forward to seeing how the whole thing performs under the stars.
How low will your mod go, temperature wise - have you got to testing it out?
Cheers
Doug
Thanks Doug. I hope to have it complete this weekend. Conservatively, I'll say 25 below ambient. I'm hoping for 40, but more importantly, aiming for reasonably tight temperature control.
Thanks Daniel. It's taking shape. Interesting to see how it performs.
The latest is a screen shot of the temperature sensor readout, polled every second. The sensor is reasonably stable with the odd spike +/-, > 0.3 degrees from average. Polling at longer intervals might be better for the control system, given that temperature doesn't change rapidly.
I will write up a more detailed description and post some more images soon. And test the results with some more cooling runs.
Here's the last screenshot. Ignore the temp control code. I edited the code just before taking the screenshot, and ran that afterward. Not happy with the resolution and think that polling the temp at 2 second intervals and using actual temperature or voltage might produce better results. Then again, I may need to change the control method.
You can read about the modification here. It is quite involved, and I'm hesitant to provide any more than a general overview. In-fact, the list of resources is substantial and the risk moderately high.
Some improvements are on the go as we speak. I'll post at a later date.
For now the mod consistently cools 34C below ambient and can be cooled to a set point temperature through the Arduino hardware/software.
DeltaT can be increased by use of a less conductive and better insulated heat sink mounting plate, which is sucking energy out of the system - 3mm acrylic should be sufficiently non-conductive and isolate cooling to the cold finger only.
Here's another little project using an Arduino to control a lens attachment that I use to take calibration frames - it illuminates for flat frames. The sequence is automatic. Screw it on, flick the switch and let it do its thing - go to bed and in the morning, nicely baked bias, dark and flat frames.
