Another thread with another 450D cold finger build. I am using the proven cold finger design but without the step.
First of all I will need to give credit to the people who already "walked this path" and who did all the designing and testing and put everything on IIS. It makes the lives of people who follow much easier. Thank you guys.
To start with a question... did you cut the tab on the LH side of the DSLR back, just above where the cold finger exits? I have more-or-less finished the copper cold finger but the back of the camera cannot be closed. There is about 1cm wide tab which hits the cold finger. Marked red in photo 4 and shown without the cold finger in photo 5.
Luka, remove the metal tab, it is only there as an earth path and there are others. If you bend it back and forth it will break off near the base eventually, that is safer than trying to cut it off with a dremel tool and creating metallic dust. Rowland may have done something different. Trim the plastic as required if it still hits the finger, but leave the long tongue which has the screw hole in it, depending on your finger width you might have to shave it slightly but you need it. I used a hack saw blade to cut the plastic along side the screw tongue which you must retain then some sharp electronic flush side cutters to nip the plastic tongue to shape. It is soft plastic and will not crack and run, just take it slowly and refit to test as you go along.
The cold finger is designed around a 1000D which doesn't have the tab.
That explains it, thank you Rowland.
I just could not be stuffed with the step after spending an hour nibbling and filing. But if I anneal the copper before bending, the step can even be done by fingers (not sure how long before copper regains the hardness after annealing).
Richard, I have not seen your build before, very nice. Thank you for the link. That is a very nice flat heatsink/fan combination, now I am considering using that one instead of my old chunky CPU cooler.
That is an excellent point Glen which I have forgotten about. I may try my old CPU heatsink first and see how much power I actually need to keep the temperature around zero.
I just could not be stuffed with the step after spending an hour nibbling and filing. But if I anneal the copper before bending, the step can even be done by fingers (not sure how long before copper regains the hardness after annealing).
Richard, I have not seen your build before, very nice. Thank you for the link. That is a very nice flat heatsink/fan combination, now I am considering using that one instead of my old chunky CPU cooler.
Copper work hardens after annealing when you start bending and working it. Anneal again and repeat. And 'annealed' isn't quite finger bendable to the degree needed for this mod.
That being said I chucked the wee board out and made my finger straight as well. Much easier. It is sandwiched between two pieces of Styrofoam as insulation. Also chopped off the bent tang, the more copper you have there the better the conduction rate.
The more copper the more weight you add, you really just need enough wide to completely cover the back of the sensor, and you can make the outside the right size for your TEC. Don't forget that you need to attach the heatsink/fan combi to the cold finger and sandwich the TEC. I used Artic Silver thermal paste but it can get messy. There are some great thermal pads around now but they cost much more than paste. I used two nylon screws and bolts on either side of the TEC to hold the heatsink onto the finger. I put a nylon lock bolt on each screw as well, you want it solid and unable to move. Don't use metal screws or bolts as they will degrade thermal performance. Don't forget to put your temperature sensor on the cold finger just where it goes behind the sensor, you have to have one in there to be able to do set point cooling control.
Brent i am not a fan of styrofoam in there as it compresses and does not stop cold finger movement. Sure it insulates but the only thing supporting the cold finger in that situation will be the three sensor mount screws and the leverage put on the sensor board by the weight on the cold finger is not good. Imho, a solid spacer, like my carbon fibre one, is preferred to stop finger movement. Really anything could be used that will not absorb condensation or compress. A piece of nylon cutting board machined to fit would be ok.
Glen, a really stupid question, where did you get the regulator for your Argon gas bottle (partially shown here).
I wasted 20min today talking to an "expert" sales person at Bunnings about the regulator for their Argon gas bottle (same bottle as yours). He said that the bottle ONLY connects directly to a MIG welder and that regulator for it does not exist and there is no way in this universe to get the gas out of it (apart from the MIG welder of course). I even showed him the photo of your bottle with a partially visible regulator and he just kept saying it is something else... amazing ignorance. I am planing to go to a different Bunnings tomorrow.
Do an ebay, or general search, for a" disposable gas bottle regulator". Lots of them available, even beer brewing sites have them. I think i paid about $29. Get one that has the nipple barb with it so you can attach a hose. Hose from Bunnings in the irrigation fittings section. I could not find a regulator for it at Bunnings either. Disposable gas bottles have a standard thread, so any regulator designed for them will fit.
By the way, I am not sure how much gas is actually needed but larger disposable gas bottles are cheaper than the ones from Bunnings/Masters. See for example here (more than double the size and higher pressure).
Well i had two choices, Sikafkex 227 or aquarium silicon, i used the aquarium silicon on the MC Clear over the sensor, you need very little and do not get it on the sensor surface. I flooded the area with argon while i was fitting it.
On the body chamber sealing MC Clear EOS Clip-in filter i used Sikafkex 227, as it is black. Not sure that level of dual chamber protection is required, especially if you are running a UV/IR Cut filter at the front.
Just a thought. If you were to place the sensor in a plastic tray/Tupperware and fill with argon while sealing and fitting the glass would this capture argon between the glass and sensor... I imagine it would?
Just a thought. If you were to place the sensor in a plastic tray/Tupperware and fill with argon while sealing and fitting the glass would this capture argon between the glass and sensor... I imagine it would?
Yes Rowland it would, i used a plastic storage box with argon in it when i fitted my cover glass. Both silicon and polyurethane require moisture to fully cure and in an argon bath there will be no moisture as it is displaced with the air. So fit it in the argon then remove it to air to cure. As long as you can see that you have no gaps between the cover glass and the sealant it should be fine. As far as outgassing is concerned, you would be using a very small thin layer of sealant, it will cure from the outside (air side) in. I have seen no evidence of moisture under my cover glass, even at -27C where it should freeze on a surface.
I am a bit confused. Isn't the MC Clear glass installed instead of the low pass filter 1, i.e. it sits in the holder that previously had the IR filter (as per Gary Honis' description in point 31 here).
In this case it can be glued to the holder, left to cure for as long as needed and then simply fitted to the sensor under argon bath.
Few photos of the assembly of the "final product".
1. I soldered the grounding wire straight to the cold finger. Needed to hold the soldering iron on it for few minutes to get the whole copper plate hot. If you are doing this make sure that the copper is on something that can take the heat.
The other end of the wire is soldered to a small piece of copper with two holes that screw into the holder for the video board (which has been removed). See images 3 and 4.
2. Finger coated with Armour All Custom Shield (same as plastidip).
3. Sensor with cold finger showing temperature sensor ds18b20 and the small copper plate for grounding.
4. Camera insides fully assembled (apart from the back display).
More to follow but I took the camera like this (with back display mounted) out last night, i.e. without the peltier and cooler. I thought that the copper in the cold finger will provide some cooling of the sensor. I left it taking 3min exposures. Ambient was 10C, the finger/sensor was 16C, the exif data in the photos was 21C. Even without a peltier/cooler the finger made a huge difference.
The support frame was 3D printed (I only had red filament left). The hex-head bolts were heat-pressed into it. Nuts were screwed on the heatsink side after drilling the thread out on the heatsink. Cooler is ID-COOLING Iceland Series IS-50.
I ran a quick test and got to -2C within 2-3 few minutes (ambient temperature was about 20C). It could have gone lots lower but I stopped it because of possible condensation inside (it is just on my desk).
Todo:
- The fan on the heatsink is oversized. This may be a good thing to keep the camera warm or even to redirect this airflow onto the field flattener to keep it warm, instead of using heating tapes.
- Also may need some rubber supports for the fan to reduce vibrations.
- Bag the whole thing and fill with Argon.
- Finish the PWM software controller.
(not sure how long before copper regains the hardness after annealing).
