thinking of doing a DSLR cooling mod

[mindburner]

I have a Canon 1100d which is full spectrum,, having the IR filter already removed.

It takes great astro pictures but noise is still an issue. MY atik CCD is great but the sensor is small for some targets, plus I woud like some widefiled images and I just lov messing around with electronics.

The solution would be to get my DSLR cooled. I have been reading up on Gina's excellent designs and experiments and have decided to have a go myself.

Gina don't go on holiday or anything

The downside is I could total a camera body but they seem to be around £220 so it's worth the risk to have a large format cooled imaging camera

I made a little bracket to secure the cooler to the cold finger. Dremmels are great arnt they?

The copper is a bit long but can be cut down down.

This was just made up from some copper plate and a used copper CPU heat sink. The peltier is a 70w one I think, running at 10v

It drops from 25 deg ambient to about 7 deg, which it great for a first attempt.

The copper does mist up quite a bit, probably due to the device sitting in a warm room. Hopefully when it's boxed and sealed this will not be an issue.

[mindburner]

started today.

removed the sensor and ground down the little peg that was in the road.

Fitted the cold finger.

One point it's worth noting is that the bends in the cold finger are best as tight right angles, otherwise they get in the road of refitting the sensor assembly screws.

I have ordered printable silicone PCB protector from RS components to pain ton the imaging circuit board (and maybe all the PCB's)

Also some fine pipe lagging tape which I will cut to cover the inside of the cold finger and the DIY clip I made to hold the cooler on.

The little heat sink and fan will now get the temps down 20 degrees, so I was happy with that.

I will hopefully be able to reuse my cooling box as the enclosure with a bit of modification, just need to cut the back pate away as it was made from a single aluminium sheet. Plus cover the square hold in side. Might be able to use this as a desiccant window of something

Some more pics...

[mindburner]

a quick test at 2 degrees. Just propped the cooler on the cold finger.

This was just a quick test as I didn't want to get any condensation as I have not prepared the PCB's

left without cooling at 25 deg (30 deg on exif) then right with cooling at 2 deg on attached sensor (9 deg on exif). (brightness increased to show the difference)

[Gina]

Looking good

[mindburner]

Thanks Gina

I made a little sensor heater using a 47 omh resistor wired to a 4v source.

The frame was cut away a little bit, being careful not to impeded the sensors view.

It just just enough heat to raise the air temp in an equivalent air space to the sensor housing by about 2-3 degrees

not sure how effective it will be or even if it's needed, but decided to install while I had the sensor out

some pics.

[Gina]

Ah yes Another modder used some insulated nichrome wire as a heater. I think this is an interesting alternative to a sealed box with desiccant. Much easier

[mindburner]

hi

I think I will do the desiccant as well just to be sure. The box as well to get a controlled space.I have been working on 1 - 6 volt PSU for the supply to the resistor. But ended up using a little DC

to DC converter from China.

By the way. I have some pretty nice pipe lagging foam (3mm) which is great for insulating cold fingers etc, I have a huge reel if you would like a meter or so, I can send it to you.

[mindburner]

some progress and some problems. Always the way with these things. Painted the PCB's with FSC silicone and reassembled the camera. No boot:(

Reseated all the ribbon cables and discovered the awkward on top right is to blame if there's no power.

Then tried my 47 ohm resistor mod. DC>DC driver was getting very warm so checked the continuity and well it was shorted. Although I used heat shrink the leads of the ressitor must have been touching metal. So camera apart agin and got that sorted.

My BIG problem is condensation on the front of the sensor. The resistor works quite well, but as there's only one, coverage is not great.

I am sitting in a 25 degree comfy room with the camera cooling to 6-7 degrees so that's part of my problem, just wondering how much of an issue this will be in rear world, night time use?

I'm not convinced 2x resistors will be the answer. I saw someone on the forum used nichrome wire. How did that work out

[RAC]

The nichrome wire works very well.

[mindburner]

Hi RAC

would you have any more details on what wire you used. I have tried 4 x 47ohm resistors but without much success. The condensation seems to be one of the major issues when doing a mod like this

many thanks

[RAC]

The nichrome wire was from an old gauge like this

And then i just wrapped it around the sensor like this and powered it with a little lm317 reg at about 5-6volts.

Heater for Cmos sensor by meg rac, on Flickr

[mindburner]

that's excellent. It looks like it would give uniform heat without raising the sensor temperature too much.

My current resistor effort does works but I loose 5 degrees in cooling, plus they are closer to the sensor than the wire seems in your picture.

I shall have a hunt for an old gauge to salvage some wire or perhaps it may be available on small reels

thanks for you help

[Gina]

Pity insulated nichrome wire seem virtually unobtainable on the retail market except in 500ft reels. I had a good search when I wanted nichrome wire for dew heaters. Ended up having to use insulation tape which made it a fiddly job.

[mindburner]

hi yes it seems one needs to pay megabucks for a large reel when just a tiny amount would suffice.

I did have better success with my resistor mod, at 12.6v with both heating and cooling on, I still get about 18 deg drop with no condensation. At least that seems to be the case in doors.

Did you have condensation issues when testing Gina?

[mindburner]

here's a few quick comparison shots of cooling vs no cooling. I did stretch the living daylights out of them

Leftmost image is no cooling 5 min ISO 800

a bit more stretching:)

[Gina]

hi yes it seems one needs to pay megabucks for a large reel when just a tiny amount would suffice.

I did have better success with my resistor mod, at 12.6v with both heating and cooling on, I still get about 18 deg drop with no condensation. At least that seems to be the case in doors.

Did you have condensation issues when testing Gina?

Oh yes! Condensation is a BIG problem, only cured by nice freshly dried desiccant bags stuffed in all round the camera then the box sealed. But because I haven't yet managed a perfectly sealed box the damp has got in and started causing problems after a couple of weeks. I'm hoping v3 will be better - it's certainly a better design I've yet to get the time to complete it though.

[mindburner]

you should have a go with the resistor mod in lieu of getting some nichrome wire. I fins at 12.6 v, 4x 47 omm resistors, one on each side of the sensor works! at least indoors.

I had to modify the old black plastic filter frame to fit them in.

I love this modding, although it can be heart breaking when camera wont switch on. It's worth getting one of those 'keep calm and carry on' mugs

A pic of the new box. It has a perspex window for desiccant viewing. Getting a little light bleed so will have to hunt that down...

[Gina]

Do you think it might be due to your perspex viewing window? I had problems with light getting into the box.

[mindburner]

hi yes definitely a possibility. I have a neat little cover though now, so it's much better. Ten min subs with no light ingress.

Did get a weird straight line down one side of the image when highly stretched. Seems to match the physical line that exists down one side of the sensor. Like a frame or something. More prominent when cooled and at 1600 plus ISO

[mindburner]

some updates on the condensation issues. I had my little 4x 47 ohm resistor heater running at 12v which was enough to remove misting at 8-10 degrees. But misting of the sensor occurred below this temperature. I have to run the resistor heater at 16v. This then hits the cooling.

I loose 4 degrees in EXIF temperatures. Coldfinger shows a drop of 20 degrees but EXIF max is about 16

I realise the sensor is on thermistor is on the imaging circuit board and not directly on the sensor so I may be loosing more.

I am persevering with the resistors at the moment mainly because I don't want to have to remove the sensor again and Insulated nichrome wire is unavailable

This is the worst issue with these mods

I have an idea to pre heat the sensor chamber for about 1min before cooling engages.

I will use a 555 timer delay circuit and a relay to switch through the 4+ amps to the peltier

[JAS]

I used heat shrink tubing to insulate the nichrome wire for my dew heaters, I ran them at full power for 4 hours before christmas with no ill effect to the tubing, wouldnt that be suitable for you aplication?

Jason.

[mindburner]

hi Jason

That may be an option. Very small bore heatshrink.

The are around the sensor is just so small, it still may be tight to get a couple of circuits of wire installed.

[JAS]

Couldn't you find a guage of wire that will allow you to go around just the once? There are lots of options on ebay.

Jason.

[mindburner]

hi yes

One lap of larger diameter wire may suffice. I need to do some experiments at different voltages to get the balance right.

Another issue is that the 1100d filter frame is made of plastic, so the wire would essentially melt into it. Some sort of custom frame would need to be devised to cope.

Just another comparison I did tonight on cooled dark frames VS non cooled images. Shows real benefits, so just need to get this condensation sorted and it will hopefully be a great astro camera

[Gina]

I must get my Mk 3 version finished and see how that does.