1100D cold finger sensor cooling with TEC and water cooling

[yesyes]

Yes, some pics would be helpful....

[nightvision]

No photos as I had just finished making 2 scopes from my bits box and decided to take apart the cooler at the end of the day to give me something to think about overnight.

I don't think I'll be using the big heatsinks (but I will use the solid block) as they are too heavy and I shouldn't need that much cooling, but it is interesting to see how they do it.

[Ptarmigan]

Need to figure out how to convert times in hh-mm-ss format to seconds Also, I can't seem to find any way of making the x-axis spacing according to the time - it just spaces it with equal divisions for each data point. Maybe I need to find another spreadsheet software.

Hi Gina,

have you had a play with gnuplot, very powerful graphing freeware but very geekish learning curve to begin with ( as are a lot of things that start out on Linux !! )

http://www.gnuplot.info/

[rcheshire]

Wow Gina great result. My mod gets down to 34C with a big heatsink. Condensation and icing on the face of the filter glass is a limitation. I sealed the assembly with silicon to prevent condensation inside.

I never cool below dewpoint +2C when imaging. Sometimes that's only 5 -7C. But with nighttime ambient temperatures of 20 - 25C the reduction in dark current is reduced 3 fold.

This logic level MOSFET runs off a 5v Arduino pin very well (NXP PSMN1R6-30PL very low rds (on) 0.0014ohms) - no heatsink required.

[Gina]

Hi Gina,

have you had a play with gnuplot, very powerful graphing freeware but very geekish learning curve to begin with ( as are a lot of things that start out on Linux !! )

http://www.gnuplot.info/

Yes, I'm familiar with Gnuplot used it for my weather station software. (Which reminds me, I must get the Linux box I run that on working again.) Sometime I plan to get back into programming again - apart from astro stuff I have more to do on my weather station software.

[Gina]

Wow Gina great result. My mod gets down to 34C with a big heatsink. Condensation and icing on the face of the filter glass is a limitation. I sealed the assembly with silicon to prevent condensation inside.

Thank you I sealed the case of my camera with silicon grease and put two 3g silica gel bags inside but still had condensation. I don't have a filter on the sensor as I did a full spectrum mod and use a CLS-CCD clip filter.

I never cool below dewpoint +2C when imaging. Sometimes that's only 5 -7C. But with nighttime ambient temperatures of 20 - 25C the reduction in dark current is reduced 3 fold.

Yes, that's certainly one way. I'll be able to do that once I have my control circuit finished and working.

This logic level MOSFET runs off a 5v Arduino pin very well (NXP PSMN1R6-30PL very low rds (on) 0.0014ohms) - no heatsink required.

That's a much lower rds(on) than mine - I'm thinking of getting a couple of logic level MOSFETs as one Peltier TEC runs on 5v. I read somewhere that logic level ones are better for lower ds voltages. That's a different part number from Chris's - I'll check the on resistance of that one.

[Gina]

I now have the same type of silica gel that Chris used. Much bigger bags with the colour change property. They look much better I also have most of the components but waiting for the box which will probably come at the end of the week. All these bank holidays get in the way of progress

[Gina]

This logic level MOSFET runs off a 5v Arduino pin very well (NXP PSMN1R6-30PL very low rds (on) 0.0014ohms) - no heatsink required.

IRL540N Rds(on) Max is specified as 0.044 ohms but they don't give a typical value. At the max value and 5A the dissipation would be 1.1W but I'm sure the typical value is much less. Where did you get the NXP PSMN1R6-30PL?

[rcheshire]

http://uk.farnell.com/nxp/psmn1r6-30pl/mosfet-n-ch-30v-100a-to-220ab/dp/1845637

I bought mine from the above, down under.

Suggest making sure cooling is off before uploading new code to Arduino - that is, on the fly. I broke several that way.

[rcheshire]

If it helps, the MOSFET gate pull down resistor value on my set up is 750R. 1k will probably do. I used a variable resistor to set the value and then replaced with a fixed value. You will probably find that the gate voltage will not get low enough to switch the FET without the correct resistance gate to GND. I posted a circuit diagram showing this earlier. My system humms along with temp changes on the fly. However, if you are using an Arduino, with delay() statements of a second or more, you may find that it messes with FET switching. Short delays of 300 - 400 milliseconds are OK. I found that camera and dither control messed with cooling - hence a separate board.

[Gina]

http://uk.farnell.co...20ab/dp/1845637

I bought mine from the above, down under.

Suggest making sure cooling is off before uploading new code to Arduino - that is, on the fly. I broke several that way.

Thanks for the link I'm not using an Arduino - I'm driving the MOSFETs from either a 555 timer or a USB interface card which provide PWM output with open collector. So it needs a pull-up resistor to pull the output voltage up when the o/p transistor is off. When on the output will be well under a volt with a saturated transistor.

[Gina]

If it helps, the MOSFET gate pull down resistor value on my set up is 750R. 1k will probably do. I used a variable resistor to set the value and then replaced with a fixed value. You will probably find that the gate voltage will not get low enough to switch the FET without the correct resistance gate to GND. I posted a circuit diagram showing this earlier. My system humms along with temp changes on the fly. However, if you are using an Arduino, with delay() statements of a second or more, you may find that it messes with FET switching. Short delays of 300 - 400 milliseconds are OK. I found that camera and dither control messed with cooling - hence a separate board.

As I said I'm using active pull down and resistor pull up and not using an Arduino.

[rcheshire]

Understood, it was mainly for other readers.

[Gina]

Understood, it was mainly for other readers.

OK fine Thank you

[nightvision]

Hi Gina, just wanted to apologise, I was fumbling about in this new dark and dingy forum environment and accidentally gave this excellent thread 2 stars! It is of course a 5 star thread (probably one of the best). Tony.

[Gina]

Hi Gina, just wanted to apologise, I was fumbling about in this new dark and dingy forum environment and accidentally gave this excellent thread 2 stars! It is of course a 5 star thread (probably one of the best). Tony.

AQpology accepted Don't worry about it It is different isn't it!! I'm gradually getting used to it - not sure whether I like it or not yet

[Gina]

Got the black ABS plastic box for the camera now so I'll be fitting the camera and other bits in shortly I'll post photos.

[Gina]

Some holes cut in box but more to do when I sort out just where things are going.

I've very nearly finished the control circuit board and tested it so far - just got to add the over-voltage crowbar components for the camera power. The regulator is giving 7.89v compared with the designed 7.6v - this is within tolerance. The PWM control I tested by connecting the output to my DMM - it changed smoothly from zero to 12v as I turned the pot - and I got it the right way round So that's working correctly. I haven't yet tried the MOSFETs under load.

I have had the camera connected. I checked it in Bulb mode, holding the shutter open and the steady current was 160mA. Peak current reached just over 1A when operating the shutter and mirror - standby current was 40mA. Output voltage didn't change (within the 10mV accuracy of the DMM) over that current range. The LM317T with small heat sink got barely warm.

Here are some photos...

1. The front of the box showing the large hole for the focal reducer adapter.
   
2. Control circuit board.
   
3. Camera from front showing draught excluder round FR adapter ring (needs a little gap closing) and on battery cover (to hold it closed in the box).
   
4. Camera in box held by tripod bush and screw, showing back of camera with cooling system.
   

[Gina]

More photos of the camera box and cooling system. The back and front of box. The back (top of box) detaches to gain access to camera and other parts. The removable back has a digital thermometer showing the cold finger temperature. The front (bottom of box) has a hole for the adapter ring into which the focal reducer screws and connectors for power (Gnd, +5v and +12v) and data (1-wire network and PWM inputs).

[nightvision]

A brilliant project Gina, that has really delivered.

I have just completed testing of several TECs and heat-sink arrangements. Managed to cool a 150x150x150 (internal size) PPA (tough structural insulating foam) box from 17c to -6c in 20 mins but found the heats-sinks and fans needed to be too big, heavy and noisy. So it looks like I will be following you down the water cooled route.

How heavy is your complete unit/camera package? I haven't managed to find a cooling block as small as yours but I have found these; the copper Zalman is 154g mm 53(W) x 63(L) x 40(H) and the blue ally block 100g :68 (L) x 78(W) x 8(H) MM, do you think they will be effective?

[Gina]

A brilliant project Gina, that has really delivered.

Thank you

I have just completed testing of several TECs and heat-sink arrangements. Managed to cool a 150x150x150 (internal size) PPA (tough structural insulating foam) box from 17c to -6c in 20 mins but found the heats-sinks and fans needed to be too big, heavy and noisy. So it looks like I will be following you down the water cooled route.

That's quite a big box so that's not bad at all. But yes I found I needed a lot of cooling for an air cooling system. The water cooling is very effective - Thermaltake water block, 160mm radiator and fan, low power aquarium circulating pump and plastic tubing. It cuts the weight on the focuser way down.

How heavy is your complete unit/camera package? I haven't managed to find a cooling block as small as yours but I have found these; the copper Zalman is 154g mm 53(W) x 63(L) x 40(H) and the blue ally block 100g :68 (L) x 78(W) x 8(H) MM, do you think they will be effective?

Just weighed my box complete with camera and all the bits and it's 875g. Standard 1100D with battery weighs 506g. I have a Zalman water block and wasn't very impressed - the circular cooling area did not cover my 40mm square TEC. The water block I'm using weighs about 90g as I recall and just nicely fits the larger (40mm) TEC (actually just slightly smaller but it covers the part that gets hot).

Just checked - this is the water block :- http://www.ebay.co.u...984.m1439.l2649 One left it seems. It's 78g it says. I think that blue one should be OK.

This is the ABS box :- http://www.amazon.co...ils_o02_s00_i00

[nightvision]

Thanks for the links Gina, off to do some shopping....

[Gina]

Box pretty much completed and been testing powered off my bench supply. Camera works fine but I need more cooling for the LM317T voltage regulator - the tiny heat sink I'm using gets too hot with long exposures. The manual PWM control is working for the 12v TEC - with only one supply indoors ATM I can test only the 12v one. The MOSFET doesn't seem to get warm though I can only run a proper test with the water cooling working otherwise the water block gets too hot. But it's looking good for the next stage of testing in situ (once I've arranged better cooling for the LM317T).

A couple more photos.

1. The insides.
   
2. The top showing the PWM switch and manual control knob.
   

[RogerTheDodger]

That's really coming together now Gina, I can't wait to see some very long images from it.

[Gina]

Thank you Shouldn't be too long now to get it finished. I can test it on terrestrial views as long as we don't get rain with strong easterly wind. Then some decent weather would be nice We haven't had a clear night for about a couple of months now and thw next week doesn't look any better