I managed to get my 450D (in side a peltier cooler box) to 0C~-4C but it takes nearly 3/4hr to 1.2hrs to reach this. Thus while my box works it's a real pain in running it and is a little ungainly attached to a focuser. I also had thermal losses thru the focuser drawtube (had to insulate it)
Thus I ask, what time period does it take for you to obtain for full temp drop? I'd expect much faster than my box and thus I'd be more than interested it all this.
Also do you use the Arduino as PWM temp feedback control or just run the TEC flat out and measure the temp?
Are you getting thermal losses thru the camera case or drawtube?
Sorry for all the questions...just interested I am!
Cooling down the full 34 C takes about ~10 minutes and it stays there. Temp drops 10 degrees from ambient in ~20-30seconds. To zero in ~2 minutes and so on.
Major heat losses are from the copper heat sink mounting plate, despite insulation (replacing with 3mm acrylic) and the distance between the TEC and the sensor - this is insulated with double sided tape where possible, otherwise duct tape where space is restricted (which doesn't do much except keep condensation at bay).
I looked at several mods including cold boxes and decided that most of the energy is going into keeping the box and camera cold, instead of cooling the sensor. Even the cold finger mods waste energy with non-essential conductive couplings. So I figured that isolating the cold finger as much as possible should produce the best results and minimize losses through the camera body, draw tubes etc.
At the moment I have tested 3 times at max deltaT - no control, and once using PWM. I need to improve on the program for better stability. It held quite well at -5C (139, 0 - 1023), but not so well as I would like.
Typically, the accuracy of the TMP 36 is ~0.3C, but I am seeing steady temps with .01C tolerance, for long periods with the occasional + or - spike here and there. I think that's electronic noise as the thermal mass is fairly stable.
BTW. The camera is still working. Happy to provide details. It's a fiddly mod. The advantage is that all the camera controls are accessible and functioning, except for the Piezo sensor cleaning - manual only.
I'm currently trying to fix a Losmandy Gemini for David (Brundah1) and I'm also just about finished a very nice PIC stepper motor focuser control (with ASCOM control, temp control and focuser wheel re-positioning control) by Dave Trewren. More here.
Just about finished this project...have some way to go on the Gemini...but am always open to "new projects"!
But If I could get lets say....-2C (exif)....with PWM feedback temp control wow! Now I'd buy that!
Happy to provide details to a fellow DIY. It's not a trivial mod. -2 +/- 0.3 is realistic, but on average, it does a little better. I suspect that sensing error is greater than actual deviation.
Hi Brendan. Keep me posted. PWM is working but PID is better. By Feb temp control should be automatic using the Arduino, and changes in ambient temperature wont affect the preset target temp.
I'm using two TMP36 temperature sensors at the moment, one reads ambient air temperature and the other cold finger temperature. PWM values are calculated as a function of the ambient and set point temperature delta.
Testing shows the need to improve on the code. Still getting my head around the Arduino PID library, which I am told will produce more accurate results. In any case PWM is still required.
An easier method is to calculate the desired PWM value and set that in place of a target temperature, eliminating the need for two sensors, with which there is a lot of noise due to read errors of both sensors with PWM chasing set point temperature. The down side to this is that as ambient temperature changes so does cold finger temperature - PWM values range from 0-255, with 0 being the ambient state and 255 max cooling. So at an ambient temperature of +25C and 0C, PWM = 0 (120 approximating a drop of 24C - with this system).
Overall I think calculating the mean based on the thermal inertia of the cold finger is probably more representative of the actual temperature (that is, temperature change is not rapid when removing power from the Peltier, except at very low temperatures operating near the maximum differential of the system - decrease in cold finger temperature is slower due to increasing power requirements).
I have a few ideas for improving the code and smoothing things out, then again I might just average all the readings and print the result - doesn't look as bad.
EDIT: Analog sensors are subject to interference. A 0.1uf capacitor strung across Vin and GND may stabilize the readings, with the capacitor located as close to the input pin as possible. So I guess straddling the pins is as close as you can get. That will be impossible for the cold finger sensor as it is well and truly embedded now.
The temperature control system is fairly easy to build. Peltier and power supply, a Logic Level N-Channel MOSFET very low Rds(on). The -ve side of the Peltier to FET Drain. PWM signal from an Arduino PWM pin to FET Gate and the FET Source to GND, controlling current to the Peltier. To avoid overheating of the FET you need the lowest Rds(on) that you can get.
Then again sourcing the right FET and getting them to Aus at reasonable shipping costs can be a headache. Love to hear of an Australian supplier.
Choosing a heat sink was tricky, in that it needs to be tuned to the system. Basically, not so big as to be cumbersome and big enough to extract as much as heat from the Peltier minimizing the hot / cold sides delta.
Theoretically, I think the system is capable of a max differential of 46C, but that is lost in inefficiencies over which I have no control. 34C is pretty good and provides flexibility and control. Most of the time the system will run at half that or less due to high dew point temperatures.
I used TMP36 sensors because they are inexpensive, linear and quite accurate without calibration. Comparing temperature readings with other electronic thermometers in the house and the local weather readings both sensors are well within the ball by 1/2 a degree or so.
Next task is to vent air from the cooling fan filtered and dried to the face of the low pass sensor, much like car defogging, to overcome high dew point temperatures. 16C at the moment, so there is little point in using cooling at all.
Addressed the points below and tidied up the cabling. Added some cooling protection to the MOSFET - not absolutely necessary - happier with it for the time being.
If interested in details PM or email. This is not a trivial mod. Still, I've had it running at -13C saturated with no affect on operation. If the weather is kind I hope to take some shots of M42.
The Arduino also schedules image acquisition and dithering, on what is an unguided set up. The Tak mount is very accurate up to 4 or 5 minutes with careful alignment. 15 minutes is the record without appreciable drift. Arduino does a calibration run once the illuminated lens cap is fitted. Pretty much set and forget.
With temperature control calibration should be fairly consistent.
Anyway, the proof is in the pudding.
Last edited by rcheshire; 28-12-2011 at 03:58 PM.
Reason: a bit more detail
I would be very interested in doing this to my 450D...it's getting on a bit and new DSLRs are getting cheaper so if I kill it...no great catastrophe. I will go to a cooled CCD, someday, but I do find a DSLR an excellent all round camera. They are very clean noise wise with low sensor temps.
I reckon this would be a very worthwhile and rewarding project....PM sent.
For anyone following this thread, this project draws to a close. And none too soon.
I realise its not everybody's cup of tea - it's DIY and satisfying as its own end.
Cost - conservatively, including all components, not including the FET's I fried, because they were the wrong parts ~180AUD, which included the power supply. If you already have one, then less than a 100AUD should do it. The biggest investment, because it was a first of, is time.
All is working as planned, with a couple of improvements outstanding.
Good points - cooling can be controlled very well within a practical range of 30C below ambient, 34 if need be. It will cool further, but progress is slow and control is difficult because the system is at the edge of its performance envelope.
Cooling is fast at 100% pwm - seconds or minutes depending on the desired setting. The bigger the differential the longer the cooling. Alternatively, setting the pwm variable, for the camera temperature selected, slows the rate of cooling, which may be better if thermal stress is an issue - though I don't think it is - this is my preferred method, as it is much smoother.
Condensation - noticeably less when using live view. Dew control is a WIP and I'm looking at a chemical solution. Used sensibly, cooling to above dew point presents no problems. My view is, that because the cold finger is in direct contact with the back of the sensor, sensor temperature is not greatly affected by heating during long exposures - however, I have no way of testing this.
I did attempt a first session and took some frames, including calibration. With 10 - 15C reduction from ambient noise is significantly less.
Dissapointments - It was necessary to employ a second Arduino board because cooling became unstable when multitasking - a consequence of my limited programming skills.
If anyone is writing a no delay shutter sketch with mirror lock-up, perhaps using the MsTimer2 library, I would like to hear about it. In any case the stepper that I use to control the GEM's hand controller pretty much stops everything else while it does its little 50 step dance.
Power supplies - Linear is better. Switch mode has been a pain with interference, requiring a fair bit of ripple smoothing to make it usable.
The heatsink is not mounted square and detracts from the appearance, in my view. Broke the golden rule on this occasion - "measure twice, cut once".
Was it worth it. Well I have to say yes. Truthfully, this project involved a very steep learning curve and I'm surprised by the results - better than expected, though it would be nice to build the electronics into a more professional package - time.
Images here. Managed to delete the previous album.
Added push button temperature control. Preset and change on the fly. An LCD would be nice. An LED provides on temperature indication and maximum differential warning (ambient -30C).
EDIT: further software improvements. Two button push registers setpoint with ambient temperature at any time. Saves keying all the way through to max and ensures setting is accurately referenced to ambient.
Having reached a point where further improvements amount to tidying up the prototype and making something professional of it, and further minimizing EMI in the system, I had the opportunity to test the cooling mod from dusk to dawn at Snake Valley a week ago.
With spectacular skies, the cooling system ran for several hours at a time, in total approx 9 hours imaging different objects - I'm happy to report, flawlessly.
An old travel alarm clock, with a digital thermometer, was used as the ambient temperature reference. The same can be done with a laptop displaying the serial output from the Arduino board - more complex and a bigger hardware footprint - the idea is to keep it compact.
With an ambient temperature of 20C, target camera sensor temperature was set at 5C, but this caused the face of the low pass filter to fog up. Raising the setting to 7C provided fog free imaging all night.
(As mentioned in an earlier diatribe, defogging of the low pass filter face is a work in progress. Reassembly of the sensor/filter array during the filter mod and subsequent sealing with silicon in dry conditions, prevented the possibility of condensation inside the senor/filter array).
An LED indicates "on temperature". It flickered at high frequency in response to read errors all night. Because the system is slow to react thermally, to changes in temperature readings, it is safe to say that the actual temperature was stable within +/- 1C.
While the system is capable of a 34C differential, all that's required is to maintain temperature above dew point and minimize dark current. The advantage of a cold finger is that the sensor is kept at or very near the desired temperature; that is, any tendency of the sensor to heat up is absorbed quickly by the cold finger. The power requirements are also very low.
I set the temp at ambient -20 (0C) and let the system stabilise. When I removed power from the Peltier, temperature at the sensor remained within <1C (0.96) of 0C for 13 seconds before increasing. As far as thermal inertia goes, the system seems very stable and resistant to rapid and excessive changes in temperature.
I thought that I'd round off by saying that this project was worthwhile if only for it's educational value. The camera cooling system has since been dismantled, the camera restored to its original condition and given to a relative, who is enjoying their first entry level DSLR - a worthy cause.
If you really want a hardcore DSLR cooling experience then I suggest heading over to Stargazers Lounge and checking out Gina and YesYes efforts with a Canon 1100. I just don't have time to continue with the project, although it would have been nice to get the sensor defogging working.
This modification was always limited to 2 degrees above dew point to be certain of a clear sensor. Works great in a dry climate summer night. Cooling down 25C or so quickly and maintaining temperature. The winter advantage was prevention of sensor heating.
One of the bug bears of this project was the dreaded low pass / anti-aliasing filter, a feature of DSLR's. It is better removed for sharp images, however, I used it, unwittingly, to seal the modified sensor assembly from moisture, but in the end, didn't have the desire to pull it all apart and replace it with coated glass or complete the messy job of sealing the astrodon filter / sensor face.
Soft images were always going to be a problem with this set up. Anyway it was great fun and a superb learning experience in areas that previously were unknown.
I am posting the code, including the oddball camera and dithering control code, which sequences the entire imaging session
The dithering mechanism has been converted to push button control - there doesn't seem to be a simple way of automatically sequencing it with the CCD camera - need to find a creative solution.
Some of this stuff might be useful with modification for the hardcore DIY'ers.
