DSLR Cooling Controller - Pier/Tripod mount

[glend (Glen)]

I have started building the controller for my DSLR cooler system. I can't possibly go to the level of integration that Rowland is achieving with his PCB system but this remote box solution should work for either a cooler box or a cold finger implementation. The plan is to mount the box on the pier in the observatory but that it will be removable to take to the dark site and attach to the tripod.

The components are pretty simple: a digital temperature controller, a PWM to be able to adjust the 12V power to the system (this one came out of a dew heater controller that Arthur made for me), a cooling fan for the PWM, connectors for the camera and battery (or AC/DC converter),and misc extras.

All of this equipment has been used during the testing of my cooler box and cold finger so I know it works, this just tidies up the whole solution. The controller and the PWM are both available on ebay (you can read the product numbers on them in the photos and search with that if you want to get them).

Photos are attached.

[pmrid (Peter)]

G'day Glen.

I'm going to watch this one closely. I have the basic bits already so I'm really going to be interested to see how you set yours up and operate it.
My temp controller is the EBay version and has outlets for both TEC and Fan (as well as the thermistor of course). Yours looks a bit more expensive. My PWM is the same as yours (the CCM6CV4 also via EBay). The idea of getting all your connectors and whatnot together in a single project box is also a good one.

Keep the pics coming please Glen. You have an audience of at least one.

Peter

[pmrid (Peter)]

Glen, before I go mad with the soldering iron, I was wondering about the PWM's role in this system.
I can understand that by reducing the voltage being supplied to the TEC you can exercise some control over the aggressiveness of the cooling. But isn't that what the Temperature Controller does anyway? Is the PWM necessarily adding anything to it? In a mobile rig, I appreciate that power saved is a good thing. But in a permanent installation, where 24o AC is available, it's not generally an issue.

And as a side issue, do you plan on placing the PWM between the temp controller and the TEC or between the 12VDC source and the temp controller. The reason for the question stems from my temp controller having a separate outlet sending power to the TEC fans. I wouldn't think it a good idea to run those fans too far below their nominal 12VDC input levels.

Peter

[glend (Glen)]

Peter there are several ways to wire this all up. Yes the temperature controller can switch (at least mine can switch up to 10A), however it is just a switch it cannot regulate - so its either on or off as per its setup. The PWM is commonly used for motor control functions so running slow or fast - so to speak, is its forte. More info here:

http://en.wikipedia.org/wiki/Pulse-width_modulation

In theory you could run your PWM through the controller relay contacts and be able to regulate when it is switched on.
Do you even need a PWM, not if you want the TEC to be running flat out all the time and drawing a continuos 6amps from your power source - maybe not a concern if your on an AC/DC converter at home. It is useful on battery power in my opinion.

Re the issue of possible differential regulation of the TEC and the fan. I see no real benefit in regulating the fan separately as it only draws 0.18A whereas the TEC needs 6 Amps and produces all the heat. It is also risky to turn the fan off or dial it down if the TEC is receiving full power - its a good way to burn out your TEC - I did that with the first TEC I bought.

I will initially set it up with the TEC powered via the PWM . I will probably run the fan off the same power feed. Initially the temperature controller will be reporting the internal camera temperature only and if it does switch anything it will be a simple Red LED to signal the threshold is crossed. I prefer the initial field testing to be done this way. I want to be able to work out the PWM setting that holds my target temperature over the length of my imaging run (and this may be different on different nights). Remember if you separate the TEC and fan power, the fan needs to work , you might want to dial down the TEC but keep the fan running hard to avoid heat buildup in the heatsink which will pull up the TEC. Rowland can probably control them separately (looking at his circuit). But I can't see any reason to switch off the fan at any time.
Keep in mind the lag time - the time it takes for a shutdown TEC to get back to your target temperature. If you do switch the TEC off then keep the fan running if you intend to restart the TEC. By experimenting you should be able to determine how long it takes for a shutdown TEC to get back into the zone - this will allow you to work out at what temp do you turn it back on. Your controller can handle the switching but leave the fan running.

Finally, reading the controller instructions, it's pretty clear that you can set it so that it will turn off when your target temp is reached and start up again when your 'd' value is reached (ie set 'C' + d = start. At 'C' it turns off again until C +d is reached again). But again, only for the TEC, I would not turn the fan off because of heat build up and its affect on recovery time.

This is just reflects my experience so far and I am by no means an expert on this subject.

[rcheshire (Rowland)]

Peter and Glen.

For this type of cooling, run the fan flat out. It reduces the thermal resistance of the heat sink for better differential.

I have stopped using PWM and now pulse the TEC at 10Hz - ON 0.1sec, OFF 0.1sec. It's the same thing as PWM but slower. Importantly, I let the last two degrees idle down at 10hz - separating any high power application away from set point to handle hysteresis and avoid large swings in cooling and conversely warming. I have found this method to be extremely accurate and reliable - couldn't believe my eyes.

I think the bottom line is this. Minimise the thermal resistance of the system for the best differential and keep large applications of power/pwm away from the set point by two degrees.

Logic is this;

if (the finger temperature is higher than set point +2C) then (drive the finger temp down aggressively)

if the finger temp is <= setpoint + 2C and >= to setpoint 0.25C) then (pulse at 10hz)

if the finger temp is < setpoint - 0.25) then (stop pulsing and set the power to 0) - TEC systems always want to warm up.

Typically, I see the finger temperature dither above and below set point by 0.5C absolute maximum. Usually, 0.3 - 0.4C.

The principle should work with any PWM or digital ON/OFF system.

[pmrid (Peter)]

Thanks Glen and Rowland.
I take Rowland's point about controlling hysteresis. With my temp controller, there seems to be a temp overshoot of about 2C when the TEC reaches its upper set point. Quite an overshoot. So perhaps as an interim measure to control that, I could wait until the temp sensor tells me it is within the range I have set. With relatively high ambient temps at the moment, and humidity that is still generally >60%, I have been setting the min/max at 6-8C. So when the temp reaches, say, 6C I would then set the PWM to drop the input to the TEC to 6-8 VDC or thereabouts and hopefully bring the hysteresis down somewhat and perhaps even stabilise the TEC temp to remain entirely within that range.

I also agree with Rowland that the fan should be kept running flat out at full power no matter what the PWM is sending to the TEC.

Thoughts?
Peter

[glend (Glen)]

I have completed my controller box and attached a few photos. Both the possible input devices are fused (the battery plug, and the AC/DC converter), and I will be putting an output side fuse on the TEC. Its hard to tell from the photo due to the small space, but I have setup the PWM +12V to run through the controller relay contacts so the controller can turn it on and off. The fan is wired directly and will be running at max all the time. The three pair shown at the camera end of the cable are the fan power, TEC power, and the temperature sensor leads. Tidy up of the camera end leads and connectors is the next step, after I finish the camera carbon fibre shoe and fan brace. Basic testing with the camera sensor attached (the controller won't work without that hooked up) showed the controller was switching the PWM on and off as I moved the control value around the ambient by the d value. The 'work' light tells you when the controller has turned on the PWM and is cooling. BTW factory setup (at least on mine) was that the relay was closed and that won't change until you program in your values.

[rcheshire (Rowland)]

Nice job Glen.

Hysteresis is probably not such a big deal with cooler boxes. The change in temperature at the sensor will be negligable.

[glend (Glen)]

Originally Posted by rcheshire

Nice job Glen.

Hysteresis is probably not such a big deal with cooler boxes. The change in temperature at the sensor will be negligable.

Thanks Rowland, mine is for a cold finger setup, although I have a cooler box as well it's not going to get used now that the finger is working. Certainly it will work for a cooler box but perhaps running the TEC through the controller relay is less benefical due to the hysteresis issue as you mentioned.

[pmrid (Peter)]

As a matter of interest for anyone planning an integrated design such as Glen's - with the Temp Controller and a PWM module boxed together, I came across this cheap digital temp controller board in a rummage through the dreaded ebay - http://www.ebay.com.au/itm/390686340888

I've ordered a couple to play with and will see how well I can integrate them with the PWMs I already have.

I'll let you know in due course.

Peter

[rcheshire (Rowland)]

Originally Posted by glend

Thanks Rowland, mine is for a cold finger setup, although I have a cooler box as well it's not going to get used now that the finger is working. Certainly it will work for a cooler box but perhaps running the TEC through the controller relay is less benefical due to the hysteresis issue as you mentioned.

Hi Glen. Yes, I was refering to Peter's cooler box.

BTW. I have posted data logged some time ago during the maximum differential test of my set up, having reduced the size and weight of the heat sink. Surprisingly, a 28C reduction vs 36C, not a significant penalty, which illustrates the dynamics of this type of cooling and the effect of excess energy supply.

The data is presented as %power, that is PWM duty cycle vs %cooling of maximum differential. Experience tends to indicate similar performance across these systems. Notably 75% of cooling at 50% pwm duty cycle. The remain 25% requires considerably more effort and time (not logged). The start point is 6C reduction without power, give or take a degree or two or 3.