A hot sensor is a noisy sensor
When acquiring any exposure, regardless of the duration of the exposure, the sensor will generate some noise. Some of that noise is electronic randomness which can be eliminated by subtracting bias frames. Some noise is actually signal generated by the chip. Dark frames are used to remove this unwanted signal. I immediately noticed how much of this thermal signal was present when I first started my deep sky experiments with the ASI174MM. I’d like to say that this is my original idea, but I actually saw people doing this hack with DSLR and other cameras. It was clear that I would be able to quickly and inexpensively come up with my own cooled ASI174MM camera hack.
Building a cooled ASI174MM
The basic materials for this hack are simple: I need a thermal electric cooler, ‘TEC’ for short. A power source, and a way to remove heat from the hot side of the TEC. Having some way to affix my heat sink, fan, and TEC to the ASI174MM was a little challenging, but thanks to a thread online, I saw someone basically already do what I had planned on. I’m not ashamed to admit that basically ripped their idea off. I found the rubber o-rings at a local family owned hardware store.
The hot side of the TEC has a thin layer of Antec Formula 7 Nano Diamond Thermal Compound which helps transfer the heat into the sink. This is a small, but important part of the cooling system as it increases the efficiency of the heat sink. The heat that I can extract from the hot side of the TEC influences how cold I can make the cooler side. When sandwiched together and held down by the o-rings, the construction is fairly simple, though it may not look very pretty. To help my cooled ASI174MM be a little more efficient, I added a layer of neoprene to insulate the camera from the warm outside air.
The positive and negative leads of my TEC are connected to a 12v DC cigar style plug. It is powered through Frankenstein. I wired in a switch so I can turn it on and off without disconnecting it from the battery.
For the initial tests, I used FireCapture because, despite it’s lack of certain features, I can read the sensor temperature after each exposure. My home office was allowed to warm up to about 80F for this test, not unlike a warm summer night in Dallas, TX. I set FireCapture up to take 5 images, each 180 seconds long. During this test, the TEC was inactive, and the camera was not cooled. After the 5 ‘warm’ exposures, I turned on the TEC. After five minutes, the cooled ASI174MM was then used to acquire 5 ‘cool’ exposures just like before.
I discarded the first and last image from each run so I could focus on just the middle of each experiment. As expected, the cooled ASI174MM tests showed a large reduction in thermal signal. The results are very compelling, and also very promising. The warmed ASI174MM had an average temperature of 107.6 degrees F. The cooled ASI174MM had an average temperature of 65.3 degrees F. That’s a 42.3 degree reduction in temperature, and it shows a huge reduction in thermal signal buildup in the cooled camera.
The only problem I had left to overcome was the amp glow. Despite the cooling efforts, the ASI174MM still showed a significant signal coming from the electronics around the chip.
Voiding my warranty
There is no real benefit to having a cooled camera if there is still a large amount of unwanted signal building up on the sensor. In this case, I’ve removed a significant amount of thermal signal in the cooled ASI174MM, but the amp glow was still present. I wanted to find a way to help remove it as well. My hypothesis was there must be some electrical component very close to one edge of the sensor. That component was somehow polluting the chip, and that pollution was building up as amp glow. The easiest way to test this theory was to have a look inside. This meant I needed to open the camera, thus voiding the warranty on the ASI174MM.
On the underside of the ASI174MM sensor board showed a large number of capacitors. Capacitors, typically, don’t generate nearly the amount of heat as resistors. I cannot help to notice, however, that there are a great number of capacitors in use, and they correspond to the same shape as the amp glow observed. My first thought was that there was sufficient heat coming from these capacitors, and if I had a heat spreader then I might be able to fix the amp glow. I’m going to warn you now that what I am about to describe should not be attempted by anyone. Ever!
Almost destroying my camera
I made my own heat spreader by cutting an aluminum strip from a soda can. To protect the electronics that made contact with bare aluminum, I coated one side with clear nail polish. I then bent, and trimmed the aluminum so that it could press against the capacitors in such a way that the heat they were generating would transfer through the aluminum, into the housing of the cooled ASI174MM. At least, that was the idea. In reality, nothing changed. I believe that while I was able to place a resistant layer of nail polish between the electrical components and the aluminum, I also successfully thermally insulated the aluminum, too. In fact, if something is a good electrical insulator, it’s also a good thermal insulator. This exercise came with considerable risk to my camera, however, as if I had not correctly insulated the aluminum, I ran a huge risk of shorting a circuit, and frying the camera.
Conclusion of the experiment
I’ve concluded that the ASI174MM can benefit greatly from active cooling. Using the TEC is a simple way to implement this cooling. I’ve also found that at exposure times that are typical during deep sky astrophotography, a cooled ASI174MM will still have considerable amp glow that needs to be addressed in another way. Firstly, dithering the exposures by a significant amount will help – which is something that FireCapture does not allow at this time. Finally, removing the amp glow may be possible with judicious use of dark frames to properly calibrate for the signal. The dark frames will also assist the removal of any noise generated by the thermal signal that the cooled ASI174MM camera still contains.
New drivers being produced by ZW Optical are showing a lot of promise. I’ve been beta testing these drivers for about a month as of this writing and they have greatly reduced the amount of amp glow. Some evidence shows that the camera is actually running cooler when taking long exposures. If this turns out to be true, then this cooling modification will actually be more effective.
It still remains to be tested in the field but I do have high hopes for the cooled ASI174MM as a viable entry level deep sky imaging camera.