Suddenly after no problems with star shapes, the next night out they were just terrible!
Below I show a gif comparing two 2 hr stacks, one with the good stars, and the next with elongated stars. In this example the distortion was vertical. But later in the night, the elongation direction changed. The difference was huge, my bad stack averaged a FWHM of 7.4 pixels, 3.7 arc-sec, compared to what I got before with round stars of 5.2 pixels, or 2.6 arc-sec. I was gutted by this sudden and mysterious issue.
I have a C925 with a Starizona f6.3 reducer, a CEM70ec mount which has encoders, so I was not guiding for my 2 minute exposures. My camera is a ZWO ASI 2600MC Pro, so has fan cooling, which I had turned on. I monitor the drift from frame to frame, but that was less than 0.5 px on average, or 0.25 arc-sec, clearly not the issue.
📷 bad-vs-good-comparison.gif
At first, I thought it was the wind. But as the wind died down that night the elongation was if anything getting worse.
I went to a bright star to look at my optics, I noticed for in live view that with 2 ms frames the star was doubled!! What could cause it to double??? Then at 20 ms exposure the star was elongated, but no longer doubled.
I realized then I had a sinusoidal vibration, likely from the fan!
Sinusoidal vibrations are interesting. As shown in the figure below, the amplitude increases rapidly and almost linearly, then around the peak hardly changes at all in amplitude, before plunging down—again almost linearly, then again nearly stopping any change in amplitude near the minimium.
📷 image.png
Thus, if we are observing a sinusoidal vibration at a slower frequency than the vibration (a longer exposure), we just see a blurred average of the position change. But if we sample faster than the vibration frequency, then when the cycle is on a rise or a fall, the star will be in a different place in each of those sequential frames in live view mode, so to the eye, they blur into invisibility except. But on the flat part of the cycle, near the peak maximum or minimum, the position of the star is hardly changing, so we see the minimum and the maximum as a star—thus a doubled star separated by the amplitude distance. Thus, sampling abit faster than the vibration frequency, we will see the star at top of the peak and the bottom of the peak, and on average we don’t see much of the rise and fall. If you use extremely short exposures, very high frequency, in live view the star seems to jump around a lot.
In this case at 2 ms, or 1/500 of a second, I basically only saw the maximum and the minimum preferentially, a doublet! At 20 ms, or 1/50 s, everything was averaged out, so I just saw a broadened smear. The fundamental fan frequency for ZWO is apparently something like 100 Hz, a period of 1/100 s, so it all suddenly made sense.
I have a ZWO 2600MC for years now, never a problem, but now it appears the fan vibration is suddenly a problem. Turning the cooling off and on (there is no control over the fan itself), showed me going from 4 or 5 pixel stars with cooling off in my 2 minute exposures, to 7 or even 9 pixel stars with the fan on, so the fan was clearly the issue.
I cleaned the fan, which was after years was incredibly dirty, but to no effect, vibration was still there.
I bought a new ZWO replacement fan, which did not do the trick at all, no improvement.
I removed the screws holding the fan to the back, and isolated the fan from the camera with silicone rubber posts, with no improvement. I have no idea suddenly why I’m sensitive to this fan vibration, even with a new ZWO fan.
Fortunately, there are people who have posted on-line a solution: there are two fans that are recommended for low noise and thus low vibration: Noctua (for 40mm ZWO fans) or Fractal Design (for 50mm ZWO fans). For the 2600 I needed the 50 nm fan, the Fractal Design. The only issue is these fans are that they have a three-prong connector, with a speed control wire, while, ZWO has only the positive and ground wire in a two-prong connector, as ZWO, in this camera at least, doesn’t use a speed control wire.
Anyway, easy enough to splice the ZWO connector onto the Fractal Design wires, as well shortening the long wires that came with the new fan. Here is a picture of the pins, make sure you connect them correctly if you need to do this. The top is a schematic and the bottom a photo of the connector from my new Fractal Design. The label 1 is the ground, the middle is red, 12V, and the bottom (2510) is the control, which we don’t use in my camera, so it is just closed off, unused.
📷 image.png
📷 image.png
📷 image.png
For the Noctua fan, it has an adapter, so you can use that adapter to do the two-pin connection for cameras that fit this fan.
Also make sure you replace the fan with the correct airflow, which is inward.
Finally, to be sure I did everything I could, I added the silicon rubber posts, and bought more of the rubber o-ring spacers, all available from Amazon.
📷 image.png
📷 image.png
Below is an example (note I forgot to get an image before reassembling, so the camera backplate is missing in my photo, but I had an extra back screen, so imagine the back plate is there to the left of the screen, then I put two of the rubber O-rings to isolate the fan from the back plate and screen. All four rubber posts were done the same way. On the inside (and outside), cut off what rubber post you don’t need. The fan seems very secure, but one can add a couple of O-rings on the post on the left (inside) of the fan, for a more positive hold on the fan
It works like a charm, no vibration, images are the same with and without cooling on! It has been working flawlessly to date, about a month now no issues.
This seems to be a particular issue with the ZWO fan, but if your fan gets very dirty or damaged, any fan could start to vibrate.
I do advise everyone to check the effect of the fan often, I worry now—and there is nothing I can do about this—that the fan may have been broadening my images for a while, just too subtle to tell compared to the effects of seeing, wind and polar alignment. I’m suspicious because it seems my “seeing” is suddenly better most nights than I had been generally in the past. Water under the bridge now, but I won’t be caught again.
Speaking of water, QHY does have a liquid cooling system to avoid vibrations, but not one has to contact them for a price, which usually means it is very pricey indeed…
From the internet ZWO seems to be most problematic for their fan choice, though QHY says all fans cause some problems, but then they might just be trying to sell some liquid cooled systems?
If you have a ZWO camera (and maybe any fan cooled camera) I suggest you check at least your FWHM with and without the cooling, make sure there is no effect during your subexposure.
What I do now every night in Live view mode is to turn cooling on and off and look at my star shape and size and movement when I am doing my sky alignment on a bright star. It only takes a few seconds to make sure all is well. With short exposures and a big zoom one can see the star wiggling around a lot more when there is vibration with cooling on, compared to no vibration with it off.
