Resolved SBIG 16803 - image train rigidity

Continued...


Star Roundness Trial 1: New Fan ON....2nd repeat, different area of the sky



Star Roundness Trial 2: New Fan OFF



Star Roundness Trial 2: New Fan OFF......repeat, a different area of the sky






Discussion:

The Eccentricity figures are fairly well backed up by a visual observation of the stars. There is some evidence of field curvature and this is to be expected given that Celestron only guarantee star shape for an image circle diameter of 40mm, where the SBIG chip has a 52mm diagonal. With the new Fan ON, the average Eccentricity is in the range 0.51-0.58. This is a large improvement on the Eccentricity figures reported wth the old Fan, which were in the range 0.65 - 0.85 and were at the worse end of the scale in recent trials, suggesting a degradation of performance.
The stars at the centre of the images where the Fan is ON show a small but perceptible amount of elongation.

The Eccentricity for the Fan OFF trials were in the range 0.40 - 0.47. The stars at the centre of the images where the Fan is OFF are virtually perfectly round.

Conclusion

The previous fan arrangement , whether the fan itself and/or the vibration insulating foam, resulted in severe vibration causing substantial elongation of the stars in images captured with the SBIG 16803. Fan OFF trials showed this elongation was virtually completely eliminated.

The new Fan and/ or insulating foam arrangement gives a substantially improved performance but it is by no means perfect and there is a demonstrable difference to the fan OFF images in terms of star roundness.

 

SBIG 16803 Fan images, before and after replacement

Previous Arrangement:

The insulating foam pieces do not cover the entirety of the side of the fan.



The foam on one of the edges was quite badly damaged and looks like the adhesive side had been ripped back on itself during insertion




New Arrangement:

Foam covers the entire side of the



Foam carefully inserted and snug in the holder.

 

A comparison of magnified images of the stars at the centre of the FOV for Fan On vs Fan OFF

Fan ON left, Fan OFF right



These two images clearly show that star elongation due to fan vibration is still occurring with the new fan and replacement vibration dampening foam, although the amplitude of the vibration has clearly been significantly reduced. The Fan OFF images have almost perfectly round stars.

These figures and those derived previously suggest that with the old fan/ vibration dampening foam configuration, that the amplitude of the vibration was generally such that the stars were being elongated by 60% - 100%, dependent on the FWHM and that the more recent trials were on the worse end of this range.

For the new fan/ vibration dampening foam configuration, the amplitude of the vibration is generally such that the stars were being elongated by 15% - 23%.

With the fan off the 9%-13%. The centre of the image is more like 5%-10%

Bearing in mind these are average figures and include an amount of field curvature from the corners of the images, and that the Eccentricities at the centre of the images are less, this broadly suggests the new Fan is responsible for a star elongation of 5-10%.

Discussion
The amount of elongation seen in the images when the fan is ON, gives a perceptible elongation to the stars and of course will reduce the resolution of any nebulae or other non-stellar features in the image.

Is there a a specification for performance in this area? Is this performance comparable with that from other SBIG 16803 cameras in use around the world? What about for SBIG 16803 cameras hooked up to C14 Edge HD OTAs? Is there any interaction, which was mentioned as a possibility by Doug, between the camera and the OTA?

How can I move forward from here to achieve an acceptable performance in terms of star roundness?

Do I need to look to water cooling as an option where the fan which is cooling the circulating water is physically remote from the camera and particularly the sensor?

 

Thanks Doug,

Would you say what I am seeing is abnormally bad? The Mount is a Paramount MX+, so pretty much a top end Mount. I am not aware of any issues with its performance. I doubt the Mount is at issue. The Camera is bolted to the adaptor which is relatively short and screwed to the baffle lock nut. I doubt there is a more rigid imaging train. That leaves the OTA.

Are there option for an ultra low vibration fan to replace the existing one? What do you recommend for the water cooled option?

Thanks,
Niall

 

Guys, I still do not have a solution to producing images without star elongation due to the vibration of the fan. I understood you were going to send me a rig for water cooling to allow me to achieve the necessary cooling without having the fan mounted to the camera directly. I have been dealing with my supplier who tells me that he still doesn't know when this rig is going to be shipped. I have sent a PM to Colin and received no response.
This forum post is marked resolved, which as far as I am concerned it is not.
Can I please have some feedback as to where things are at, as after 1 year I still have not produced a satisfactory image from this camera.
If you are unable to help me with the water cooled option, could I at least ask that you support me with the identification of a suitable ultra low vibration fan to replace the standard one.

 

Thanks to you both for keeping me apprised. I understand the difficult situation with the pandemic. I am also going to buy a 12V ultra low noise/ vibration 60mm fan by Noctua as a fall back. It is only 45AUD. It only has half the volume throughput of the existing fan but should work OK during the cooler months. You may recall Andrei Ioda said he solved his problem with such a fan. The liquid cooling as a supplementary system will allow me to achieve temperature in the summer months.
There are a number of forum posts referring to deep sky imaging with C14s where the focus is done at the primary mirror, mainly because of the lack of back focus space. There is some discussion that mirror flop can occasionally cause the loss of sub where the auto-guiding can't keep up with the shift. Clearly that means the mirror can't be locked. I haven't come across anything that suggests fan vibration is an issue when the mirror remains unlocked. Besides that, my trials with the mirror locked, showed no benefit to star elongation at the centra of the FOV when the fan was operating. All I did get was some field curvature (concentrically elongated stars in the corner of the image) as the mirror locks clearly put an amount of pressure on the mirror inducing some tilt and a loss of collimation.

 

Gentlemen, I bought a 12V Noctua fan of the same dimensions as the one that came with the camera (60mm x 25mm). It is reputedly very low noise and vibration and only draws 1.5W compared to the 5W for the existing model. Of course I don't expect the cooling to be as good, but hopefully enough to get me going at least in the winter. However the new fan has a number of different 3 pin plugs and three wires the go to the fan. Unfortunately, they are different to the connector for the fan that comes with the camera. Can you give me any advice on connecting the fan to the camera?

 

Thanks Doug. So I can just cut the wire from the original fan with the connector that attaches to the camera, then attach the red and black wires to the red and black wires on the new fan?

 

Thanks Doug,

I can assure you that I would rather not have to do this. But I don't have a camera that I can use at the moment and I think you'll agree the risk is low. I am quietly confident this will allow me to produce acceptable images in terms of star elongation or hopefully the lack of it, at least in the winter months. It is just a lower powered fan of the same dimensions. If the power draw is any measure of the potential for vibration, then the result will be satisfactory. The fan design is, from what I understand, amongst the best out there for low noise and vibration. I am hopeful that this option will get me by until a liquid cooled option is ready.

 

Round stars at last!

I installed the Noctua fan in my SBIG 16803 by joining the red and black wires from the new fan to the clip connector of the old fan.


I carefully recovered the anti-vibration insulation you sent me and reused that.



I then installed the fan as normal and it worked beautifully.


The noise and vibration are massively reduced versus the old fan. The 1.5W power draw versus 5W is a pretty good measure of how much less energy would be going into vibrating the camera. The Noctua is also an extremely well designed fan in terms of managing the air flow to avoid noise & vibration. Of course it doesn't have the air flow punch of the old fan, but in my trials of last night, when the temperature here was ~ 3C, the power draw of the peltier cooler was ~ 25%. It was ~ 16% with the old fan under similar conditions of temperature. It will be interesting to see how it goes at higher ambient temperatures. A trial I did at 12C showed a power draw for the peltier cooler of ~ 48%.

My first series of trials produced very satisfactorily round stars, with eccentricities from 0.35 to 0.45. Oh what joy!
I produced these with SGPro and was surprised to see the RBI pattern, as I thought I had turned off RBI pre flash, but perhaps I made a mistake there. In any case that doesn't affect the star shapes.


1) NGC 6744 Red filter 10 min exposure





2) NGC 6744 Luminance filter 5 min exposure





Discussion

It is clear now that vibration from the fan supplied with the camera has been the cause of unacceptable star elongation in my images. I cannot say whether the issue is peculiar to my set up, however, feedback on this forum and others suggest that I am not alone but also that the problem is not peculiar to the SBIG 16803. Clearly as Colin said, this has rarely been identified as an issue with the camera, suggesting that there is some peculiarity, perhaps resonance or some such factor, that has resulted in the problems I encountered. Nevertheless I believe it would be beneficial to entertain the idea of redesigning the cooling fan to mitigate the risk of this causing an issue, or perhaps offering a lower vibration option with the compromise on cooling capability.

The Noctua fan is barely audible where the San Ace 60 is very loud. Noise and vibration are linked.

The question for me here will be whether I will run into issues in summer where night time temperatures can be 20C at night. Having the liquid cooling as an option will be a great fall back. In the meantime I am in business, assuming I can solve my RBI issue.

Niall

 

Thanks very much Colin. I look forward to receiving that.
For your interest these are the power consumption figures versus observatory temperature for the Noctua fan:
20C......80%
10C......48%
5C........36%
3C........28%