Tuesday, November 24, 2015

The Dark Cube: new monochromatic 5mp CMOS camera

This is the second light, I've placed a simple convex lens between the BF10 and the camera nosepiece, this is effectively a reducer.

The result is that, the chip can hold the smaller solar disc with ample spaces so that I could move the sun around while imaging.

I've set sharpcap to get 600 frames, i.e. more than 2 minutes at around 4-5 fps with full 5mp resolution.  Since the solar disc was moving around, the Newton's ring was averaged out and the dust removed too.

The file is 8.5G in size!  Transfering the file from my tablet to my PC took nearly an hour via wifi!  Crazy indeed, and I've tried to move that file via a micro SD card but then the system complains that the file system on the SD does not support that large a file! 

Registax is moving slowly...

The time stamp was not from the camera and not from the driver too, as the manufacturer said, I've to double check with Sharpcap to see if I might turn that off.  A bit annoying...

The image was a bit soft, I suspect it could be a result of:

1. bad focus, you know, I'm effectively imaging at very big image scale despite I'm working at less than around 400mm since the pixel density is oh so high!

2. poor seeing, again, due to the big effective image scale... also the relatively long capturing window will blur fine detail.  I shall check.

In summary, if I were to take a fine full disc shot, use full 5mp resolution.

If I were to capture fine high resolution detail, I should go for ROI, and that will allow much faster frame rate and then, I could freeze seeing.

Monday, November 23, 2015

New monochromatic camera, setting up

Got a new one, I guess not many people use it, it's based on CMOS which is quite risky.  I don't know if it has Newton's ring... if there is, I will use the drift and average method.

It has 5M pixel so that I no longer need to go for barlows and the chip is quite big so that mosaic work is minimized.

The problem becomes frame rate due to the large volume of data.

With my Windows 8.1 tablet, it could go up to 5+ fps when saving to the internal SSD... it drops to lower than 1fps when saving to a good micro SD card, but then I know it's due to the internal card reader rather than the card itself.

When my Eee PC, the frame rate is slightly less than 3fps....

So, the tablet will be used for imaging, and I will have to free up space from the internal SSD to store the video clips.  I was thinking to remove the android partition which I seldom use.  But then I will want to make a partition dump as backup.


First light, since the frame rate is low, it could be hard to gather too many frames to stack, for example, to stay within a 100s time window, we could have only around 500 frames to play with:-

Note that the camera driver will insert time stamp on the upper left... !  Anyway, most of the newton's ring is averaged out except on the upper right area... I shall explore way to improve.  The amount of detail is nice, and we don't need to go for mosaic.


To free up extra space for AVI storage, I have decided to erase the Android of my Teclast.  However, I reserve the right to go back later, so I only remove the user data partition which is around 16G in size.  Before formatting that partition, I use:


to backup the partition image, so when I need it again, I will restore that particular partition.  Actually, I have another Android tablet so I will not use that on my Teclast.  And in fact, the Teclast Android is buggy, for example, the wifi password would not be saved and I will need to enter it every time after rebooting.

By removing that partition and reformatting to NTFS, we could use it for image capturing but then I found that the drive letter than I assigned via Computer Management (Windows 8.1) will be lost every time I reboot.  After some investigations, I found that partition was set hidden.  To unhide it, use diskpart commandline to unhide... select disk, select partition, partition detail, attribute volume set/clear hidden

too cryptic?  not already for old age IT guy.


To remove the Newton's ring, I will try to DIY a tilting device and there are some C mount extension tube from taobao, and they might be good candidate for the modification work, and it might be a good idea to sandwich a small reducer lens between the tilting mechanism so that I could have full disc in a single shot!

Before making the tilting adapter, I will try to image the sun without tracking...  i.e. just let the solar disc drift from one side of the CMOS to the other side slowly, and hopefully, the rings will be average out.

Wednesday, November 18, 2015

Canon FD 300mm f2.8 conversion: internal reflection

When using as a telephoto lens, I found internal reflection if there is bright light source around.  With careful inspection, it was due to a silver ring inside the adapter chain.  Therefore, I've blackened that ring, and now the problem was solved.

Tonight, I tried to use it again on the moon, and the result is very good.

Then, I placed the EOS-1.25" eyepiece adapter in, and I will need to insert a diagonal between the lens and the diagonal, the internal reflection issue comes back!  I soon realized why, it's because the silver barrel of the barlows!  On shielding it with black cover, it's all solved!

It's a good astrograph, it's a good day time telephoto lens.  However, it's not so good as a telescope for:

1. a diagonal has to be inserted, and thus, it has effective focal length of more than 900mm... so it's no longer a wide field instrument.

2. for high power observation, it's not wide to use a f2.8 objective

Therefore, I am going to sell it.

Monday, November 16, 2015

Replacement camera for my Lu055M

Since the death of my Lu055M, I had been trying my ASI 120mc on H-alpha imaging but the result is acceptable at best. I had an ASI 120mm before, but the newton's ring is too annoying so that I sold it soon. The image taken with a 120mm is great, however. I was impressed from the image on the real time screen, not to say the stacked version! I had been looking at those CMOS camera again, and I had one in mind and it features a CMOS sensor which should have newton's ring problem! The chip size is bigger, big enough to fit in the whole solar disc with my Ranger and it has 5 mega pixels and so I have no more need to make mosaic. With my Ranger which has 480mm focal length, the solar disc image would be smaller than 4.2mm but of course, with the prominences at the edge and it might exceeds the size of the CMOS sensor a bit. To elminate the newton's ring, the simplest way is to let the solar disc to drift round the sensor while capturing. Flat field might not work since the pattern of the newton's ring might change rather rapidly even if the solar disc moves a little bit in the field. Let me see how it works out, and the bottomline is that, it's going to be a fine lunar camera.

Saturday, November 07, 2015

Concept: Motorized focuser for my Ranger

Formerly, I was using a NGF-S and then upgraded to a NGF-CM. But those were sold due to the exit from solar imaging, and after getting the Solarmax II, I had been looking for a JMI focuser, but so far I could find any good deal. I had bought a Event horizon focuser but then it was incompatible with the old interfacing adapter. With the recent jump in shipping cost, and the lack of good second hand offering, I am thinking to motorize the original helical focuser of the Ranger. Going up to 2" for the Ranger is not very attractive and that also lowers the motivation towards JMI focuser, too. So I will be looking for a timing belt (同步皮帶), and a suitable timing pulley (同步輪/同步皮帶輪) and DIY suitable mounting adapters. The mechanism should be better be removable, and I could engage and disengage the motorizing mechanism too. Let's see how to move on. -- At the mean time, I will to use my adapted Canon FD 328 for visual purpose, see if it is acceptable or not for wide field scanning and high power observation.