Strange behavior in new LodeStar X2

[acf9...@gmail.com]

I took delivery today of a second LodeStar X2 guide camera.  I also own a two year old LodeStar X2.  I was testing the new X2 and noticed it behaved differently than my original.  Keeping all settings the same, I switched between cameras.  I have the cameras connected as "Starlight Xpress SXV."  The "brightness" slider is full right.  The cameras are being tested at my desk and they are capped; 2 second exposures are being taken.  No dark libraries are being used.

Original X2 shows black background and hot pixels as you would expect:

New X2 shows so much background the hot pixels are not visible as shown in the screen shot below.   As I watch frames download, I can see substantial variation in the brightness level of the background.  About one frame in fifty will have a black background with four hot pixels showing; then back to the screen you see below.

I switched the new X2 to the ASCOM driver and saw the same results.  I don't think I'll be able to see guide stars with this new camera.   The only guide cameras I have ever owned are these two LodeStars X2 -- is this new one defective?

Charlie 

[Andy Galasso]

Hi Charlie,

I don't think you should make any judgments based on the screen display.  The noisy-looking image could be the result of fewer hot pixels in the new camera. For screen display PHD2 does an automatic histogram stretch based on the lowest and highest ADU values in the frame. With few or no hot pixels the stretch will be more extreme and highlight the ordinary camera noise.

To do a true comparison, you should disable darks then take a frame or two with each camera saving the image to disk (menu: File, Save Image).  Then, open the FITS images in your image processing software and use that software to evaluate the camera frames.

Andy

[acf9...@gmail.com]

Andy:

Thanks, I'll give that a try to check it.  But I believe the answer may be staring me in the face -- the power LED is red.  On my original, the power LED is a faint green color.  I thought Starlight had replaced the faint green LED with a bright red one so you could actually see it in the dark.  I suspect RED is DEAD.

Charlie

[acf9...@gmail.com]

Andy:

I followed your advice.  I connected my old X2 to SGPro and took a 2 sec dark frame.  Looked like a normal dark frame with a lot of hot pixels.  Then I connected the new X2 to SGPro and took the same 2 sec dark frame.  It also looked like a normal dark frame but with only a few hot pixels.  

So, now I need to find out why the LED on the new camera is red instead of green like on the old one.

Also, if the new X2 is not broken, I still have concerns about how it is going to perform on the actual guide scope.  I will take it to my observatory Sat night and just try it.

Charlie

On Friday, March 24, 2017 at 8:59:20 PM UTC-5, Andy Galasso wrote:

[acf9...@gmail.com]

More info this morning.  I decided to try the "LodeStar Guider (ASCOM)" option with the new LodeStar X2.  Initially, there was no difference with the new X2 having a huge amount of back ground noise that completely drowned out the hot pixels.   I decided to take a look at the ASCOM driver configuration dialog box.  On a chance, I enabled the "Gaussian Blur" option.  It made big difference.  The back ground is now a uniform grey and I can see the hot pixels -- there are only 10 (that I can see).  The new X2 camera still has a much brighter back ground than the old one but the new one appears to be usable now.

So, big difference in behavior between the old X2 and new X2 when using "Starlight Xpress SVX" selection and using "LodeStar Guider (ASCOM)" selection.  If there are some diagnostic files you would like to see, let me know.

Charlie

[Andy Galasso]

Charlie,

Could you upload your 2 raw camera frames:

1. connect the original camera as Starlight Xpress SXV, disable darks, set 2 second exposure, loop, stop, File => Save Image

2. repeat with the new camera

Post a link to the 2 FITS files (or attach them to a post).

Thanks,

Andy

[acf9...@gmail.com]

Can do this Monday, 3/27.

Charlie

[acf9...@gmail.com]

Andy:

I ran the tests with 2 sec exposures and no dark library.  I saved the FITS files taken by the old and new LodeStar X2 cameras.

Here is a Dropbox link to the zipped file:

https://www.dropbox.com/s/vz5sndk52a2vv0t/OldNewLodeStarInfo.zip?dl=0

It contains the following saved images:

1.  Old X2 camera with ASCOM driver

2.  Old X2 camera with SXV driver

3.  New X2 camera with ASCOM driver

4.  New X2 camera with SXV driver

5.  ASCOM driver dialog box screen shot

6  Screen shot of download image with Old camera and ASCOM driver.

Note that both cameras looked virtually the same on screen under ASCOM except the new camera has fewer hot pixels and has a brighter background.  Under SXV, the two looked dramatically different.  

I viewed the four FITS files in PixInsight.  The two ASCOM files actually look noisier than the SXV files but the PHD2 screen displays appear just the opposite.  Seems to be some screen stretch issue between the two cameras.

I was able to test the new camera under ASCOM last night and the auto guiding was completely normal.  I was using v2.6.3dev1.  Interestingly, if I enabled the dark frame library, the down loaded frames were very noisy looking.  Disable dark frame library and the frames were totally smooth.  v2.6.3dev1 completely ignored the hot pixels, so that new feature worked well.

[Andy Galasso]

Charlie,

Thanks for posting the files.  This confirms the explanation I suggested in my first response to your original post. The old camera has an ADU range of 1251 - 9715. The new camera has a range of 1186 - 1384 ADU because it has virtually no hot pixels. PHD2's automatic histogram stretch is based on the range of ADU values, so the new camera gets a much more aggressive stretch and the noise is highlighted.  The noise as measured by PixInsight is actually lower on the new camera (σK = 1.575e-04) than the old camera (σK = 1.714e-04).  The screen stretch in PHD2 is for display purposes only, guiding is using the raw un-stretched pixel values.

Andy

[acf9...@gmail.com]

Andy:

Thanks for the explanation.  The differences in the image display for the two cameras had me convinced the new camera was bad.  Turns out, its actually better.   As always, your time in helping out all of the PHD2 users is greatly appreciated.

cm