Hi, I've owned my STF for 2 years mostly I use the camera for NB with a small refractor. Last december I imaged the Orion Nebula in RGB and I was trying to bias calibrate 20s subs. The calibrated subs look horrible, completely over corrected. I inspected the master bias and the mean value of the master bias was higher than the mean value of the 20s lights. I thought that it was time to create a new set of calibration masters. I shoot a new set of calibration master and the bias level still higher than the 20s lights. I blamed my capture software (Sequence Generator Pro), so I did the same test with TheSkyX and to my surprise TheSkyX images agreed with SGP, so the capture software was not the issue. I blamed the firmware/dirvers, I updated them to the latest and greatest version and the issue still there. For instance the mean value of a master bias is greater than the mean value of a 30min master dark. This makes impossible bias calibration or dark scaled calibration (darks needs to be bias calibrated) I am running the lates and greatest driver/firmware: - Firmware 2.48 - sbigudrv: 4.90 build 1 This graph is the median value of master darks and bias taken @ -20C, the overscan region was not use and the bias level dark correction option was not set in the driver (by the way this option doesn't make any difference in the calibration frames) As you can see the bias frames median level is the highest! Any idea what could be wrong? Cheers, Jose
Hi, I haven't heard back regarding this issue. Do you want me to run more tests? I can even grant you remote access to my system in case you need to perform specific tests. Cheers, Jose
Hi, I made a quick test to see how overscan correction works with this issue and the issue didn't go away. I created 3 sets of calibration masters, bias, 20s and 60s darks with 32 lines of overscan area (32 subs per master). Then I calibrated each sub by adding a small pedestal of 200 ADU then subtracting the median value of the overscan area and finally adding back 200 ADU. After I calibrated each frame I created a master frame. The bias mean value is the highest among the 3, making bias calibration meaningless. It looks like the offset is not linear is some sort of multiplicative offset, making it impossible to calibrate with additive overscan correction. Could you at least acknowledge the issue? Cheers, Jose
I've been out of town for a few days, and am playing catch-up. First please make sure you SBIG Universal Drivers are up to date. Then check if the same is true for the camera firmware. If not then please update and retest. If you still have an issue let us know and we will look into it further.
Hi Doug, When I found the issue the first thing that I did was to update the drivers/firmware. AFAIK I am running the latest version of drivers/firmware. I am attaching a couple of screenshots with the driver and firmware information. This issue us reproducible with different capture software, PCs, usb cables, power sources, overscan calibration etc. Thanks for picking this up! José
As I recall, there was an issue with how the bias frames were implemented, which was fixed. If your images were taken with the old firmware, that might have been the cause. But pictures taken before the update would still show the effect. Can you confirm these were taken with the latest firmware?
Doug, The issue is reproducible with the newest firmware/drivers. I just made a test I took a bias and 20 darks in increments of 1s from 1s to 20s. I am attaching the results. Cheers, Jose
Hi Doug, FWIW, the external temperature (or the cooler duty?) greatly affects the median value of calibration frames. I've seen bias frames between 1000 and 1300 ADU, all taken a cooler temp of 0C however the external temperature was different. I know this is a known issue but probably related to this particular problem. Overscan calibration doesn't solve the problem. Regards, Jose
