Why is the bias frame higher ADU than the dark frame?

Discussion in 'Image Capture' started by Elizabeth, Sep 24, 2015.

  1. funkmeisterfred

    funkmeisterfred Standard User

    Aug 12, 2015
    St. Louis, MO
    Greetings all -

    I am an STF-8300M user, and have more data and information to add to this topic.

    First, I can confirm that the ADU of my 10s dark images is about 40 counts higher than the ADU of my 0.1s dark images. This is found to be true both with and without "Auto Bias Level Correction" enabled.

    Camera and driver details:
    • Model: STF-8300M
    • Serial: 83F011323
    • Driver version (stf-8300usb.bf2): 2.47
    • Driver Version (stf-8300ga.bf2): 2.18
    I only discovered this, however, because I was looking into another phenomenon that has me equally as puzzled. I'm glad I found this topic, as I suspect there's a chance they are related. Here's the story:

    I recently shot 20' Hα frames of IC342. The Hα in this region is incredibly sparse, so as expected, there are large areas of "nothing" which would be nearly equal to having a cap on the camera for a dark frame. The ADU of these regions is about 920. I was alerted to an issue when bias correction completely zorked the data in these frames during calibration, and I subsequently found my bias master ADU to be 1060. This bias master was shot and generated in August 2015.

    I presumed that, perhaps, the characteristics of my camera have changed over the course of a few months, so I re-shot my bias frames. The new master has an ADU of 960 -- lower than my August run, but still higher than my 20' Hα frames!

    I should note that I shoot all exposures (calibration and light) at a sensor setpoint of -10C. I should also note that for all of the above frames, exposures were acquired via Sequence Generator Pro.

    I shoot my bias frames indoors (in August, inside temp was ~24C, while inside temp is ~21C in November). My Hα frames were captured at an ambient temperature of roughly 3C.

    To add to this story, today I ran more 0.1s bias frame tests using CCDOps, and chose to do so maintaining a sensor temperature close to room temp (at 20C). These results are further remarkable:
    • 12:46pm: Initial hookup, and CCDOps reports a sensor temp of 20.4C. With cooling disabled, a 0.1s bias frame shows an ADU of 835.
    • 12:47pm: Cooling is enabled at a setpoint of 20C. I wait for temperature to stabilize.
    • 12:51pm: 0.1s bias frame shows ADU of 903
    • 12:56pm: 0.1s bias frame shows ADU of 912
    • 12:59pm: 0.1s bias frame shows ADU of 915
    • 1:19pm: 0.1s bias frame shows ADU of 923
    In other words, what I'm seeing here is that the ADU increases asymptotically after initial camera power-up.

    In summary, I see 3 specific issues as relates to bias offset:
    • ADU of longer dark frames (e.g. 10s) is found to be lower than shorter dark frames (e.g. 0.1s)
    • Hα exposures of length 20' at 3C ambient are found to have a lower ADU than 0.1s bias frames at 21C ambient
    • 0.1s bias frame ADU is found to increase after camera power-up.
    The only explanation I can muster is that the temperature of the analog read circuitry (which I understand is not temperature controlled as the sensor is) is directly affecting the bias offset. Perhaps 0.1s exposures generate more analog "activity" and heat than 10s frames, causing a greater effective offset. Perhaps the 3C ambient cooling of my Hα exposures knocked that offset down as compared to my indoor 21C bias frames. Perhaps after powering on the camera, the analog circuitry slowly warms up and results in an increasing offset over time.

    Or perhaps I'm completely mistaken.

    In any event, I hope this is helpful to this discussion. I look forward to hearing any responses, and hopefully some feedback from a staff member. Currently, the only solution I see would be to enable the overscan regions in order to compensate for these changes, regardless of what's causing them.

  2. edkammer

    edkammer Standard User

    Sep 22, 2017
    I have a similar problem with the STF-8300c, using Maxim DL. All that I discuss below is with 2x2 binning ( although problem occurs with other binning setings). At -15c the bias frame is about 1230 average, while dark frames out to 600s are less on average.
    At room temp the bias is the same, and dark frames decrease for small times, then increase as expected with increasing time.
    I did not see a solution to this bias problem in the thread above. how can this issue be resolved?
  3. Doug

    Doug Staff Member

    Sep 25, 2014
    First step: make sure you're running the latest firmware, and update your drivers.
  4. edkammer

    edkammer Standard User

    Sep 22, 2017
    Camera firmware update resolved the problem. Thanks

Share This Page