0.90
#...
·
 |
|---|
|
After downloading the new driver of the above mentioned camera, i was eager to try out the new FullWell mode. Sharpcap sensor test showed me that the Full Well had gone up to 10100 ADU, 14,5 stops dynamic range, at the same time read noise had only increased to 4,4. But after trying on bright stars ,exposure time 300s, the histogram would only go as high as max 50000 ADU. How can it be? Surely bright stars would saturate very quickley? I had to increase gain to reach max ADU of 65535, but i think that defeats the purpose as maximum fullwell and dynamic range sinks. Does anyone know what is going on? Does anyone use the the FullWell read out mode? Linearity was only 70%. I suppose that is also not good.. CS Haakon |
7.90
#...
·
|
|---|
| A slight of hand from their part since you avoiding saturating the dynamic range without actually increasing anything by tweaking the response curve. I still can't understand how SC achieved 100100 ADU (I assume this is the figure you wanted to quote) with a 16-bit data range. |
|
0.90
#...
·
|
|---|
| Yes i quoted it from SC-sensor analysis with read out mode FUllWell/LCG, low noise. I suspect something is not right with this read out mode at gain100 (0) as saturation is not possible. Check it out: FullWell_LCG_SensorTest.txt |
|
0.00
#...
·
·
2
likes
|
|---|
|
Most people seem to not understand "fullwell". Not that they are dumb, it's more that "fullwell" somehow became a way to express the quality of a sensor. In reality, fullwell is a simple math relationship between gain (e/ADU) and the bit-depth of the A/D converter. Any 16-bit sensor that can operate at 1e/ADU will have a fullwell of 65535. You can see that the sensor analysis you linked makes perfect sense. Gain 100 = 1.553 e/ADU, fullwell = 1.553 x 65535 = 101778 (so within some small measurement error). Of course you can saturate star cores at gain 100, EFW - it will just take a longer exposure than the same star at gain 100 with EFW off. Why? Because each detected electron is converted to a lower ADU value with EFW on (I guess you need to think in terms of 100s of detected electrons, because at 1.5e/ADU some electrons will increase the accumulated ADU and some will not). EFW mode is not some magic trick - it is simply lowering the gain. |
|
7.90
#...
·
·
1
like
|
|---|
Haakon Rasmussen: That solves the mystery, the FW is in fact FWC+, that is the full well in e- not in ADU. Checking it out against my ASI294MC it does not perform really that well. |
|
0.90
#...
·
|
|---|
| Thank's to both of you. Is the bottom line that this read out mode does not bring anything new (and good) to the table? I will then continue to use the LcG mode without the FullWell option with a fullwell of 50000. Has been working really well for me. |
|
0.00
#...
·
·
1
like
|
|---|
|
It is a quite specialized mode. I used it last year for a very particular rendition of M42. 739x15s LCG/EFW gain 177 I wanted very clear resolution of the core and trapezium, and I think it succeeded. I could have done the same with even shorter exposures without EFW, and probably even better (lower overall noise / better SNR in the outer dusty clouds) at HCG but it would require thousands of *really* short exposures. EFW saved disk space and processing time. As far as I can see, that is the only reason to use it. |
