Forums » Instruments » Cameras »
How often are gain/dark references acquired on Falcon II and how many frames to include ?
Added by Anchi Cheng over 10 years ago
We have used the default 160 frame image and 10 images for a while but got a recommendation for at least 50 images (took forever). Any input from experienced users ? We also retook it before a three day run ends when images didn't look as good, but that could just be our sample. I would like to know what people do.
Replies (2)
RE: How often are gain/dark references acquired on Falcon II and how many frames to include ? - Added by Matthias Wolf over 10 years ago
Hi Anchi,
our engineer recommended 20/20 as a compromise.
There is no help to this dialog in the gain reference tool and it is also a bit confusing to me.
My current understanding is, "160 frame image" means an average from a series of 160 frames - these are sent at approx. 18 fps from the Falcon to the Falcon controller box, which averages them before sending a single image to the Krios PC (unless saving intermediate images). This would mean there will be 10 exposures, each with an exposure time of 160 frames / 18 fps = ~9 seconds. These 10 exposures are then averaged once again on the PC.
In principle, more frames should result in a better gain reference.
But it should be equivalent to take 40/40 or 160/10. Only the total averaged dose should matter.
However, if one is using a high dose rate of 50-60 elec/pixel/second as recommended by Richard, it might be better for the Falcon chip to use shorter exposures to prevent it from heating up too much.
Also, the usual practice is to match the total dose in a single image to the actual total dose used during data acquisition (this is less of a problem with FEI's linearization these days, but nevertheless). So if your typical exposure is 1 sec or less, then 20/20, 20/40, or something like 10/80 might be more suitable than the default 160/10.
I'd like to hear what FEI has to say about this.
MatthiasRE: How often are gain/dark references acquired on Falcon II and how many frames to include ? - Added by Anchi Cheng over 10 years ago
Mathias,
Thanks for your input. I agree with you about more frames should be better, but your final "usual practice" goes against that, and I would not do that myself since it is a CMOS frame-based camera. We've had early cases where users did that for other direct detector camera, the resulting references were rather bad.