New user - trying to determine my effective focal length

[steve bobey]

Plate solve is working fine be the Pixel Scale displayed has just 2 decimal places. I am trying to determine my actual focal length. It's  a 12" RCT F8, FL should be 2432. Camera Pixel Size is 3.8um. Astrometry.net results say pixel scale is 0.32 arcsec/px, which is correct for any focal length from 2376mm to 2450mm. In the log tail, I see this  "sqrt(det(CD))=0.320105 [arcsec]" which tracks the displayed pixel scale pretty closely and the units seem right. I'm wondering what that item is. Is it possible to get more decimal places on the pixel scale value? 

[steve bobey]

So I was able to look in the FITS header of the new.fits files and did find pixel scale to 6 decimal places. I am long by 16mm. I expected some because the pull screws in the primary mirror were pulled way in. However 16mm seems like a lot. Is there a +- error figure for pixel scale? Anyway this feels like progress, Thanks 

[Hans]

Hi Steve,

Just a friendly warning in case you intend to use the focal length measurement
to set the correct mirror distance on an RC:

"Be careful about using the scope focal length, measured using a plate solve,
to infer mirror spacing. This is only a coarse predictor since there are
inevitable tolerances in the mirror figures leading to few percents of focal
length differences between otherwise identical scopes, this is just normal and
to be expected."

by Dr. Gaston Baudat from Innovations Foresight, copied from
https://www.cloudynights.com/topic/703752-gso-16-inch-rc-collimate-with-an-autocollimator/?p=10961496

ps. do you happen to have an STL file for the baffle extension needed for this
exact model ?

-- Hans