How to enter transformation coefficients

8 posts / 0 new
roe
How to enter transformation coefficients

I use three filters: V, B and I.  I took images in each of these filters of M67 and used Tom Krajci's spreadsheet to analyze the data.  I got lots of slopes, etc.  Reading the Transforms-Sarty.pdf available from AAVSO I see three "cases" that might apply: BVI case, VI case and BV case.  Which case should I use (or should I use more than one? - note that I would probably only use two of the three filters on any given target)?

Given that, how would I enter the respective values of Tv. Tvi, Tbv, and Tb as defined by Sarty?

MZK
Transform Coefficients

Hi Jim:

under filter band coeff., click add transformation rule once, row opens, enter data for Tb by first entering filter coeff value (near 0), select B in next column,  in next column select which other standard filter was used for that value (e.g., V), for (B-V)

repeat by clicking transformation rule again, new row opens, repeat for next filter coeff., for example you may enter a different Tv value in which you may have used more than one pair of standard filters such as VR and VI.

I use Canopus/Photored to calculate my filter coeff. and I populate a full set of T(filter) for several pairs of standard colors, and 3 color index slopes for B-V, V-R, V-I. I even have a set of filter coeffs for my clear filter.

For color indexes, under color index coeff, click add transformation rule, enter the value near 1 and select which color pair you used.

repeat as necessary for all your coeffs.

Ken

roe
MKZ wrote  "enter data for

MKZ wrote

"enter data for Tb by first entering filter coeff value (near 0), select B in next column,  in next column select which other standard filter was used for that value (e.g., V), for (B-V)"

OK, I did this using Sarty's definition of Tb as the slope of (B-b) vs (B-V).  When I went to transform a pair of images taken in B and V, VPHOT said there was no value for Tv <----.  Now what?

MZK
Tv

Plot V-v versus B-V. see what it looks like.

roe
Tv

I had done that, very close to zero (-.00007). As I read farther down in Sarty I see the UBV case, which I hadn't considered as I have no U filter.

So, new problem is VPHOT won't save the value. When I try to transform two images (B and V) it comes up with a blank in the Tv field. I can enter it manually but that doesn't seem right as I went to all the trouble to set it up in my telescope file?

However, when I try to transform two images in V and I, VPHOT fills in the value I registered in the Ti slot?

Further confusion. I can guess what Ti is (slope of (I-i) vs (V-I)?) but Sarty doesn't define nor use a Ti. If VPHOT doesn't follow Sarty, what does it follow?

kge
Tv

That is because you have entered Tb, not Tv.

VPHOT does not follow Sarty as such, although the basic principles are the same. VPHOT is a bit less elegant in that you need to specify more parameters, but at the same time gives you more freedom to combine any filter. So if you have BVI filters you should enter Tb, Tv and Ti. You should have Tv based on both B-V and V-I if you plan on doing B-V and V-I transformation (meaning plot V-v vs. B-V and find the slope, then do the same for V-v vs. V-I).

Geir

roe
Tv

OK, the dawn has broken in my dim mind.  Where I got off track was MZK's first reply to enter B in the first box (should have been V).  It seems to be working now and I am cross checking the results against the stars in M67.

roe
How to enter transformation coefficients in VPHOT

Things seem to be working well now.  Thanks to the urging and helpful spreadsheet from Tom Krajci I calculated my transformation coefficients using V, B and I images of M67.  I checked them by transforming the V and B pair and the V and I pair against 25 of the well known stars.  My results are that the average V-V* and B-B* for the V-B case are 0.039 and 0.020 respectively.  The average V-V* and I-I* for the V-I case are 0.040 and -0.003 respectively.

There were, of course, two independent measures of V* which I will call V*b and V*i.  The average difference was only 0.002.  Which raises new questions which I will take up in a new thread.