NAME:
lrcmp - computes heights,angles given laser ranging distances.
SYNTAX: istat=lrcmp(pcdat,retdat,ext=ext)
ARGS:
pcdat[]: {lrpcinp} data read in by lrpcinp().
KEYWORDS:
ext : if set then return extended lr data structure: Included
space for az,zagr,zach (but they are not yet filled in).
RETURNS:
retdat[]: {lrdat} data with computed offsets.
istat: int always returns number of entries in pcdatnumber of entries computed
DESCRIPTION:
The routine computes the platform positions from the distomat distances.
It is called by lrpcinp after reading in the raw data from the laser ranging
PC.
The coordinates and matrices are:
lr6 \ / lr1
\ /d1
d2 \ T12 /
\ /
\./
/ \
/ \ |
/ \ |
\lr5 d3 / \d9 / lr2 | y (+ north)
\ / \ / |
d4 \ / \ /d8 ---------X
\ / \ / x (+ west)
\ / \ /
T8 ----------------- T4 Z (+toward the bowl)
/ d6 \
/ \
/d5 \d7
/ \
/lr4 \lr3
The d's are the 9 distances (6 measured , 3 known) that will determine
the 9 coordinates of the corners of the triangle
delta_d=[d1,d2,d3,d4,d5,d6,d7,d8,d9]
The matrix equation is:
A*dx = delta_d where
dx = [dx1,dy1,dz1,dx2,dy2,dz3,dx3,dy3,dz3]
1=t12,2=t8,3=t4 is the motion of these corners.
then
dx=delta_d * ainverse
If you compare this with any c code remember that idl has the
first index varying most rapidly and C has the second (mainly for the
init of ainv below).
The direction cosines look to be (cornerpoint - point)/dist
The platform corners seem to go in the opposite directions then mmd,jayaram
memo: T12,T8,T4 rather than T12,T4,T8.
Lets hope the ref distances were also switched.
the distomat ordering looks like lrn on the plot (i think).
NOTES:
let [] be a vector.
[Cm] is the position of one of the 3 corners.
[Crm] is the reference position for the [Cm].
[DRn] be one of the 6 distomat reference vectors. It points from
the laser ranger to the target when the 3 platform
corners are at [Crm].
[Dn] be the instantaneous position from laser ranger to target.
deltaDn : magnitude of ([Dn]- [DRn])
[dxm] be the displacement vector from corner m : [Cm]-[Crm]
Any motion of [dxm] perpendicular to [Dn] will not change the distance
The change in distance will be the projection of [dxm] onto
the [Dn] direction.
Since the relative motions are small compared to the length of
[DRn], use the unit vector of [DRn] rather than [Dn] when computing the
dot product.
The unit vector for a [DRn] is just [DRn]/magnitude(DRn)
[Drn] points from the distomat to the target.
As an example,
DR1 is at -x,+y1,+z
CRm is at 0,+y1a,0 where y1a
(See /pkg/rsi/local/libao/phil/lr/lrcmp.pro)
LRCMPDIST3 - COMPUTE PLTAVGHGT FOR ENTRIES WITH MISSING DISTOMAT READINGS
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrcmpdist3 - compute pltAvgHgt for entries with missing distomat readings
SYNTAX: n=lrcmpdist3(fitI,dist6,cmpI)
ARGS:
fitI : {} struct holding fit coef. see lrdist3inp()
dist6[6,m]: float distances to check for computation
RETURNS:
cmpI:[] holding the fit results
DESCRIPTION:
Search dist6[6,m] for the entries that don't have 6 lr measurements.
For those that have valid 3 135 or 3 246 reading (but not 6 good ones) ,
compute the average height from the fit coef in fitI (see lrdist3inp()).
Return the indices into dist6[0,I] in ii[] for these entries.
cmpI will contain:
IDL> help,cmpi,/st
** Structure <8c3d18>, 4 tags, length=6480, data length=6472, refs=1:
N LONG 462 number we found to compute
II LONG Array[462] index into dist6[0,*] for this
AVGPLTH DOUBLE Array[462] avg hght in meters
SETUSED INT Array[462] 1 dist135, 2 246
(See /pkg/rsi/local/libao/phil/lr/lrcmpdist3.pro)
LRDIST3INP - INPUT THE COEF FOR THE 3 DISTOMAT FITS
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrdist3inp - input the coef for the 3 distomat fits
SYNTAX: istat=lrdist3inp(fitI,yymmdd=yymmdd,fname=fname)
KEYWORDS:
yymmdd: long date for data. If not supplied use
current date. The coefs can be updated
occasionally.
fname: string filename for the coef. If not supplied
use the default name.
RETURNS:
istat:int 0 ok
-1 error inputing file
-2 no data for requested date
fitI: {} holds fitinfo
DESCRIPTION:
Input the coef's used to fit the avg plat height to 3 distomat
distances.
(See /pkg/rsi/local/libao/phil/lr/lrdist3inp.pro)
LRPCINP - INPUT PC LASER RANGING DATA.
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrpcinp - input pc laser ranging data.
SYNTAX: npnts=lrpcinp(yymmdd,b,daynum=daynum,year=year,ext=ext,use3=use3,quiet=quiet)
ARGS:
yymmdd: long year,mon,day to input. -1 --> today
RETURNS:
npnts : int number of entries found
0 file not found
b[npts]: {lrdat} array holding the data
keywords:
daynum: long if supplied then use daynum for day of year
year : long if daynum supplied, then year to use. If not supplied,
use current year.
ext: if set then return extended info.. az,gr,ch positions.
This uses the tdsummary info. It is not available for the
current month.
use3 : if set, then also compute hght if only 3 dist good
either 1,3,5, or 2,4,6
For those points using only 3 distomats,
only the avg height is computed,
the cornerheights, offsets and rotations are not computed
quiet : if set and /ext, and current month, don't output an error message.
DESCRIPTION:
The routine inputs a days worth of laser ranging data. It returns the
data from the laser ranging PC as well as the heigts converted to feet
above sea level (the conversion factors were measured in 1990 and have
probably changed!). The data available on disc goes back about 8 months
from the present. New samples are taken once every two minutes. During
periods of rain, the distomats have no readings and the distances are
set to 0.
The data structure for the returned data is:
The coordinate systems used are:
C1: x-west,ynorth,z-down cm no corrections to move to Sea level crdSys
C2: x-west,ynorth,z-up feet above sea level
name type coordSys value description.
DATE DOUBLE 1.0057176 daynumber with fractional day
TEMPB FLOAT 0.00000 bowl temp F (not recorded)
TEMPPL FLOAT 70.1000 platform temp degF
DIST FLOAT Array[6] distomat distances [meters].
DISTTM FLOAT Array[6] secs when reading completed.
AVGH FLOAT C2 1256.33 Average hght Ft above sealevel.
CORNERH FLOAT C2 array[3] Corner hght [12,4,8]
DX FLOAT C1 -1.50626 Avg x-translation [cm]
DY FLOAT C1 4.52196 Avg y-translation [cm]
DZ FLOAT C1 -100.191 Avg z-translation [cm]
XROT FLOAT C1 -0.00127837 rotation about x-axis [rad]
YROT FLOAT C1 0.00106705 rotation about y-axis [rad]
ZROT FLOAT C1 -0.00022241 rotation about z-axis [rad]
PNTS FLOAT C1 Array[3,3] [xyz,T12/T4/T8] corner points
DOK INT 1 if all 6 distances measured ok
-1 use dist 135 if use3 set
-2 use dist 246 if use3 set
SECPERPNT INT 120 spacing between measurements [secs]
NOTES:
You need to do @lrinit once before calling this routine to define the
{lrdat} structure.
History:
28feb18: include computation if disti 1,3,5 or 2,6,4 are valid
this will use the model fit for 3 dist.
update dok to be:
dok=1 all 6 ok
dok=-1 1,3,5 used for computation
dok=-2 2,4,6 used for computation.
(See /pkg/rsi/local/libao/phil/lr/lrpcinp.pro)
LRPCINPRANGE - INPUT A RANGE OF PC LASER RANGING DATA.
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrpcinprange - input a range of pc laser ranging data.
SYNTAX: istat=lrpcinprange(year,mmdd1,mmdd2,b,ext=ext)
ARGS:
year: long 4 digit year of interest
mmdd1: long mon,day to start
mmdd2: long mon,day to end
KEYWORDS:
ext: if set then try and return extended info.. az,gr,ch positions.
This uses the tdsummary info. It is not available for the
current month.
use3: if set then use 3 disto for avg hght if needed
RETURNS:
b[npts]: {lrdat} array holding the data
istat : int number of entries found
use current year.
-1 if an error
DESCRIPTION:
The routine inputs multiple days worth of laser ranging data. It returns the
data from the laser ranging PC as well as the heigts converted to feet
above sea level (the conversion factors were measured in 1990 and have
probably changed!). The data available goes back to 2000. The routine
constrains you to 1 year at a time. see lrpcinp for a description
of the data.
NOTES:
You need to do @lrinit once before calling this routine to define the
{lrdat} structure.
(See /pkg/rsi/local/libao/phil/lr/lrpcinprange.pro)
LRPLOTDIAG - PLOT DIAGNOSTIC INFO
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrplotdiag - plot diagnostic info
SYNTAX:istat=lrplotdiag(yymmdd,tmr=tmr,diag=diag,v=v,psname=psname)
ARGS:
yymmdd: int yymmdd to plot
KEYWORDS:
tmr[2]: float time range for plot. hour of day [hr0,hr1]
if only 1 value, then plot from hr0 to end of day. def: all day
diag : int which diag to plot
1 - integration for each distomat vs hour of day (default)
v[2] : float vertical scale for plot, if not supplied then autoscale
with 5% slop on each end
psname :string if supplied then generate a ps file with this name.
(no terminal output)
RETURNS:
istat : int 1 ok
-1 illegal diag code
-2 illegal time range
-3 illegal vertical scan . none or 2 numbers
-4 no lr data for date
-5 no lr data for time range
DESCRIPTION:
Plot diagnostic data vs hour of day. To use the routine:
* The user specifies the date to use (yymmdd format).
* You can optionally select a time range (in hours of day) using the tmr= keyword
* Next you select the diagnostic to plot (curently only diag=1 is supported.. integegration time).
* If you specify a filename for psname=filename then the plot will be sent to the
postscript file
Examples:
idl71
@phil
@lrinit
psname='lrdiag_180514.ps'
istat=lrplotdiag(180514,tmr=[14.,16.5],psname=psname
.. to plot to screen
istat=lrplotdiag(180514,tmr=[14.,16.5]
(See /pkg/rsi/local/libao/phil/lr/lrplotdiag.pro)
LRPLOTDISTFAIL - PLOT DISTOMAT FAILURES BY AZ,ZA
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrplotdistfail - plot distomat failures by az,za
SYNTAX: lrplotdistfail,d,year,gooddist=gooddist,includehr=includehr,$
exclHr=exclhr,totcnt=totcnt,badcnt=badcnt
ARGS:
d[npts]: {lrdat} laser ranging data input via lrpcinprange with the /ext
option set.
year: long 4 digit year for data in d (d had daynumber but no year)
KEYWORDS:
gooddist: int number of distomats that have to have good data
before the measurement is used (default is 3).
includehr:fltarr[2] range of hours (min,max) to use. default is
0 to 24.
exclhr:fltarr[2] range of hours (min,max) to exclude.
0 to 24.
title : string title to use on top of plot.. default is date range.
RETURNS:
totcnt:lonarr[2]: totcnt[0]: number of measurments made (after applying
includehr, exclude hr, and the ch at stow.
totcnt[1]: number of measurements that had at least
1 distomat bad but at least goodist ok.
badcnt:lonarr[6]: number of bad counts for each distomat.
also outputs the plot..
DESCRIPTION:
Make a plot of distomat failures versus az,za for the data set passed in.
To input the dataset use:
lrpcinprange(y,mmdd1,mmdd2,d,/ext)
Data is made available after the monthly processing of each month ( so the
current months is not available unless the processing is run manually).
The points in the dataset will be constrained so that:
1. there are at least "gooddist" distomat measurements for each point
used. This is to eliminate heavy rainfall where all of them were
out. The default value is that 3 distomats must be working.
2. The hour of the meausurment must fall within includehdr (default
is 0 to 24) and it must not fall within excludehr (default is
to not exclude anything.
The hour include/exclude allows to to look at daytime or nighttime
eg. includehr[6.,19.] will use 6am to 7pm (daylight)
exclhr=[6.,19.] will exclude 6am to 7pm (to keep nighttime)
The routine then plots the distomats that were not working versus
azimuth and zenith angle. The distomats at the same platform corner are
grouped together (1,6 td12), (2,3 td 4), (4,5 td8). Colors are used to
differentiate the two distomats at each color. Yellow dashed lines are
plotted at the azimuth of the corner that contains the two distomat
targets and at 180 degrees away. This is the azimuth where the dome
can cause the maximum platform tilt for this pair.
Example:
istat=lrpcinprange(2004,1201,1231,d,/ext)
pscol,'lrfail_dec04.ps',/full
lrplotdistfail,d,totcnt=totcnt,badcnt=badcnt
hardcopy
x
(See /pkg/rsi/local/libao/phil/lr/lrplotdistfail.pro)
LRPLOTHGHT - PLOT THE PLATFORM HEIGHT VERSUS TIME
[Previous Routine]
[Next Routine]
[List of Routines]
NAME:
lrplothght - plot the platform height versus time
SYNTAX: lrplothght,b,defhght=defhght,sym=sym,year=year
ARGS:
b[npts]: {lrdat} laser ranging data input via lrpcinp.
KEYWORDS:
defhght: float value to print out as default height (1256.22)
sym: int symbol to use for each point (-1 to -8). def: no symbol
year: long 4 digit year to use. Default is current year
DESCRIPTION:
Make a plot of platform height versus hour of day. The data in b[] must
first be read in using the routine lrpcinp. The output is in feet above
sea level and includes:
white: the average platform height
red: tower 12 height
green: tower 4 height
yellow: tower 8 height
The date at the middle of the plot and the time,temp, and average height at
the last point of the plot are printed (note that the last point is usually
from the start of the next day).
The plots show that the tiedowns are trying to keep the average height of
the platform fixed while the corners move depending on the unbalanced load
(which is a function of the dome/carriage house positions).
Example:
istat=lrpcinp(010505,b)
ver,1256,1257
hor,0,24
lrplothght,b
(See /pkg/rsi/local/libao/phil/lr/lrplothght.pro)
LRPLOTTEMP - PLOT PLATFORM TEMP FOR THE DATA SET PASSED IN.
[Previous Routine]
[List of Routines]
NAME:
lrplottemp - plot platform temp for the data set passed in.
SYNTAX:
lrplottemp,d ,title=title,cs=cs,xp=xp,off=off,sep=sep,clip=clip,year=year
ARGS:
d[n]: {lrdist} data input from lrpcinprange
KEYWORDS:
title - string title for top of each plot
cs - float scale factor for labels. For multiple windows you may
want to increase this to 1.5 or 1.8. default is 1.
xp - float xposition [0.,1.] where dates are printed. default:.02.
For hardcopy setting this to 1. puts the dates on the
right column.
off - float number of degrees to offset each day within it's
window so the plots don't lay on top of each other.
defaut:0.
sep - if set then make 1 plot per day user should set
!p.multi before calling the routine.
clip[] -float clip the data to [mintemp,maxtemp]. allows auto scaling
with bad data points..
year - long for the first point of the file. If not supplied then
assume the current year.
DESCRIPTION
Plot the platform temperature by hour of day for the data passed in.
(the routine limits the output to a maximum of 31 days).
The plot will have N subwindows with up to 7 consecutive days per window.
Each day will be color coded within a window. Any extra days (>28 will appear
in the last window). The routine internally uses daynumber of year for
computations so it will not cross year boundaries gracefully.
You should call ldcolph to setup the colortables for indices 0-10. You
should also set the vertical scale to temperature range you want via
ver,vmin,vmax. The horizontal scale should be set to hor,0,24.
The extra keywords that may be used are:
charsize=cs. If the multiple windows cause the letters to come out
too small.
EXAMPLES:
hor,0,24
ver,70,90
lrplottemp,d,title='platform temps for jul01', cs=1.6
NOTES:
The routine is setup to plot 1 month of data at a time.
(See /pkg/rsi/local/libao/phil/lr/lrplottemp.pro)