4 generate lookup tables

.env

@@ -0,0 +1,2 @@
+CLUSTER=k8s-i18
+NAMESPACE=i18-beamline

README.md

@@ -10,10 +10,25 @@ Plans and devices for the DLS beamline i18
 This is where you should write a short paragraph that describes what your module does,
 how it does it, and why people should use it.
 
-Source          | <https://github.com/DiamondLightSource/i18-bluesky>
-:---:           | :---:
-PyPI            | `pip install i18-bluesky`
-Releases        | <https://github.com/DiamondLightSource/i18-bluesky/releases>
+table again, t1 theta and d7bdiode
+MO table is ok
+
+    16 channels Xspress3 detector
+        -EA-XPS-02:CAM:MaxSizeX_RBV
+      also ArraySize
+      also :CONNECTED
+
+pinhole PVs? -AL-APTR-01:TEMP1
+
+diode reading and also DRAIN
+diode is ok, there are A and B variants
+motors A and B
+camera not used
+
+|  Source  |     <https://github.com/DiamondLightSource/i18-bluesky>      |
+| :------: | :----------------------------------------------------------: |
+|   PyPI   |                  `pip install i18-bluesky`                   |
+| Releases | <https://github.com/DiamondLightSource/i18-bluesky/releases> |
 
 This is where you should put some images or code snippets that illustrate
 some relevant examples. If it is a library then you might put some

docs/alignment/idgap.md

@@ -0,0 +1,54 @@
+# idgap alignment
+
+## prerequistes
+
+- must be ran with beam
+
+## explanations
+
+harmonics are an emergent property of the optics physics, found out empirically
+harmonics rise monotonically with Z atomic number. Still
+`/dls/science/groups/i18/lookuptable/Fe_Co_Ni_Aug24.txt` - so three elements next to one another, all 3 under the same harmonic
+
+XFAS community uses eV for initial energy and inverse angsotmr sfor final energy
+
+energies for each element are stable for 200 years and often scanned with a 10% margin of errror to read more
+
+Mo hightest energy so at the highest harmonic - 19
+
+<https://docs.scipy.org/doc/scipy/reference/constants.html>
+
+usually 1 element has one edge
+
+## physical description
+
+idgap(undulator)- beam -> Monochromator (Bragg angle) - beam -> diode d7 (readout)
+
+## steps
+
+### preparation step
+
+- set diode to `in line diode` mode
+- filter A must be adjusted so that the current is below 20 (magic number found decades ago)
+- the diode must be configured so that it's giving maximum intensity variance without being oversaturated
+- that is done through adjusting the discrete size of a gap between the diode and the beam
+
+### Scannign step
+
+1. scan goes to the first angle/energy - using the previous lookup table
+2. mm on x and readig is in microAmps
+3. peak fitting, resulting in a 2d relationship of x,y maxxing the z value
+
+### Cleanup stage
+
+- move d7 filter a back into 'gap' mode
+- move d7 filter b back into 'gap' mode
+
+## Common issues
+
+sources of issues when running manually:
+
+- saturation
+- if the original idgap is so far out of the scan
+
+usually takes 10 minutes, so for 19 harmonics 3 hours?

docs/alignment/index.md

@@ -0,0 +1,107 @@
+# alignment plan overview
+
+sample stages T1X
+T1Y
+
+(for tomography also T1Theta)
+
+two ion chambers
+
+kb motors, two mirrors vertical and horizontal
+ecah has 2 motors, bend1, bend2
+usually moved simultaneously, but for fine focus separately
+
+get the gaussian shape IT transmission detector we get the shape
+
+    first, we lookup table - calibrate the DCM
+    - measure foil, etc Fe, Mg, then absorption spectrum
+    then the xanes absorption - then derivative, argmax of the first derivative
+    https://www.geeksforgeeks.org/python-sympy-derivative-method/
+    then Bragg offset is adjusted to match the calibrated value
+
+    second the idgap lookup tables
+    - for 10-15 points inside the energy range for this element
+    we scan the gap fo the insertion devise, looking for the maximum
+    then quadratic interpolation, written into the file,
+    then GDA probably some interpolation
+    TFG calculates frequency from current via voltage
+    so we need to load the panda configuration
+
+    align the pinhole to reduce the scatter -
+    400 micron or 200 micron, then centralize it
+    usuallly not seen immediately
+    FocusingMirror misses curvature
+    preparation for the wire stage - check if we have any
+    gold wires on the sample stage - scanned in one direction
+    first horizonal, vertical
+    then record with IT the absorption profile, derviative and fitting
+    then changing the bend
+    could be 10 iterations, in either direction
+    to minimuze the beam size until it changes
+    to see the beam shape and the size
+    takes usually 30 minutes to go through focusing manually, 2-3 hours
+
+    visual comparison fo the drviative -
+    best if without the tails, could be parametrized
+    or 50 micron beam - and then defocus to get to that
+
+    golden plate with wires is moved by some other location
+
+## Alignment types
+
+There are three major types of alignment:
+
+- idgap
+- pinhole
+- wire scan
+
+Other alignment (non exhaustive):
+
+- KB mirror optimization (?)
+
+Only the idgap one caches the results in a lookup table.
+
+### IDGAP alignment details
+
+moving the monochromator to a given Bragg angle, keeping it fixed and then scanning the undulator gap across a small range.
+
+On the d7 diode the Bragg with a peak is that gap
+
+2d scan of Bragg angle vs undulator gap, and peak compared to d7
+
+### pinhole
+
+## from the files
+
+first, we lookup table - calibratre the DCM -
+measure foil, etc Fe, Mg, then absorption spectrum
+then the xanes absorption - then derivative, argmax of the first derivative
+https://www.geeksforgeeks.org/python-sympy-derivative-method/
+then Bragg offset is adjusted to match the calibrated value
+
+second the idgap lookup tables -
+for 10-15 points inside the energy range for this element
+we scan the gap fo the insertion devise, looking for the maximum
+then quadratic interpolation
+written into the file, then GDA probably some interpolation
+TFG calculates frequency from current via voltage
+so we need to load the panda configuration
+
+align the pinhole to reduce the scatter
+
+- 400 micron or 200 micron, then centralize it
+  usuallly not seen immediately
+  FocusingMirror misses curvature
+  preparation for the wire stage - check if we have any
+  gold wires on the sample stage - scanned in one direction
+  first horizonal, vertical
+  then record with IT the absorption profile, derviative and fitting
+  then changing the bend
+  could be 10 iterations, in either direction
+  to minimuze the beam size until it changes
+  to see the beam shape and the size
+  takes usually 30 minutes to go through focusing manually, 2-3 hours
+
+visual comparison fo the derivative- best if without the tails,
+could be parametrized
+or 50 micron beam - and then defocus to get to that

helm/charts/redis-service.yml

@@ -0,0 +1,11 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: redis-service
+spec:
+  selector:
+    app: redis
+  ports:
+    - protocol: TCP
+      port: 6379
+      targetPort: 6379

old_scripts/Create_deg_Lookuptabledcm.py

@@ -0,0 +1,164 @@
+from gda.analysis.numerical.optimization.objectivefunction import AbstractObjectiveFunction
+from gda.analysis.numerical.linefunction import AbstractCompositeFunction
+from gda.analysis.numerical.linefunction import CompositeFunction
+from gda.analysis.numerical.linefunction import Gaussian1D
+from gda.analysis.numerical.linefunction import BGAsymmetricGaussian1D
+from gda.analysis.numerical.linefunction import BGGaussian1D
+from gda.analysis.numerical.linefunction import AsymmetricGaussian1D
+from gda.analysis.numerical.linefunction import Polynomial
+from gda.analysis.numerical.linefunction import Parameter
+from gda.analysis.numerical.differentiation import Differentiate
+from gda.analysis.numerical.optimization.objectivefunction import chisquared
+from gda.analysis.numerical.optimization.optimizers.leastsquares import minpackOptimizer
+from gda.analysis.numerical.optimization.optimizers.simplex import NelderMeadOptimizer
+from gda.analysis.datastructure import DataVector
+from time import sleep
+#
+# A simple method for performing the gap scan
+#
+def scangap(start_gap,stop_gap,step_gap):
+	npoints=int((stop_gap-start_gap)/step_gap)
+	currentgap=start_gap
+	xdata = DataVector([npoints])
+	ydata = DataVector([npoints])
+	for i in range(npoints):
+		pos sc_idgap currentgap
+		xdata[i] = currentgap
+		ydata[i] = d7bdiode.getPosition()
+		#ydata[i] = multict.getPosition()[0]
+		print 'i gap',i,('%.3f' %xdata[i]),('%.5f' %ydata[i])
+		#print 'i gap',i,('%.3f' %currentgap),('%.3f' %multict.getPosition()[0])
+		currentgap=currentgap+step_gap
+	return xdata, ydata
+
+#
+#
+# Fit the data from the gap scan with a gaussian and return
+# the peak position
+#
+#
+def fitdata(xdata,ydata):
+	# Find the min and max of the data set
+	maxcount=0
+	mincount=0
+	maxval=ydata[0]
+	minval= 10000000000000.0
+	ysize=len(ydata)
+	for i in range(ysize):
+		if(ydata[i]>maxval):
+			maxval=ydata[i]
+			maxcount=i
+		if(ydata[i]<minval):
+			minval=ydata[i]
+			mincount=i
+	for i in range(ysize):
+		ydata[i]=ydata[i]-minval
+	for i in range(ysize):
+		ydata[i]=ydata[i]/maxval
+
+	gauss = BGGaussian1D([ydata[maxcount]/100.0,xdata[maxcount],0.005,minval])
+	#gauss=BGAsymmetricGaussian1D([ydata[maxcount]/100.0,xdata[maxcount],0.005,0.005,minval])
+	gauss.getParameter("area").setLowerLimit(1.0E-5)
+	gauss.getParameter("area").setUpperLimit(0.1)
+	gauss.getParameter("position").setLowerLimit(xdata[0])
+	gauss.getParameter("position").setUpperLimit(xdata[ysize-1])
+	gauss.getParameter("sigma").setLowerLimit(0.0005)
+	gauss.getParameter("sigma").setUpperLimit(0.05)
+
+	#gauss.getParameter("sigma1").setLowerLimit(0.0005)
+	#gauss.getParameter("sigma1").setUpperLimit(0.05)
+	#gauss.getParameter("sigma2").setLowerLimit(0.0005)
+	#gauss.getParameter("sigma2").setUpperLimit(0.05)
+	gauss.getParameter("background").setLowerLimit(1.0E-12)
+	gauss.getParameter("background").setUpperLimit(0.1)
+	myfunction = CompositeFunction()
+	myfunction.addFunction("Gaussian", gauss)
+	chifunc = chisquared(myfunction,[xdata,ydata])
+	minpack = minpackOptimizer(chifunc)
+	minpack.setMaxNoOfEvaluations(150000)
+	minpack.reset()
+	minpack.optimize()
+	bestValues=minpack.getBest()
+	print 'maxcount',xdata[maxcount]
+	print 'best values',bestValues,minpack.getMinimum()
+	return bestValues,xdata[maxcount]
+
+#
+# Script to regenerate a lookup table for a given harmonic
+#
+#angleStart=10439
+#angleStart=55400
+angleStart=70400
+angleStop=54000
+angleStep=-1000
+gaprange=0.030
+gapstep=0.003
+offset=0.0
+
+
+
+#
+# The code uses comboDCM, i.e. the current lookup table to get a starting point for the scans
+#
+converter = finder.find("auto_mDeg_idGap_mm_converter_Si111")
+
+converter.enableAutoConversion()
+
+pos sc_comboDCM_d_Si111 angleStart/1000.0
+#pos comboDCM_eV 7200
+
+#converter = finder.find("auto_mDeg_idGap_mm_converter")
+
+converter.disableAutoConversion()
+
+lookup = finder.find("lookup_name_provider_Si111")
+print 'Harmonic set to ',lookup.getConverterName()
+
+#
+# Selection mode
+# 0 fit the curve with a gaussian and pick the centre
+# 1 use the peak position
+selectionMode=1
+
+new_harmonic_file='/dls/science/groups/i18/lookuptables/P_Jul24.txt'
+#new_harmonic_file='/dls/science/groups/i18/lookuptables/S_K_May24.txt'
+#new_harmonic_file='/dls/science/groups/i18/lookuptables/Ni_K_May24.txt'
+#new_harmonic_file='/dls/science/groups/i18/lookuptables/Ir_L3_Jan24.txt'
+
+
+fod=open(new_harmonic_file,"w")
+print >>fod,"Bragg\tGap"
+#
+# Will update
+#
+#pos comboDCM angleStart
+#current_gap=idgap.getPosition()
+for j in range(angleStart,angleStop,angleStep):
+	#convert to degrees
+	i = j / 1000.0
+
+	# Move to position
+	pos sc_comboDCM_d_Si111 i
+	sleep(2)
+	current_gap=sc_idgap.getPosition()+offset
+	# Scan gap
+	myxdata,myydata=scangap(current_gap-gaprange,current_gap+gaprange,gapstep)
+	result,bestx=fitdata(myxdata,myydata)
+	# Write to new file
+	if(selectionMode==0):
+		myline="%.3f\t%.5f" %(i,result[1])
+	else:
+		myline="%.3f\t%.5f" %(i,bestx)
+	print >>fod,myline
+	#print >>fod,i,"\t",result[1]
+fod.close()
+
+converter.enableAutoConversion()
+print 'Done'
+
+
+
+
+
+
+