Today, I started to reduce the coupling between DOF at the IP stage, measuring the LVDTALIGN (S) matrix and ACTALIGN (D) matrix as reported in klogs 2236 and 22337.
The updated matrices MEDM are:
S = [ 0.8210 0.4172 0.0687
-0.3629 0.8152 -0.0127
0.0815 0.0195 1.0064]
D = [ 0.6565 -0.4965 -0.0129
0.4545 0.6567 0.0098
0.0016 0.0251 1.0005]
I will test the efficiency of those matrices as soon as the suspension will be available.
Today I tested the efficiency of S and D matrices injecting broadband noise along L, T, and Y and measuring the residual coupling between d.o.f.
Pic.1 shows the residual coupling, as a function of the frequency, before and after changing S and D.
Pic.2, Pic.3, and Pic.4 show the comparison of the TFs before and after changing S and D.
Looking at these figures it is possible to note that the mechanical response of the IP is not dramatically changed, but the couplings in all DOFs are reduced. It seems that the new matrices are working fine.
Pic5, Pic6, and Pic7 show the comparison of the measured TFs and the fits along L, T, and Y.
The zero pole model used to fit the TFs has been uploaded in MEDM DCCTRL filter bank as susmod.
