[Jenne, Rahul]
With LLO's initial success using the new R0 tracking (LLO alog 50897), we have also put in infrastructure for us to try here at LHO. This became a somewhat bigger project than we were intending, since there were discrepancies in what we are running versus what is in the SVN for the common library parts. This has been an ongoing conversation, but today we separated the ETM and ITM Quad_Master library parts by copying them into the h1-specific folder so that we didn't pull in LLO's changes that they've made for 3.3 Hz damping. Then we made the changes to implement the R0 tracking.
- Copied QUAD_MASTER in uapps/sus/common/models/ to QUAD_MASTER_H1 in uapps/sus/h1/models/
- This is the main library block for the ETMs
- Replaced the main lib block in h1susetmx and h1susetmy with this new version.
- Copied QUAD_ITM_MASTER in uapps/sus/common/models/ to QUAD_ITM_MASTER_H1 in uapps/sus/h1/models/
- This is the main library block for the ITMs
- Replaced the main lib block in h1susitmx and h1susitmy with this new version.
- Copied FOUROSEM_DAMPED_STAGE_MASTER_WITH_DAMP_MODE.mdl in the common folder to FOUROSEM_DAMPED_STAGE_MASTER_WITH_DAMP_MODE_H1.mdl in the h1 folder
- This is the library block for the L2 stage of the quads, and is used in both the ETM and ITM master libraries. (We started this only intending to change the ETMs for today, but in order to modify the L2 library part, we also had to put the changes into the ITM master).
- Replaced the L2 block in both the ETM and ITM quad masters with this new L2 library block.
- Checked all of these things into the svn, since at this stage we shouldn't have seen any functionality change, just separating out our files from the common folders.
- Added outputs for the L2 witness channels for L, P, Y in the new version of the L2 library part.
- Note that LLO only added L and P, since those should be more important (L is most important it seems, but then we might want P). But, to future-proof ourselves, we've also pulled the output for the Yaw witness just in case.
- Modified the R0 block in both the ITM and ETM quad masters (since the R0 isn't a library part, although perhaps it should have been so we didn't have to make these mods twice).
- We added 3 inputs for the 3 witness channels from L2.
- We created, as LLO did, filter banks to take the L, P, Y witnesses and send them to the input of the ADD block in a frequency-dependent way. Copying LLO, we also created a Roll filter bank that takes as the input the same pitch L2 witness as the pitch filter bank.
- We pulled those filter bank outputs over to the ADD blocks.
- Made tags for these new outputs of the L2 block, to take the signals over to the R0 inputs. Did this for both ETM and ITM quad masters.
- Made tags to pull the L2 witness outputs to the new R0 inputs.
- Compiled all 4 quad models. (did the make step, but not the make-install step)
- Confirmed that all new changes are in the svn.
On Tuesday, we'll do the make-installs, and reboot all 4 quad models. As LLO did, we must ensure that the gains to these filter banks are OFF before we take the suspension out of SAFE after the model is restarted, so we aren't sending the raw L2 witness channels straight out to the R0 actuators.
LLO has shared with us the filters that they used in LLO alog 50889, so we'll probably use those as a first try for implementing the R0 tracking. We might need to check if the ITM filters can be the same as the ETM filters, since there are some differences between the ETMs and ITMs.
Then, perhaps with injections (so that we're not reliant on the environment to stay constant), we'd like to check with some on/off tests to see the effect on scattering arches. Probably we'll ask for some commissioning time Wed AM, after we've got these installed on Tuesday.
