40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  PSL  Not logged in ELOG logo
Message ID: 1747     Entry time: Wed Oct 12 13:18:19 2016
Author: awade,Aidan,rana 
Type: Notes 
Category: Other 
Subject: Meeting: discussion of thermal controls and actuation 

Rana, Aidan and I had a discussion about the design and requirements for thermal controls
 
 There are number of things to address:
  • Making the tank windows smaller (probably 1.5” is sufficiently small once the aperture of the mount is taken into account);
  • Shielding the inside of the tank Al foil plus a mask over the cavity ends made of polished metal (this can go in almost strait away when we install the cavity shields);
  • Need to investigate windows: for a really good seal double viton or a flange directly braised onto a metal fitting. We actually need to THINK about this design;
  • At 18 W the steady state thermal load on the tank is very high (according to rana), need to check with Tara what it was previously and improve on this;
  •  TCS heater circuit units are overkill for what we need we can build our own (see picture below)
  • Offset of signal into ADC and noise might be an issue. Whatever HF noise there might be, this can be averaged out and also LP filtered.  The Average response will be a lot more stable.
 

Windows:

There are a few options for windows, it would be good seek out designs for 1.5” windows that are already done. Off the shelf options are really not very good: we want something wedged and coated with decent AR coating for minimal loss and reflections. The windows associated the TCS and the LIGO pre-modecleaner come to mind but they are probably not the only examples. 
 
There are a variety of sealing options including viton (single or double o-ring) and methods where the window is directly braized to a metal mechanical fixture. The latter technique is fraught with difficulty given the high temperature of the weld technique and the damage this messy process can do to coatings. Maybe we could get pre-braized and then coat. However, this comes with issues, the obstruction of metal part can get in the way and cause ‘shadowing’ of the optic around the edges. Most manufactures are tooled to do naked substrates.
 
Lead time and design on updated windows may be months. Maybe we should kick off the process for obtaining quotes etc for optics now and push forward with design of a mount.
 

Dumping of ghost beams:

We want to collect promptly reflected beams within the vacuum can. We sould think about maybe putting some beam catchers in there. Would be good to make quick calculation on estimations of beam wedge and work out where/how these could be put in. The fact that windows will be possibly more wedged with smaller aperature means that we really need to address how this is dumpped.  It will also be very important for the final science measurment and noise hunting, beyond just getting a basic beatnote.
 

Additional in-tank shielding:

The large glass windows and inner surface of the vacuum can are high emissivity. We should make two shiny polished masks to go over each end and line the inside of the tank. The inside of the tank can be lined with UHV foil.  We just want to cover most of the inward facing surfaces with shiny material, there is nothing particularly tricky here, designs need to be made and sent out for making.
 
Fixing insulation of the tank:
Presently there is a heating load of 18W to maintain the tank ~25C above room temperature. This is with the tank insulated everywhere except the fittings on the top. For the 50Ω heaters this is equivalent to 600 mA required just to hold close to its operating point.  We should look at where heat might be leaking out and fix it. 
 
There was some argument (way back) that maybe insulation should be limited to ensure there was enough gain on the downward actuation PSL:261, I’m not sure if this is relevant here.
 

Heater drivers:

The TCS heater driver is over the top for what we need.  Something along the lines of what is illustrated below would do the job
I will build a prototype on vector board for +/-10V 200 mA output from 0-10V input, we will order PCBs later if the need arises. The buffer stage could be made up of ganged BUF634s, two of these would give us  0.5 A. However, the resistance of the present heaters is still too high and would require more than 10V with the current tank insulation.

 

Task to add to the CTN workflow:

  • Masking shields to go inside the ends of the vacuum can: shiny with two holes and some method of fixing them in place + Al foil lining with 3x springy expander bands to hold it in place
  • Current driver circuits for 2x heat shields (~200 mA)  and 1x Vac tank (~500 mA)
  • Coated window optics + mounting for 10” CF flange: consult with cryo people
  • Work on improving insulation
 
 
 
Attachment 1: SchetchHeaterDriver_rana_.pdf  66 kB  | Hide | Hide all
SchetchHeaterDriver_rana_.pdf
ELOG V3.1.3-