RF Checklist
This checklist should be totally, unbelievably, almost obnoxiously, complete. Every detail, including making all the tiny cables inside the Frequency Generation and Distribution Boxes and tightening every individual SMA connection, should be included on this page. In this way, everyone will understand what needs to be done, and coordination of tasks will be much easier.
Contents
Things to do before we can consider the RF system complete:
General / Background
Jenne read Alberto's RF wiki page Main RF wiki page () Jenne read Alberto's RF box wiki page Frequency Generation Box
() Read through Alberto's svn pages for the RF system - should be the most up-to-date. See elog 3565
- () Replace the cable-tie scheme to something more rigid. Heliax cables should have strain relief before / near switchover to SMA (near PSL rack).
- () Check out the Sigg RF Design page in aLIGO wiki for stuff, and maybe test procedures.
Frequency Generation Box
- () Make sure it's tested (Alberto says it should be completely tested and characterized already)
- () Phase noise of each freq out
- () Amplitude and phase noise at several (~2) different knob settings.
() Measure Harmonic Distortion (peak height at diff freqs.) using some standard Total Harmonic Distortion (THD) definition.
() Copy Sigg test procedures item 18 (and maybe others), in addition to saving spectra. Repeat for several knob positions. Sigg doc = E1000059-v2 E1000059.pdf (Note dBc = dB with respect to to the carrier.)
- () Remove 2 of the 3 Marconis in 1X2 rack.
- () Install box where the Marconis used to be.
- () Find cable to make power cable. Should have ~4 cores.
- () Make power cable. +18V, GND, +28V.
- () Make cable to connect Temperature Sensor to computers. Temp sensor's Dsub connector should have power (the +/-15V) and the signal out.
- () Make EPICS channels for temp sensor.
- Replace temporary front panel with the nice front panel.
- Replace temporary back panel with nice back panel.
- () Connect heliax cables in a good, non-temporary way.
- () Make new heliax cables, long enough to go all the way to the EOM.
- Should be seriously clamped to the PSL table, no wishy-washy clamping allowed.
- Clamp should include G10 insulation between connectors and optical table.
- Only a short SMA cable should be used to go from heliax to EOM.
- EOM mount and driver, etc, should be insulated from the optical table using G10.
- () Make new heliax cables, long enough to go all the way to the EOM.
Frequency Distribution Box
The parts for this have all been ordered, and are in-hand, but nothing has been assembled. This is going to need to be done very soon, otherwise we can't lock any of the central cavities.
- Examine the Freq Gen Box for methodology of connecting things, particularly electrical insulation while maintaining thermal conductivity. Don't want anything to be grounded to the box itself.
- Gather all components for Distribution Box in one place.
- Cut a piece of teflon to fit inside the box.
- () Prepare standoffs to connect teflon to box (don't attach yet).
- Lay all of the parts out, to decide how they'll fit. There should be a bit more space than in the Freq Gen Box.
- () Figure out lengths and quantities of SMA cables to go inside the box, to connect all the components.
- () Make said SMA cables.
- () Use Mica sheets when attaching heat sinks to box (thermally conductive, electrically isolating).
- () Connect power to voltage regulators (LNVRs)
() Cartoon of what needs to go in the box is in elog 3532
- () Iteratively prototype front panel for the Box.
- () Order a fancy front panel (this will happen as a last thing, after everything else has been finished).
- () Test the box.
- () Output to demod boards should be somewhere between -1dBm and +2dBm for each output.
- () Make a breakout board to fit in rack near Freq Dist(?) Box.
- N-cables will be seriously (not wishy-washily) strain relieved to the rack.
- Short path from strain relief place to breakout board
- "Inside" (non-accessible side) of breakout board will be N-connectors
- Outside will be SMA.
- Short SMA cables will go to input of Box (Freq Gen or Freq Dist??)
- Box (Gen or Dist??) will have isolated SMA feedthroughs.
- Inside of box will be entirely SMA.
Demodulation Boards
- () Upgrade the band pass filters. These are surface mount chips that go into open spots on the demod boards (so we never had band-passing in the past?). Band pass filters are combo of LP (or HP) on the board, and HP (or LP) on the output connector, on the outside. Need a different band pass freq for each demod board. All of these should have already been ordered.
- () Decide on, lay, and connect cables from Distribution Box to the Demod Boards.
LSC PDs
- () Put them together
- () Test them
- () Harmonic Distortion measurement (similar to Freq Gen Box)
- () Optical TF using calibrated Jenne laser (this is the 'real' definitive answer for the TF)
- () Electrical sweep TF (use test input on each PD). We can then use these as the comparison TFs for future diagnostics
- () Power spectra of output - should be 10 or 20 nV/rtHz. Use low noise amplifier for these tests since this is close to analyzer noise
- () Look at time series of output. No DC offsets. No crazy oscillations - must use 500MHz 'scope in order to look at high enough frequencies.
- () Attach cables as necessary
- () Calibrate the Jenne Laser using a 1GHz PD (in order to take Optical TFs of the PDs)
Refl 1 (11 MHz)
(To be used for CARM)
(Should be mostly complete, stored in RF cabinet or under table in IFO room)
- () Take optical transfer function using Jenne Laser and network analyzer, up to ~500 MHz
- () Compare measured transfer function with LISO model
- () Confirm that we have a peak at 11 MHz, and notches at 55 MHz, 22 MHz and 110 MHz
POP 1 (11 MHz)
(To be used for PRCL)
(Should be mostly complete, stored in RF cabinet or under table in IFO room)
- () Take optical transfer function using Jenne Laser and network analyzer, up to ~500 MHz
- () Compare measured transfer function with LISO model
- () Confirm that we have a peak at 11 MHz, and notches at 55 MHz, 22 MHz and 110 MHz
POP 2 (55 MHz)
(To be used for SRCL and MICH)
(Should be mostly complete, stored in RF cabinet or under table in IFO room)
- () Take optical transfer function using Jenne Laser and network analyzer, up to ~500 MHz
- () Compare measured transfer function with LISO model
- () Confirm that we have a peak at 55 MHz, and notches at 11 MHz, 22 MHz and 110 MHz
- () Check elogs. Might need to re-fix bad oscillation peak using same method as AS55.
AS (55 MHz)
(To be used for DARM...acquire at 55MHz, detect at DC)
(Should be mostly complete, stored in RF cabinet or under table in IFO room)
- () Read elog re: oscillation problem with op-amp (MAX4107). Will need to repeat this fix on other PDs.
- () Take optical transfer function using Jenne Laser and network analyzer, up to ~500 MHz
- () Compare measured transfer function with LISO model
- () Confirm that we have a peak at 55 MHz, and notches at 11 MHz, 22 MHz and 110 MHz
POP 22, POP 110 (22 MHz and 110 MHz)
(To be used to measure sideband power. Frequencies are 2*f1 and 2*f2. Would like to make one doubly resonant PD, but might have to split the light and have 2 separate PDs. This is less desirable because we don't have a lot of space on the table.)
This (these) PD has not been designed at all. There might be some LISO action around, but maybe not. Going to need to do these soon.
- () Check if we have any LISO model at all.
- () Make / tweak LISO model to figure out if we can do one doubly resonant PD or if we need 2 PDs.
- () Modify one or two PDs to match LISO model.
- () Take optical transfer function using Jenne Laser
- () Make sure we have resonances in correct places, notches at other frequencies (11 MHz, 55 MHz)
POX (11 MHz)
??? We need these!
Will this be identical to Refl 1 (11 MHz) electronics? Are there any differences we need to take into account, or can we make it completely identical?
POY (11 MHz)
??? We need these!
Will this be identical to Refl 1 (11 MHz) electronics? Are there any differences we need to take into account, or can we make it completely identical?
Other Photodiodes
- () Might have to make 3f PDs. Use 1mm diodes, which are stored in the RF cabinet.
- Refl 33 MHz?
- Refl 165 MHz?
- Refl 55 MHz? Is this one used for 'main' IFO, not 3f?
- AS 165 MHz?