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: 1848     Entry time: Thu Aug 3 01:24:05 2017
Author: awade 
Type: DailyProgress 
Category: FSS 
Subject: Wiring up FSS boxes into acromag cards 

To date we have been manually engaging the TTFSS boxes for the PDH locking of the lasers to reference cavities.  There is a sub-D 25 pin output on the back that can be used with binary I/O for switch engage and analog voltage inputs for variable gain/offsets.  As soon as we install the mode cleaners the lock of FSS loops will be conditional on mode cleaner state. As Aidan packaging the MC electronics up with Acromag controls, it makes sense to also do the FSS loops. All the control will then be python scriptable.  

In the far future it might make an interesting project for an undergradeuate to look at applying neural networks etc at this relativity simple locking problem. If we have a ready to go system with access to all the controls, structuring a small project around this might make sense.

After digging around the pinouts for the D25 can be found on the TTFSS Interface Power Board: https://dcc.ligo.org/LIGO-D0902051/public

To engage the lock both  RMPEN and T1EN need to be pulled high.  Here ramp closes the loop at a number of different points on the box (it is the main engage fo the loop) but the T1EN when left low defaults opening the loop on the RF board (mixer+LP filter) and instead engaugs the test SMA on the front panel (note, this is not the common path transfer function test point).  So to engage the loop both have to be held high. A total of 4 binary I/Os are required.  Three analog voltages signals are needed to set the common gain, fast gain and offset. 

TTFSS sub-D 25 pin out connections
Control +ve -ve Load Type
T1EN (RF board test point, switches out mixer for SMA)* 23 13 (GND) 820 Ω Binary (+5V)
RMPEN (Ramp loop engauge)* 11 13 (GND) 820 Ω Binary (+5V)
T2EN (Common test point engauge)* 24 13 (GND) 820 Ω Binary (+5V)
AOEN (offset field box connector engauge) 12 13 (GND) 820 Ω Binary(+5V)
Common gain 1 14 NA Analog (+/- 5V)
Offset 2 15 NA Analog (+/- 5V)
Fast gain 3 16 NA Analog (+/- 5V)
Remote 25 13 820 Ω

Binary, not used


*Note: some of the binary engauge states are actually to switch out the test point, so the logic may be inverted. All this can be changed in software

Also, important: there is a 10 kΩ pull up resistor in the acromag unit binary I/O channels that is too high when combined with the TTFSS box impedance to pull the state low. Aidan already found a fix to this problem (PSL:1573) by tying a 820Ω resistor in parallel with the acromag 10 kΩ to pull down to 0.669 V when set to off.

I've hooked up 4 binary I/O channels and 3 analog channels and mapped these to soft epics channels in FSScontrols.db for the south path.  The loop engages and the gains and offsets can now be tuned from a medm channel.  With Jamie's python tools we can also now script them.  Also with ssh -Y forwarding its super convenient for locking from wherever in the lab.


I've ordered more acromag cards that are due to arrive Friday August 4.  Hooking the rest of this up for the north path is low priority.

ELOG V3.1.3-