Laser Tuning
LC noted that the output beam quality of the laser (
figure 1) was poor and not as good as it had been before (
figure 2).
Figure 1 : Current spatial profile of the laser taken after the regen Nov 2010.
Figure 2 : Previous Gaussian-like profile of the laser taken after the regen in Nov 2007.
Figure 3 : Beam profile out of oscillator for comparison.
Oscillator notes
The oscillator needs cleaning fairly regularly (maybe every month or so) with ultraclean acetone, although don't clean the small SESAM at one end of the cavity. This usually increases the power by 30 - 50mW. If the oscillator won't lock to the external frequency it might be so far away it needs tuning to the reference. Use a spectrum analyser and tune the large screw on the SESAM mount to move the laser frequency close to the reference. If the lock is noisy, sometimes it is possible to reduce the noise by very very carefully tweaking the vertical adjustment on mirror M1 (opposite end of the cavity to the SESAM) but be careful that the laser beam doesn't separate into two or spots when this is done. This can be monitored using a green ir card held after the output coupling dichroic mirror in the cavity. If you tweak M1 in either plane the beam starts to separate into multiple spots, which should be avoided. Tuning this mirror for a less noisy lock should not involve moving it enough to see this effect though.
Regenerative Amplifier (RGA)
Energy
During the visit of LC, LN in Sept 2010 the regen output energy was ~ 3 - 6mJ, well under what it should have been (~ 16mJ). The Coherent engineer restored this energy output while noting two problems - the damaged etalon and a damaged turning mirror between the seed and the regen. The mirror was taken out and replaced with one of the CVI Nd:YAG ir mirrors (Y1 coating). The damaged mirror had a crack at one side which didn't really affect the light transport of the seed but as we have plenty of 1" ir mirrors it seemed sensible to change it to avoid any problems in the future e.g. the damaged mirror breaking. The etalon is discussed below.
Notes on tuning regen for energy
There are several parameters that can be changed - do one at a time. The energy is very sensitive to the timing between the firing of the flashlamp and the SDG box that controls the Pockels cells - need to inject a pulse into the cavity when sufficient gain has built up but before it decays. This timing is controlled by the light blue TD2 module at the left hand side of the crate. The best way to optimise this is to get the regen lasing with PC1 and PC3 on then alter the timing until the comb of pulses is as early (to the left on the scope) as possible. PC3 should be around 18 - 20ns behind PC1. This gives a clearer train of pulses. Optimising the cavity mirrors, etalon angle and seed injection into the cavity is the same - push the build up time (pulse comb) early. The switching out time (i.e. PC2 timing) should be ~ 100ns or less after PC1. If it is a lot more than this it means the cavity is not optimised and the pulses are taking too long to build up inside the regen. It should run stably at ~ 15 - 16mJ when everything is aligned properly.
Etalon
It was noted by the Coherent service engineer that the etalon in the regen was damaged as shown in
figure 3, but he didn't have a spare to replace it with. This may affect the beam quality.
Figure 3 : Etalon from RGA showing several damaged spots.
LC cleaned etalon but the spots are definitely damage not dirt. The etalon was replaced in the regen (NB do not run regen without it - the short pulse length will damage the laser rod) but with four washers under each corner of the etalon mount to raise it up. Now the beam through the regen cavity passes slightly below the damage spot on the etalon. The beam quality has not improved though - is something else in the cavity damaged or dirty? LC optmised regen cavity (angle of etalon, both cavity mirrors, injection angle of seed) and by midday 20101115 the regen was running at ~ 17mJ output i.e. slightly more than before.
Linear Amplifiers
The lower green output of the laser around 19/11 is not poor SHG conversion efficiency (crystal orientation has been optimised). There is less energy after the linear amps than previously. Measurements after amps on 28/3 are:
- ir 200μs = 88.4mJ
- ir 160μs = 312mJ and
- green 160μs = 109mJ.
Measurements 19/11 are:
- ir 200μs = 42.2mJ
- green 200μs = 7.42mJ
- green 160μs = 40.6mJ
The energy gain through the linear amps has dropped since March. The output of the regen is nearly the same in both cases (March 2010 18.4mJ, November 2010 16.5mJ) so the problem is in the amplifiers. Have tweaked the alignment through the amps on the camera and optimised for energy but this gives ~ 40mJ. Suggested issues are either that the poor spatial quality is not extracting the gain efficiently (but the beam is not that much worse than in March), or maybe the flashlamps are getting old - not sure when they were last replaced. If this is the case the energy will continue to drop - needs monitoring. One way to deliver more energy to the IP is to go to shorter flashlamp delays to get higher ir energy and thus more green. In the logbook it is recommended that the ir energy before the SHG crystal does not exceed 600mJ to avoid damaging it. Need to do a check of ir energy v. flashlamp delay with the attenuator on the energy meter for current setup to see what is the shortest delay that will not damage the crystal.