Yesterday I got a call from Ed that the TCSY flow rate was reporting under 2gpm, lower than the usual 3gpm (alog51806). While the flow at the flow meter was reporting lower, the chiller was still reporting its usual flow rate, ~4gpm. I was immediately concerned of a leak, but Ed found no water missing in the chiller. We are still full today as well.
My next suspect was the flow meter. Our water in the system is not the cleanest, and we have had issues in the past with the paddles in these flow meters getting gunked up (see Peter's alog32777 for pictures). This particular flow meter was replaced December 20, 2016, for similar issues, and the TCSX flow meter was replaced October16, 2018. The situation looks a bit different now because we have less of what we have been calling glitches in the flow. The difference is that yesterday and on the 5th we had this flow meter report the flow as ~1gpm lower for a few tens on minutes, before returning to normal. Meanwhile, the chiller flow doesn't seem to be affected. The laser temperature goes up slightly when this happened, about 0.5C, but this has been happening more frequently when we replaced the chiller.
The current, running TCSY chiller is actually our spare. We swapped it out on January 15, 2019 due to a leaky shaft seal. The flow on the current chiller seems to be less consistent than on the normal TCSY chiller. As Jason notes in alog50789, this may be part of the reason we have had more relocks from this laser. The normal chiller has be repaired with a new motor and pump, but still needs to be tested.
Going forward:
It definitely looks like we need to swap out and rebuild the TCSY flow meter soon. This isn't hugely invasive, but runs the risk of breaking pipes because the meter is dry-fit into the pipe and each time has been more difficult than expected. Because of this, we are hoping to push this off till the October vent, when we have more time. In the mean time we would like to swap the TCSY chiller as soon as we can get a long term test on the backup unit. The inconsistency of the current chiller can't be helping the flow problems, and it's an easy swap.
Attachment 1 - The flow rate from the flow meter, the flow from the chiller, the laser temp., and the laser output from the last few days.
Attachment 2 - Same channels as above, but in December, before the chiller swap.