EPD FEE temperatures pre-run 2024
Updated on Wed, 2024-04-17 10:33. Originally created by lisa on 2024-04-17 09:57.
Context:
As discussed on this page, the temperatures in the FEE box are of some concern and may (not clear) increase the dark currents. Three observations were made:
How the cards heat up:
The only way to power down the cards is to turn off the Power Supplies. Pushing `OFF' on the GUI only turns off the SiPM bias, not the cards. So, I had the shift crew go out and power down the supplies and I waited for several hours (overkill surely) for the cards to cool off. Then I had them turn it on and quickly did a 1-Wire Scan on the EPD GUI. Then I saved manually (keeping track of time and pressing the buttom) the temperatures to a file every minute or so. Pressing `SaveAs' on the bottom of the GUI will store the FEE values to a file on your local computer.
I wrote a little macro which will shows the temperatures as a function of time. It is attached at the bottom of this page. Here are two measurements. Figure 1 is the nicer one.
Figure 1. Morning of 17 April 2024. Here, the measurements started within a minute of the P.S. being turned on. (I don't know why the West side temperatures did not seem to update, but there were some TUFF errors around that time.) It nicely shows that all temperatures start at the same value, and then diverge as the cards heat to different asymptotic values. It also shows a characteristic equilibration time on order 15 minutes.
Figure 2. Evening of 16 April 2024. Here, there was a few minutes delay after the PS were turned on, and the measurements started. Also, I did not record the temperatures as frequently. I am mostly showing this to show that the West boxes do read out usually.
Context:
As discussed on this page, the temperatures in the FEE box are of some concern and may (not clear) increase the dark currents. Three observations were made:
- The temperatures were different for different FEE cards, and the pattern clearly indicates that it is a function of the position of the FEE card relative to the cooling fans in the box.
- On a warmer day (15 April 2024), the temperatures were higher (30C) than on an earlier day (5 April 2024, 27C). In the mornings, the temperatures in the box are often only 22C. This is relevant, as STAR hopes to run through the summer.
- The temperature does not increase when bias is turned on, so it is not a function of dark current, just as a functions of the Power Supplies (on the racks) being on.
How the cards heat up:
The only way to power down the cards is to turn off the Power Supplies. Pushing `OFF' on the GUI only turns off the SiPM bias, not the cards. So, I had the shift crew go out and power down the supplies and I waited for several hours (overkill surely) for the cards to cool off. Then I had them turn it on and quickly did a 1-Wire Scan on the EPD GUI. Then I saved manually (keeping track of time and pressing the buttom) the temperatures to a file every minute or so. Pressing `SaveAs' on the bottom of the GUI will store the FEE values to a file on your local computer.
I wrote a little macro which will shows the temperatures as a function of time. It is attached at the bottom of this page. Here are two measurements. Figure 1 is the nicer one.
Figure 1. Morning of 17 April 2024. Here, the measurements started within a minute of the P.S. being turned on. (I don't know why the West side temperatures did not seem to update, but there were some TUFF errors around that time.) It nicely shows that all temperatures start at the same value, and then diverge as the cards heat to different asymptotic values. It also shows a characteristic equilibration time on order 15 minutes.
Figure 2. Evening of 16 April 2024. Here, there was a few minutes delay after the PS were turned on, and the measurements started. Also, I did not record the temperatures as frequently. I am mostly showing this to show that the West boxes do read out usually.
»
- lisa's blog
- Login or register to post comments