[Hallb-engineering] [New Logentry] Follow-up Re: Follow-up Re: Hall B Torus and DBX Nitrogen System

kashy at jlab.org kashy at jlab.org
Fri Aug 14 08:50:03 EDT 2020

Logentry Text:
This entry is intended to explain the "Batch Filling" of Hall B LN2 Tranferliine.

After turning off HTR8554 we can see that the time between high LN2 flow to Hall B LN2 system goes up. With 30% heater on we required 3 fills in 46hrs  (`15.5 hrs), but with the heater off the time between refills increased to 2 in 44 hrs (22 hrs). See Plot 1. This makes sense because we add 600W of heat load to the system. (Actually we could do a detail study and determine the LN2 consumption of Hall B including its distribution system by taking data at a few more heater power levels.)

After the heater was allowed to go to zero I did a little extra tweaking of controls was done to see if I could smooth the level of the LN2 reservoirs in the DBX (distribution box) and Torus, but these resulted in marginal benefit.

Figure 2 shows the LN2 flow rates of all 3 Halls (CFI6711A, B, C)   and it appears (to me, other Hall operators who have a closer/deeper view please chime in) that having the heater on was of no obvious benefit to Hall A or Hall C Nitrogen stability

Finally, figure 3 shows most clearly the transferline refilling. This is a non tiled plot and all signals are plotted on a scale of 0-100 (except the pressure PT8552 on a scale of 0-10).

Note for future discussion:
EV8553 (Cv=3.0) is the LN2 supply valve to the reservoir. Any flow that does not go through that valve goes directly to the Torus LN2 shields through EV8555T. (Cv=0.27). 

So what is going on?
At ~11:30 EV8553 begins to open and the level drops below 54% on LL8554CP (capacitance probe) which is the liquid nitrogen reservoir in the DBX. 
Notice that the level drops faster than it was prior to the valve opening. This indicates that warm gas, not liquid is entering the DBX and the level drops for 21minutes until the transferline delivers liquid to EV8553. 

At 11:48 the flow rate measured at ESR CFI6751B peaks out at 29g/s (max it can read, not the real max). This is 14 minutes after the valve started to open. 
AT 11:54 the liquid gets to EV8553 and During the next 9 minutes LL8554CP rises and EV8553 valve closes.

The best indicator that the transferline is still filling is described next!

For 33minutes after EV8553 is on its minimum the flowmeter to the Hall remains pegged (>29g/s)!

How can the flow go in at this high flow rate and not go out through a valve? 

Lets consider what is in the transfer line!
It is either COLD gas or Liquid nitrogen 
If it were Liquid, then the flow would have to stop when EV8553 closes because there is no place to put more mass (LN2) in the system.
If it was gas in the TL then mass could be added by increasing the density in the volume of the transferline. 
How much volume does the TL have?
Lets calculate it!

The LN2 flow circuit is an annulus. The outer pipe is 8NPS sch 10 so 8.625"OD with 0.148" wall so 8.329" ID
The inner pipe is 6NPS so 6.625" OD

Do the math and you find the area between these is 20square inches
The length of the TL is approximately 160ft or 1920inches
So the volume of the line is 38,400cuin = 630liters (not small!) (The LN2 reservoir volume in the DBX is 78liters)

The density of gas nitrogen at  3 atm is .013g/cc
while the density of liquid nitrogen at 3 atm is 0.76g/cc

The most mass one could add to the system would 630liters x1000cc/liter x (.76-.013)g/cc = 470,600grams

Looking back at plot 3 the full flow in the TL is happening for approximately 48minutes = 2880seconds

Dividing 470,600/2880 would be a flow rate = 163g/s


How can liquid get into the transferline if it is full of gas at the same pressure? Where is the gas going if the exit valves are shut?
For those of you who read the original post in this thread I described a subcooler in the ESR Valve box.
The answer to where the gas goes is not simple.

The gas actually turns to liquid. 


The temperature of that gas is reduced by the incoming subcooled liquid to below the saturation point and it condenses into liquid  thus allowing the flow into the transferline without flow leaving!



This is a plain text email for clients that cannot display HTML.  The full logentry can be found online at https://logbooks.jlab.org/entry/3827230
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/hallb-engineering/attachments/20200814/755d3c07/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: application-pdf.png
Type: image/png
Size: 346 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/hallb-engineering/attachments/20200814/755d3c07/attachment.png>

More information about the Hallb-engineering mailing list