mass flow rate 2

[Dengr1]

My apologies if this is a dumb question but I can’t seem to see the answer.

If I have a pipe with known length and cross section that is empty on one side of a valve and full of a known substance (Liquid nitrogen in this case) at a known PSI what is my flow rate as I open the valve up at the exit of the pipe.

Assume m(dot-in) = m(dot-out) with no phase change of liquid

 

[IRstuff]

Do you really think that what you've given is sufficient for answering the question?

TTFN

FAQ731-376

 

[Dengr1]

My problem can get very complicated very fast and I did not want to get marred down with assumptions and approximations. I was looking for a general way to solve this type of situation. I really do not know why people must be rude on this site but oh well. Also due to security clearance stuff it is hard to provide all the details. I will do my best to clarify.

I am designing a chill plate for a system that needs to have say 1000W removed. I can assume a known length of pipe or pattern in the chill plate (See attached file). At the inlet of the piped chill plate I have Liquid Nitrogen at say 70 Kelvin. I say “let’s say” because there is a lot of variability in the system. The storage container is located far from my chill plate entry and all the charts I have found just give me properties of liquid nitrogen at 1 atm. My system is pressurized to 50psi which is about 3.4 atm which shifts the boiling point of the liquid nitrogen which will change them temp at which nitrogen will still be at a liquid and not change into a gas. If I have to I could probably put a thermocouple in the system to get the inlet temperature.

The liquid nitrogen is held from entering the chill plate by an on-off valve. Once the thermocouple feedback reaches the set temp the valve is opened and the liquid nitrogen removes the heat until it again reaches set value then the valve is closed. The system is constantly oscillating in a temperature band width. Phase change can happen somewhere between the inlet and outlet of the chill plate. As you can see the chill plate has two zones. The heat generating system does not always distribute the heat evenly across the chill plate. Hope that gives you an idea of the system.
In order to get some sort of base line if I know the following: inlet temperature, liquid nitrogen properties, length and volume of pipe, and flow rate of liquid nitrogen how much heat can I remove. I suspect it will be way more than I need but I still want to check. My question is how I can get the flow rate with only knowing the psi of the liquid nitrogen. Once I get a flow rate value (with the valve open constantly) I can start going back and seeing what pulse time I will need to remove my specific load.

Anyway just thought I would try this forum but if nothing turns up I am sure I will find a solution some other way.

Thanks

 

[IRstuff]

Given sufficient information, you can calculate the liquid flow rate, based on pressure, orifice size, pipe resistance, etc. At some point, it may pick up sufficient heat to boil.

Based on your drawing, it looks to me that your design will have a hot zone in the center of the plate, which would seem to be the exact opposite of what one might desire, assuming your heat source is essentially centered on the cooling plate.

In general, it would seem to me that LN2 is not that good a choice for heat transfer with such high power levels. I'm not sure that you can run enough LN2 through the plate to make it work well.

You didn't include any dimensions, but kW level cooling really screams for water, and lots of it, although that depends on how much temperature rise you're willing to tolerate. A system that will only allow, say, 5°C rise, will need about 4 gpm of water to cool a kilowatt.

TTFN

FAQ731-376

 

[zdas04]

Denger1,
Let me start by saying IRStuff was absolutely not rude. You've been coming to eng-tips for weeks, have started two threads, and responded in this one. You've awarded zero stars and received the same. We see a lot of first-time posters here that ask half a question snark at someone because they didn't do the new guys work for him for free and then they're gone forever. You may not be one of these. You may end up being a worthwhile contributor to this site, but you haven't started off that way, and chastising members for "being rude" is not a strong sign. Your last paragraph in your second post sounded like a threat or a challenge. No one cares about your threats and your challenges may have unintended consequences.

Your problem is that mass flow rate at any point in a continuous stream with no intermediate inputs or outputs must actually be the same as soon as most of the transients have passed. If I slam open a valve into an open pipe, the mass flow rate at the valve will be different than the mass flow rate at that instant 2 miles away. It looks to me like you are trying to apply steady-state rules to transient phenomena. On top of that you know you have a phase-change within the process so most flow correlations are worthless.

The way you described it, the valve is either fully open or fully shut. I've never seen a system "control" that way, it looks like you are pretty much assured of overshooting every target you set every cycle.

As to nitrogen conditions, there are many free or inexpensive tools available to help you understand your phase conditions. Personally, I like REFPROP.EXE from NIST.com It cost about $125, but I use it every day and it has been money very well spent.

If this isn't adequate to help you start in a productive direction, then good luck at answers.com.

David

 

[Latexman]

It appears to me the initial inlet leg is smaller flow area than the rest of the path. This is counterintuitive to assuming a liquid flow, sensible heat problem. Or, maybe I'm making too much out of the sketch.

Good luck,
Latexman

 

[Dengr1]

zdas04

I have tried to use this site for about the last 3 years (machine design, cosmos forums, algor forums, FEA forums, Seal Forums, and now this heat and mass Forum) each group has their own regulars. I have had great success and gotten very valuable help and information but have also gotten very rude and unwarranted comments. Since you are a regular user I am sure you have read many remarks that are hurtful and just plain unwarranted. I use to take offence but now I just realize that some people just don’t care about others and there is nothing I can do about it. Due to the fact that I have had success about half the time I will occasionally try and get some help. The last time I tried to get some help someone like yourself decided I needed to be told off because I accidently posted something in all caps when in fact I just need some help.

My comments were never intended to threaten anybody but simply state if you want to be rude to me I am not going to get angry and I just will have to find the answer someplace else. You are correct though, I should not have not reposted and just chalk this up to another unsuccessful try.

We have over 200 chill plate stands that have to remove a variety of heat specs from different systems. The heat generation is almost never in the center. We use to use water but do to other systems that have to remove much greater heat loads they were switched over to Liquid nitrogen. I have to use the existing set up.

I the end this is why I tried to keep my problem definition simple because of all that I have stated, lots of details, but still do not have an answer to my question.
As for IRstuff if he/she was not being rude my sincere apologies for not seeing their point.

 

[zekeman]

For somebody looking for free engineering advice, your attitude is, quite frankly, unbelievable.

You should be paying upwards of $250 per hour for the advice you seek here and NOBODY owes you the kind of information you seek, especially with your attitude.

Your response to a simple query by IRstuff was simply unacceptable.If you think by telling us virtually nothing about the problem simplifies the problem, well, it doesn't.
If you want answers give as much as you can and let us simplify the problem.

And, in the future, a little humility would be most welcome, although in answering we sometimes aren't as respectful as we should be.

 

[zekeman]

If you could give us an idea of the internal flow cross-section and piping length of this chill plate and the set-point temperature, it would be a little easier to help you.

To minimize N2 usage you must have phase change which will reduce flow rate.

The problem is tractable and looks like a heat transfer and fluid flow problem, which would need a flow area and surface area to allow sufficient coolant to cool the plate.

 

[Dave41A]

Might I suggest a more advanced control system for the liquid nitrogen? It seems like you are trying to control temperature with an on-off type control. If you had a different valve, then you could meter the nitrogen to achieve some semblance of steady-state.

Another idea is to analyze full, 100% flow of Nitrogen. Assuming phase change, what is the rate of heat removal when this is the case? Then divide your true heat rate by the 100% rate to obtain an approximation to the level of duty cycle you'll have to run (i.e. open 40% of the time, closed 60% of the time). As far as how fast you want to cycle between the two, you could probably approximate this system as a lumped capacitance system and compute how long it will take to warm up to the threshold temperature (to trigger the valve to oepn again).

Good luck,
Dave