Does an orifice control mass or volume flow? 5

[jennyrabbit]

I am trying to understand the effect of ambient pressure on venting a gas from a pressurized vessel. I understand that conductance of an orifice has units of torr-litre/sec. Does this mean that if there is a pressure difference of 1 psi, regardless of the absolute value of the input/output pressures, the same mass will flow? That is, if I hypothetically pressurized a vessel to 14 psia and vented it through a 1/16 inch diameter hole into a 13 psia ambient, the mass flow rate would be the same as if I vented that same vessel from an initial internal pressure of 4 psi into a 3 psi ambient? If this is the case, then an orifice controls mass flow. I'm having a hard time visualizing this as I am not a fluids person. I'm assuming that the gas is incompressible in this case, and flow is viscous. Any insight much appreciated.

 

[zdas04]

Greg,
People use Volume Flow Rate at Standard Conditions as a surrogate for mass flow rate and it has all the good features of mass flow rate (i.e., if velocity is not constrained then changes in mass flow rate are not a function of local density changes). If density goes up with constant area, then local velocity will go down by a proportionate amount and mass flow rate will stay constant. But, you must know both density and area at some point to be able to determine your mass flow rate and therefore your volume flow rate at standard conditions.

Volume flow rate at actual conditions is pretty much worthless for calculations with dramatically changing pressure or temperature. The only place I ever use it is in empirical compressor hp calculations.

David

 

[BigInch]

Condensed matter is beyond my scope, so can you deal with a short discussion from the Lattice Gas Cellular Automata-CFD perspective?

http://rapidshare.com/files/33908432/Lattice_cell_gas_flow_CFD.pdf.html

http://virtualpipeline.spaces.msn.com

 

[Latexman]

jennyrabbit,

This seems to be pertinent to your post:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19830015079_1983015079.pdf

Good luck,
Latexman

 

[jennyrabbit]

OK, what I'm hearing is that if I measure a mass flow of 1 g/sec flowing across my orifice for a fixed upstream pressure of 15.7 psi and a downstream pressure of 14.7 psi, THEN I expect to get 1 g/sec flowing across my orifice with an upstream pressure of 5.7 psi and a downstream pressure of 4.7 psi. I seem to be hearing that a fixed orifice, like a hole, allows a certain mass flow depending on the pressure differential across it. Period. Any dissenters?

 

[jennyrabbit]

Thank you, Latexman. I had a copy of that and had been so frustrated with the poor copy quality, but I did take another gander at it. Alot of great stuff in there....although "area" still has to be determined for any vent which is not a straight shot hole.

 

[Latexman]

You are welcome. When you did not say you had looked at or downloaded some material from NASA the first time I mentioned them, I just had to do a quick search.

If you got 1 g/sec from 15.7 psi to 14.7 psi (29.4/30.4 = 0.967), should you expect 1 g/sec from 5.7 psi to 4.7 psi (19.4/20.4 = 0.951). Without doing any detailed calcs, I'll say no. You are not near sonic velocity, so that's not it. You'll get < 1 g/sec from 5.7 psi to 4.7 psi due to density differences (19.4/29.4 = 0.66).

Good luck,
Latexman

 

[jennyrabbit]

I agree with you, Latexman. What makes me kind of crazy is that Scialdone in that reference you gave me gives a dependence of dP/dt on sqrt(deltaP * P). Analytical equation for flow through a simple orifice shows a dependence of dP/dt on P and sqrt(expression of ratios of upstream and downstream pressures). My empirical results give dP/dt depending linearly on P and a power law dependence on delta P (close to sqrt, not quite). At least I'm getting clearer on what is not clear!

 

[Latexman]

Did Scialdone assume isothermal expansions/conditions in that article? If so, his sqrt(dP * P) is equivalent to sqrt(dP * rho) 25362 mentioned above for orifice mass flow rate equations. If not, then I'm as lost as you.

Good luck,
Latexman

 

[FACS]

So, if this thing launches, as you say; won't there be a temperature difference, and therefore a change in density?

Charlie

http://www.facsco.com

 

[jennyrabbit]

Scialdone assumes an isothermal process as the launch is pretty fast (order 100 seconds). And I misspoke when I said that an orifice had a different dependence. I was looking at dependence for a circular pipe.

 

[crysta1c1ear]

Does an orifice control mass or volume flow? I am trying to understand the effect ...on venting a gas ...

[highlisht]Massflow.[/highlight]

I'll keep it simple jennyrabbit.

eg if 3 kg/s of gas molecules were to hit a sealed orifice from the left and 2 kg/s were to hit the sealed orifice from the right, then what happens if the seal were to magically disappear?

3 kg/s of gas would pass from left to right, and 2 kg/s would pass from right to left. We would see that as net flow of 1 kg/s from left to right.

 

[johnp]

Hiya,

Don't ever confuse mathematical theory with reality. The reality is what you are dealing with - the imperfect gas. If it is natural atmosphere it will contain moisture, as well as a mix of Nitrogen, Oxygen, and other gasses and/or pollutants. That will have pressure, density, and humidity (which can cause internal corrosion).

By some fluke of nature most of the gasses approximately follow Bernoulli's equation. If you assume the gas to be "incompressible" it will act more like a liquid. Then you can make more simplifying assumptions. That approach is usually used in the design of Air Conditioning and Ventilation applications.

When you start looking at viscous effects, then the gasses approximately follow Reynolds equation, and viscous effects are approximated by the effects of Reynold's number. In that case you can use the Darcy Weisbach formula, together with the Moody Diagram. By some fluke of nature this is applicable in a general way to both liquids and gases, provided that your assumption of "incompressability" is adequate. The tables published in the ASHRAE Guide, and other air handling text, are based on an interpretation of the Moody Chart.

The same approach is used in the published tables for the flow of compressed air and other gases. Steam tables, however, are used to allow for the effect of partial pressure of water vapour in the air, and hence a "humidity ratio" - i.e. what part of the volume of gas follows the steam table.

When you put a magnifying glass on the flow effects, you have to ask - "Does the gas, in the range I am interested in, follow Newton's Formula, or is it non-Newtonian?"

You see - the real world is much more amazing that your engineering Professors would have you believe. Everything in nature is calculable. But do your numbers make sense in the range you are interested in? Most suprises come from making assumptions which ain't necessarily so.

Enjoy life - that's what it is all about.

Regards - Sgt John.Rz

http://www.latviantourists.com.au

Johnp.Rz

http://www.mets.net.au

 

[sailoday28]

johnp (Mechanical)
If as I understand the problem, the real gas has a max pressure of 1 atmosphere. Most gases will follow perfect gas behavior at those conditions. Wheter specific heats can be considered constant can be questionable, dependent upon temperature variation. With regard to being a mixture of "gases", most gases that I am aware of will follow Dalton's Law of partial pressures. Therefore the physical mathematics apply. Whether the process is isothermal or quasi-static can be questioned.
My experience is that for a fast process, heat transfer is not fast enough to maintain an isothermal condition. Further for a fast process, a transient model may be more applicable than a quasi-steady one.

An engineering professor who fails to include real world effects in analyses should not be in the teaching profession.

 

[BigInch]

I thought that's why they were.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch (Petroleum) 3 Jun 07 2:21
I thought that's why they were.

That has not been my experience.
Are you speaking up for your schooling?

 

[BigInch]

When you are trying to be humorus on the internet, due to the nature of this form of communication, it is generally accepted convention to use a smiley face -)), otherwise somebody might think you are making a personal attack or something. Do you know how to make a smiley face? If not, I would suggest you learn how to make one... quick.

http://virtualpipeline.spaces.msn.com

 

[jennyrabbit]

My philosophy is to try to understand the theory and combine that understanding with experimental results specific to the problem I am trying to solve. The posts have provided me food for thought and I am now better able to understand the problem and have been getting experimental results which I am using in my model. Heck, we have all had good profs and terrible profs...! Thank you, everyone, for your help.

 

[crysta1c1ear]

You may never have more than 1 psi ?P, but as the telescopes approach 0 psia won't the mechanism change to molecular flow?

jennyrabbit (Materials) 26 May 07 23:06
You are correct, Latexman. But at that point, I'm not worried about the telescope blowing up. ; )

In my application, I will never have more than 1 psi difference between upstream and downstream pressures.

So test the telescope with a bigger pressure difference than 1 psi and if it survives that without blowing up, or distorting (or whatever you suspect might be a problem), then it should be okay, shouldn't it?

Or am I missing something?

My understanding so far is that the flow rate is a worry because if air doesn't leak out fast enough there would be a pressure difference between the inside and outside which could cause problems.

Yet you also seem certain that the maximum pressure difference will not be more than 1 psi. (I don't know how you can say that, but I'll trust you.)

So if you test a 1 psi difference, you will have tested the worst case won't you?

 

[jennyrabbit]

It would be quite impractical to test the telescopes at 1 psi pressure. Mechanical designers specify maximum pressure and provide that to me. Reasonable suggestion but not really feasible or necessary.

 

[sailoday28]

BigInch has stated
An orifice, like a control valve, does not control either flow or pressure. Both allow a certain passage of mass based on the pressure differential... nothing else.

Not so for choked flow.