Thermal mass flow meter for fuel gas measurement 1

[FIG Journey]

I have been reading that a thermal mass flow meter does not require any pressure compensation for accurate flow measurement in fuel gas applications. Can anyone verify this?

Thank you.

 

[zdas04]

A Thermal mass flow meter relies on the momentum (i.e., the product of the velocity and the density of the fluid) to carry heat away from a heated probe. The software then compares the temperature of a reference probe to the temperature of the heated probe and the supplied current to the heated probe to convert that to a rate of heat transfer that can be related to a mass flow rate. Since density is a function of pressure and temperature and velocity is a function of pressure and temperature, I believe that you interminable quest to save a few hundred dollars on pressure and temperature instruments has failed yet again.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[controlnovice]

FIG Journey:

Compared to the tyipcal gas flow measurement device: dP measurement...no, Thermal Mass Flowmeters do not need pressure or temperature compensation.

What kind of pressure swings are you expecting? Thermal Mass Flowmeters generally do not need pressure compensation. They do need some information on the type of gas to know its "heat capacity"...or the ability to take heat away.

Sage
"Is pressure or temperature correction needed for thermal mass flow meters? No. Thermal mass flow meters replace all of this with one instrument that directly measures the mass flow rate of the gas."

https://sagemetering.com/back-to-basics/thermal-mass-flow-meter-correction-factor/

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[zdas04]

controlinvoice,
You really need to take that statement from Sage Instruments with a very large helping of salt. There are a significant number of assumptions that go into the calculation (not the least is that the probes have not fouled, which is often an invalid assumption), and if all of the assumptions are met then they do a good job. Density is an input, not a measured parameter. Deviation of pressure, temperature, or gas composition (e.g., more or less water vapor can change the density 5-10%) causes wild swings in density. Just the day-to-night ambient temperature swings can be enough to mess up a small fuel-gas meter.

You cannot measure fluid velocity. Period. You can measure the effect of velocity on the environment (i.e., how fast does a wheel spin, how much dP is there across a known restriction, how much does a heated probe cool, etc.), but that is measuring an effect that could be caused by changes in velocity or changes in another parameter. The instrument DOES NOT "directly measure the mass flow rate of the gas", and the engineers at Sage know that very well while the marketing people do not care.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[itsmoked]

Lots of fuel gases are mixtures that change so the density is going to change constantly.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[controlnovice]

Thanks for the clarification...

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[John Good]

David, THANK YOU! You have restored my faith in humanity. This has been a pet peeve of mine for years.

I once sat on an ISA panel titled, "Measuring Mass Flow - Direct vs. Inferential". I was representing DP,which was classified as "Inferential". Thermal mass and Coriolis were classified as "Direct". Newsflash - Every flow meter, mass or volumetric, is inferential as you so eloquently explained in your post. Thermal mass INFERS mass flow through delta T and the Coriolis infers mass flow with the Coriolis effect (vibrating tube).

The only true direct mass flow method is to weigh it vs. time(which is actually done at some gravimetric flow labs (opposed to them master meter method).

Needless to say, I wasn't too popular on the panel when I questioned its subject title before we even started on the technologies.

Again, thank you.

John

 

[cavbilly]

OK, what about positive displacement flow meter where the output pressure is ambient (yes, I know barometric pressure varies slightly), and the temperature must be recorded with the flow through the instrument? Would that be absolute, or inferential?

 

[zdas04]

cavbilly,
You didn't state what type of PD meter you are referring to. If it is a nutating disk, then the instrumentation is counting the number of times the disk rocks. Rotary vane is counting revolutions. Diaphragm is counting up/down cycles. The lobe-type meters are counting revolutions. All of them are correlating the transition of a volume of space to a volume of fluid assuming the density of the fluid is known and constant and that the volume of space is totally filled each step. Neither assumption is terrible, but neither is 100% accurate 100% of the time. Gas in a PD meter is especially problematic because the density of the gas changes so dramatically with small changes in pressure and temperature (to drag the discussion back to the need for temperature/pressure compensation).

YOU CANNOT MEASURE FLUID VELOCITY, ONLY THE IMPACT IT HAS ON THE ENVIRONMENT. Therefore all fluid measurement is inferential. Any manufacturer that says differently is either (1) mistaken (unlikely); or (2) lying (way the most common).

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[cavbilly]

I was thinking about Hyde flow meter calibrator, where the volume included in each rotation is very well known. The density of the gas is inconsequential in Hyde meter, only the rotation matters. OK, well if the "gas" had the same fluid density as the liquid being displaced from each quadrant cavity during rotation were the same, then it might not separate and displace.
These are also called, "wet-test" meters.

 

[IRstuff]

"where the volume included in each rotation is very well known"

That's irrelevant if you're trying to measure MASS flow; you still need to know density, and to know density accurately, you need temperature and pressure.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[zdas04]

I'm not familiar with the Hyde instrument (and neither was Google). But you say "the volume included in each rotation is very well known" so I'm going to step out on a limb here and assume that the SPACE included in each revolution is known and the "flow" is a count of revolutions. Measuring angular velocity and converting that to a mass flow rate. Just like all of them. The meter has no way of knowing if the "known space" is full each revolution or if the density of the fluid is suitably close to the density in the database. It just knows that it moved a fixed space from one side to the other. Not really "direct measurement of flow".

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[cavbilly]

Since the gas is being measured at atmospheric (local) pressure, one merely needs to convert to absolute pressure (correcting for elevation above sea level mainly), and the instrument also has a thermometer in every case to allow correction of the density to standard density.

There is no meter that directly measures "mass flow", since each so-called mass flow meter has a calibrated response to mass flow, based upon a physics principle of the mass flow interacting with a process in the meter, such as radiation, etc.
Hyde wet-test meter usually has an even number of chambers (4 is most common), and gas comes in on a port just above the hub of the rotation shaft holding the curved chamber dividers. As gas enters a chamber, it reduces the amount of water (or other liquid such as a light mineral oil in some cases) in the chamber by the volume introduced, and this results in a torque on the shaft. Simultaneously, the gas on the opposing chamber is being displaced out the exit slot by the liquid fluid during rotation. It is pretty simple, but not so easy to fabricate, although several companies sell them.

https://www.elster-americanmeter.com/assets/products/products_elster_files/Wettest.pdf

[/url]
or just google "Hyde wet test meter" and see what pops up?

 

[FIG Journey]

OK, pressure compensation it is. You guys are some smart cookies!

Thanks.

 

[cavbilly]

Pressure compensation depends on the ratio of as measured pressure to one standard atmospheric pressure (as a standard state).
Temperature compensation depends on the ratio of as measured temperature (absolute °K), to standard reference temperature of 273 °K (0 °C).
Let Q = corrected flow rate as mass/time units.
and Q' = as measured

Q = Q' x P(meas)/P(std) x T(std)/T(meas)

suppose Q' = 205 LB/Hr , P(meas) = 20.2 psig (measured at sea level to avoid elevation correction), and t = 104.8 °F

First we convert P,T to unite consistent with reference standard conditions.
P(meas) = 20.2 psig x 1 atm /14.6959487823 psi + 1 atm (to convert to absolute pressure)
P(meas) = 2.37 atm (absolute pressure)
T(meas) = (104.8 -32) x 5/9 + 273.15 (convert to °C then add 273.15 for °K)
T(meas) = 313.59 °K

Q = 205 LB/Hr x 2.37/1 x 273.15/313.59
Q = 423.2 LB/Hr, 191.96 Kg/hr, 3.20 Kg/min

 

[zdas04]

Cavbilly,
There really isn't any meaning to a "pressure compensated" mass flow rate. Mass is mass. We pressure and temperature compensate volume flow rates. Your adjustment would be correct if the input was a volume flow rate at actual conditions. The symbol is usually "q" (most people reserve "Q" for energy flow rate).

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[cavbilly]

OK, maybe I was off-topic in referring to volumetric gas measuring equipment.
True, with mass-flow instrument, these P,T corrections/compensations are not required.

 

[zdas04]

cavbilly,
I don't mean to bust your chops, your answers just lead me to pointing out problems that I see across the industry across the world every day.

If it were possible to measure mass flow, then pressure and temperature compensation would not be necessary, but (in spite of manufacturer's brochures to the contrary) it simply is not possible.

If you had a volume flow rate that was going to be used in isolation and never aggregated with any other volume and in a perfectly constant pressure/temperature environment you wouldn't need pressure/temperature compensation.

But the world of a volume in isolation and constant conditions does not exist. Someone will always add the 15 psig instrument control gas to the 145 psig engine fuel gas to the 30 psig burner fuel. Then they will add that meaningless number to the standard volume at the sales meter. 2 apples plus 9 Volkswagens plus 2 shovels plus 200 MSCF really doesn't equal 211 MSCF.

That observed fact is why I committed three pages of my 680 page book to "standard conditions", it is something that "everyone understands" and people mess up more often than any other "simple" concept.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist