Relative density vs specific gravity = ? pressure

[danthemeterman]

One of the previous posts reminds me of a situation I ran into last week. Is there a chart which could at least loosely suggest a nominal density of propane based on visual indicators of pressure and specific gravity. I am for the 1st time testing a mass flow meter, we can not get repeatability after making changes to correct the errors seen in a volumetric, small volume or loop/ballistic test. I am suggesting the counter (registering indicator) is at fault, just my gut instinct and 32 years of meter calibration an repair. My experience tells me that 99.8 % -+ .505 specific gravity propane at 35 Deg. F. should be approx. 60 PSIG. If this isn't enough information, let me know and I will try to give better examples to anyone in the refining industry.

 

[zdas04]

Acording to REFPROP, the saturation pressure for 35F is 72.383 psia. At 5300 ft elevation that is 60 psig. I'm not sure how you know specific gravity (I'm assuming you mean specific gravity of the liquid relative to water), but it doesn't seem relevant.

I would approach this as a simple saturation problem and get the temp and use a saturation table to get pressure. Or measure pressure and use the table to get temperature. One thing is pretty certain, at most ambient temperatures propane is going to be saturated.

David

 

[BigInch]

Isn't that exactly why you changed to a mass meter? Now that you have mass, I'm not sure why you want to convert to volume; proving the mass meter?

You can calculate a density at a given temperature and pressure for propane. How accurate that would be depends on the equation of state you use and the 0.2% of the gas you don't know what it is, so I would think density would be better supplied by an explicit measurement from a sample.

What's the question?

Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[danthemeterman]

OK, I was hoping a simplified question would suffice. You have more knowledge of the chemical properties than I do. I didn't believe the mass meter's reading of .5535 +- density. My experience is listed above. The reason for measuring volumetrically is my competition was in with their small volume provers, they were not able to get an acceptable repeatability. I do not walk away and tell the customers they need to fix their system with out offering a solution. This mass meter has been in place for 4-5 years and has never been right. They consistently lose inventory. While testing for gross output we were able to get it to repeat after changing the specific gravity in the register, which for arguments sake, I considered to be a problem in the 1st. place. After running 4 tests were easily repeated with in Handbook 44's requirements, that said, it did not fall within regulated repeatability when we did a test at a lower flow rate(50%). So, I was hoping someone would have the experience with a mass meter and knew of a solution. BTW, the system is PD pumps, and know control valve. Manual start of the pumps(IE: preset in PLC.). Minimum flow rate of the meter is 20. GPM, Max = 1,000. Let me know if you need more info. Thanks

 

[BigInch]

I'd say you have to do actual weight from a sample. That will tell you if the density reading is correct. As I say, you can calculate it, but so what. There is no guarantee that the calculations will be correct. You have an actual gas out there that could differ from the constants and variables used in a 98.8% pure propane calculation. Commonly the constants used in a calculation would be based on 100% pure propane anyway. Any other must be measured and fudge factors applied accordingly.

Losing inventory? That's a good question. As compared to what? Are they reading input to the plant with volume? With mass, or volume to corrected mass, or mass to corrected volume and are they calibrating properly. Are they comparing to weight measurements and mass from an actual sample? Is the inlet reading really any better than your outlet reading?

Not to mention propane systems leak, so it's not uncommon to lose inventory from somewhere, either from inlet-outlet measurements, or from two different meters, or from process losses to valve and pump packings, etc. The question is how much are you losing ... relatively speaking.. to what goes into the plant and how you determine the accuracy of each meter in the mass balance equation for the plant. I doubt that checking your meter by calculating the density is a good place to begin. Get samples and measure them first. Compare enough samples with actually tested densities with the density meter's reading and you'll know how accurate and repeatable your meter is straight away. Balancing against another meter is a waste of time. As I always say, a man with two watches never knows what time it is. It is better to have confidence in a (one) watch.

Plant losses are another story. First get some confidence in the inlet and outlet meter readings first.

Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[BigInch]

This US Gov site, the National Institute of Standards
is probably better than just any old equation of state method, and definitely is a lot handier.

http://webbook.nist.gov/chemistry/fluid/

gives for pure propane,

35.000 F
75.000 PSIA (NOTE THE ABSOLUTE PRESSURE)
32.864 LBM/FT3
0.030429 FT3/LBM

Density of pure propane would be about 0.5267 when compared to water at 60F

If the other was 0.2% by weight Butane(?) that would make SG around 0.5268

http://webbook.nist.gov/cgi/fluid.c...it=Btu/lbm&WUnit=ft/s&VisUnit=cP&STUnit=lb/ft

So, your mass density meter reading looks high, while your experienced value sounds low. We have added another "watch" to the soup.



Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[danthemeterman]

Simply put, the Liquid Propane(you would visualize leaks) comes in on rail, the rail cars are allowed to be static for a period of time so as to allow them to settle(sloshing), they are stick read(bleed out)and compared to the figures sent by the loading facility(refinery), they temperature compensate the gross amount in the rail car (unfortunately we cannot get management to believe there is value(inventory) in the vapor space of the plant and the rail car. This we consider at best about 3% at any given time during the year. Again, we are estimating over the course of the year, daily readings could come much closer, and would/could be much greater in the HOT months. These rail cars are unloaded through compressors which draw vapor off the plant creating higher pressures in the car being unloaded, then the compressors are put into reverse, drawing the residual vapors off the rail car. There is no set standard as to how far they are drawn down, most companies develop their own scenario depending on the cost of kwh. rates. I doubt if this company does although I do remember asking the terminal operator. This plant is approx. 600,000 gallons storage, 2 pumps, 2 compressers and 1 load out meter. IT DOES NOT USE A MASTER METER, IF IT DID AND MY JOB WAS TO BASE MY FINDINGS ON THE MASTER METER, THEIR WOULD BE ONE LESS SERVICE COMPANY AT THEIR DISPOSAL. There are no leaks, well if you know anything about gas plants I don't need to explain. Feel free to ask any more questions. We have no choice at this time than to trust the readout of the mass meter to tell us the density, my problem is I am trying to determine if the propane is at the density it says, is there any way to determine the specific gravity? I assume the specific gravity is approx. .505-.510 -+. based on the temperature and PSIG.

 

[BigInch]

Specific gravity, SG, of propane is referenced to the density of water with both at the same reference temperature at some reference temperature. Let's say the reference temperature is 60[°]F, so we know water density is 62.37 lb/ft3. By making 60[°]F the reference temperature we have also just set the value of water's SG_60[°]F = 1.000

Now if we have some specific gravity of propane, say 0.505, that propane's density is 0.505 * 62.37 lbs/ft3 or 31.50 lbs/ft3.

If you know the value of the propane's density, then divide that by 62.37 to get its specific gravity relative to water at 60F.

Density: You first weigh when empty, then fill a sampling cylinder of a known volume and weigh it again when completely full of liquid. Subtract the weight of the cylinder to get the weight of propane alone and divide that weight by the known volume of the cylinder at whatever pressure and temperature it has at that time. It must then be converted to the standard reference pressure and temperature.

I agree it is hard to tell where you're leaks are when the stuff turns to gas and disappears as soon as it gets away from you.



Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[danthemeterman]

BI, I do appreciate all the statistics which I have memorized from years ago, 75% of my business is in calibration of LPG meters, so you can understand that compensation, SG, pressure correction are all very very familiar territories for me. A hydrometer/densitometer are not in my list of equipment inventory, although it is getting to the point especially now I will need to consider it. That said, I was hoping to determine if the mass meter density indicator was even in the realm of possibility. With just the instruments at hand, I cannot determine the actual density. I cannot use the sample you give, I am already assuming the propane to be .505-.510 do to the pressure/temperature instruments I have available. My goal was to see if the density the mass meter is telling me the liquid is within the real of possibility. When programming the PLC there are broad fields of choices for density so as to allow the mass meter to narrow down the field to determine the density, that is wher I believe the meter may be lacking or the PLC ? Thanks again, DTM.

 

[zdas04]

REFPROP.EXE has a simple tool to generate a saturation table. I've attached 20F - 80F at 0.5F increments. The specific gravities are a bit higher than you predicted.

David

 

[danthemeterman]

Z,
Unfortunately my predictions are based on the product shipped by the refiner, IE: BOL. Thanks I'll look at the link you sent. Thanks to all of you.