Problem with pressure transmitter readings on a vertical slurry pipeline 2

[Discreet544]

Hello Folks;

I am pumping sand-water mixtures in a vertical pipeline. I assume I ve considered all the design parameters. I use a Fuji differential pressure transmitter (0-6 kPa) on a closed loop comprised of vertical and horizontal test sections. Now the transmitter is mounted on the vertical section. I have a mixing tank and an agitator to well mix the mixture too. The readings for water are perfectly matching with theory. However, the readings for sand-water mixture are 3 times more than the range expected to be. This has been tested with glass beads as well and the results are off by 200%. Any guess what might be the source of error? or if this is an error at all?

 

[Compositepro]

How about a few more details, like a piping drawing. Otherwise we would just be unnecessarily speculating.

 

[Discreet544]

Absolutely! The pipeline is 25 m long and 50 mm in diameter. The vertical section is 2.4 m long and the transmitter impulse lines are connected to pipe over a distance of 40 cm (the length the pressure is measured over). The sand D50 is 500 micro meter and slurry concentration is 5% volumetric. Temp is constant 15C. There is 35D and 4D space before and after the test section (D=pipe diameter).

 

[KevinNZ]

The mixture requires more force (pressure) to move than just water. The density is higher and there (I guess) will be more friction to over come.

 

[Compositepro]

Is the DP transmitter for measuring flow, level, density, or what? Results are off by 200%. What results?

You may be having problems with settling of solids in your piping.

Slurry is denser than water. If the fluid in the pipe is denser than that in the impulse lines, there will be a zero shift error in the dP readings, which you will not observe at zero flow because at zero flow the sand will settle out in the pipe. Still just speculation.

This is beginning to sound like a college lab exercise.

 

[danw2]

What is the Specific Gravity of the water and sand mixture/slurry?

I'll repeat Compositepro's question - are you measuring flow (DP across a flow element) or just low pressure (gauge pressure reading no high side, low side open to atm)?

The pressure tap is in a vertical pipe run. What diameter is the impulse tube/pipe? How does the impulse tube/pipe connect to the pressure transmitter? horizontal run for 40cm? Sloping down or up for 40cm? 3-4 cm horizontal, elbow, then a vertical drop to the transmitter?

What are the numbers you're getting? Water only pressure? Expected slurry pressure? Actual slurry pressure? glass bead actual pressure?

 

[LittleInch]

"Any guess what might be the source of error? or if this is an error at all?"

At the moment none at all as you've given us virtually nothing in the way of data.

All we know is you have a 25m test loop of 2" pipe (unknown material)
There is a vertical bit 2.4m long and nin the vertical bit you have two tappings for a DP transmitter ranged at 0-6kPa (0 to 60 mbar) 400mm apart. We can only assume you are measuring frictional losses in the pipe as you don't say anything about an orifice or other restriction.

So far so good.

Now
" I assume I've considered all the design parameters" Great, care to let us in on the secret?

Flow rate perhaps?
Measured density of the slurry

"The readings for water are perfectly matching with theory. However, the readings for sand-water mixture are 3 times more than the range expected to be." - how about sharing this with us so we can see if / where you might have made a mistake.

This appears to be a lab type experiment - what does anyone else think (professor / supervisor etc)

What actually are you trying to determine here by these tests?

LI


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

LI - you need to polish up your crystal ball by the sounds of it.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

My Chrystal ball has melted in the heat here at the moment.

Clearly not the right material.

discreet - Apologies if I went a bit far, but you really need to read my strapline...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[BigInch]

Power failure? All crystal balls appear to be on the blink today.


Richard Feynman's Problem Solving Algorithm
1. Write down the problem.
2. Think very hard.
3. Write down the answer.

 

[Artisi]

Thinking back many many years, I recall that a slurry of like size material is /can be problematic, 5% volume isn't normally a problem but factoring in all material of 500 micron could well be the problem (might be plug flow).
The OP need to research slurry pumping for answers and not ask for help without researching what could be the problem.

Edited to add

http://www.pumpfundamentals.com/slurry/WeirSlurryPumpingHandbook.pdf

(the Bible of slurry pumping originally from Warman Australia)


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[SandCounter]

As previous comments mention, you have not presented much information.

Using 5% by volume and assuming standard 2.65 SG sand, the slurry density will increase dynamic pressure by 8.25%. You will see additional losses due to friction (this may account for the 50% variance between sand and smoother glass beads). I believe you have left out the static pressure along the vertical 40cm between taps.

Based on your 3x bust, I calculate your pure water pressure at the lower tap to be 130.6 kPa and your expected sand slurry reading to be 141.4 kPa (that's only a 10.8 kPa differential). I calculate your actual sand slurry reading to be 162.9 kPa (that's 32.3 kPa). This 3x difference between your expected and my calculated is in the 40 cm static head between the taps which I find should increase by 21.6 kPa.


I used to count sand. Now I don't count at all.

 

[1gibson]

Unless I am missing something, 2.0 specific gravity fluid from the same pump at the same flowrate in the same system will have similar head loss as 1.0 SG fluid. But that means about 2x the pressure loss due to the difference in SG.

SO convert the results to units of head, not pressure, and see if they makes sense then?