How to verify liquid level in tank 1

[mp08]

We currently have several large storage tanks with radar level transmitters we are having issues with. The problem is that the level (which we then convert to weight) is never correct. For example, when loading a trailer the level difference has the weight calculation at 41,000 lbs and when we weigh it, it actually weighs 45,000lbs. There have also been times when we've opened the tank and you can tell that the level measurement isn't correct.

I plan on having the transmitters reranged since I think its range is larger than the entire tank height (no one can tell me what the actual calibration range is). In order to check though, I want to do a test where for a certain amount of material in the tank, I can make sure the level transmitter/weight calculation is correct. My question is does anyone know how to do this for a large tank? Is there a device where I can find the liquid level from the outside and do some hand calculations? I remember for some tanks where we had radiation levels and thought we had issues, we used geiger counter (maybe?) to help us figure out where the level was .

Thanks
-Marisa

 

[itsmoked]

Sounds like the existing level sensor is failed, failing, or out of its application area.

Of course the most true method for a back-up check is a dipstick.

I see ultrasound used more than radar. Perhaps a change of sensor?

Pressure sensors work very well unless there's debris in the fluid. Then the venerable bubbler is pretty fool proof.

If all you ever care about ultimately is the "weight" why not just weigh the tank with a strain gauge setup?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[mp08]

I've had issues with radar transmitters at my other jobs, but this new company I work for likes radar transmitters. Maybe I'll try ultrasonic.

These tanks are very tall - the shortest one is around 36 feet tall (250,000lbs+) and have to kept hot (150F+) since they are polyester storage tanks and the product can get quite thick. I think the height is what kept them away from bubblers and pressure sensors. The height is also the reason why we can't use a dipstick to verify the level.

 

[ScottyUK]

Old-school, but what about a Whessoe servo gauge? Anything which survives crude oil service has to be fairly rugged.

What's the product in question?

 

[mp08]

All the storage tanks are polyesters, some of which have solids.

 

[IRstuff]

One problem, it would seem to me, is that a heated polyester tank is not going to have the same diameter as a cold polyester tank. Since you are presumably using the height and an assumed diameter to determine the amount of material delivered, this process might be rife with potential errors.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[hacksaw]

a level measurement with less than 6" error in 36 ft (1.5%), and in an operating plant is pretty good given all the variables that you've not mentioned.

unless you are measuring the density, multipoint temperature in the tank and those same variables measured again with the flow as delivered to the truck you are going to find it difficult to resolve...

 

[roydm]

Have you allowed for the tank expanding with temperature.
You should be able to determine if the echo is bouncing off the true level by looking at the signal profile. Could it be hitting a vapour layer.
Polyester, does that have a large dielectric constant that radar needs?

I doubt you will get better results from ultrasonic, the speed of sound is effected by change in temperature.

 

[danw2]

If the radar works, meaning it reliably provides level or distance readings and does not fault into 'loss of echo' mode, then your best bet is to do some research, find the critical data and then do some calculations to determine whether the radar is configured properly or not.

The statement "no one can tell me what the actual calibration range is" is, unfortunately, all too common. All too frequently no one seems to know where the strapping table or tank dimensions are; what the instrument configuration parameters were, or exactly what the mass conversion factor was.

Changing measurement technologies is fruitless unless you have your hands on all the critical information needed to get the measurement technology to perform properly. And when you have access to that information it makes far more sense to me to review the data to see whether the radar and the receiver (DCS/PLC) are configured properly to produce a precise measurement.

You can always use a weighted rope to get physical level measurement to check against the radar measurement.

 

[phteven]

like hacksaws response.
But depending on type of radar and year they can be very preferable. If you get the software (companies have many: AMS/Fieldcare for example) a curve can be seen and deciphered. Many times radars can be found to be calibrated differently then expected. CRUDE Example: radar reads distance of 9feet from 36feet expect 25% but the calibrated range is 1ft to 31ft so reading 13%.(but expecting you know this) To test you can get laser levels (ABB for example makes one). But the weigh difference is probably due to assumed calculations. You state that the material needs to be hot...does the "DCS/PLC' take a temp reading or imply densities? Someone else mentions strain gauges and load cells are great for reading weight. Taking a level then calculating weight will always give an exceptional error if density is assumed and not taken. NOTE: even some of the "higher end" strain gauges will have 3% error of max rating.

 

[LydiaMiller]

Do you have contacting or non-contacting radars? I work for Emerson Process Management. We have both contacting and non-contacting radar devices.

I am also curious if you have the same devices on both your liquids and solids? Or do you have no dry solids, only liquids and some liquids with solids (or slurries)?

It sounds like you are having problems with the actual measurements since when you look in the tank the level is not where you are reading it. On our devices you can look at an echo curve, which shows where the microwave signals are being reflected. This graph helps quite a bit in determining what is happening inside the vessel and where you might be having a problem with getting a good reading.

Radar devices are fairly simple to set up, but you really need to follow best practices for installation and configure the devices correctly. Additionally, if you were getting repeatable level readings that just don't correspond correctly to the weight, it may be that your calculations for the weight are not taking something into consideration. For example, if you use a straight volume calculation for a cylinder and use the density to calculate the weight you may have errors from the tank actually expanding when it gets filled, changing shape due to temperature changes or not actually being a perfectly shaped cylinder. Additionally, the density changes with temperature (and pressure) and if you are not compensating for those changes you would also have some errors.