Level sensor for low density, under pressure

[Gauss2k]

Hello,

I'm looking for a sensor that could measure the level of a tank containing polypropylene beads (solid). Those beads have a density of approximatly 1 pound per cubic foot and the tank is under pressure between 60 and 100 psi.

The tank itself is made of metal (probably steel) and we suspect the inside was coated to avoid friction and rust. It's a cylinder: about 5' diam, 13-14' high

We presently use a low level probe that is like a fork. When there is no more beads between the metal bars, it activates. But because of the static electricity that builds up, it's not always working well (even though we try to eliminate as must static as we can). And operators would like to have a real time measure instead of just low level.

I saw ultrasonic sensors that could be interesting, but will they work under that pressure and with that low density?

Any other possibility?

Thank you!

 

[xnuke]

A lot of times, vendors will be more than happy to have a "trial attempt" at measuring something. They'll loan you a transducer for a trial period. If it works, you buy it. If it doesn't meet your needs, they take the demo model back. I've done this in the past with more than a few vendors and a variety of different instrument brands. I covered the installation/removal labor, so I wasn't out much if the experiment wasn't successful. If it was, I paid for the transducer. I wouldn't try this if I was measuring something in an environment that would contaminate the sensor. I'm sure you understand that vendors would not want a contaminated device back, but it doesn't sound like that is the case with your application.

Contact a few vendors and see what they say.

xnuke

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[cbarn24050]

Hi, a load cell is the simple way.

 

[bmsg]

You may look into guided radar (TDR) type level instruments. These are non-contact type instruments and should work fine in your application. Some vendors who supply this kind of instrument are Krohne & Magentrol.

You may also look into electro-mechanical type level instruments.

Good luck.

 

[Gauss2k]

Someone here told me they tryed the radar wave approach some years ago (before I was hired) and it did not worked well due to the low density and the wave reflections on the silo walls.

The load cell could be a solution (you mean measuring the weight of the silo to know it's level, right?). But there are several pipes connected to the silo and I'm not sure how this could affect the readings. I'll try to get more details about that.

I searched the Internet and contacted several vendors and one technology seems interesting. It's the laser level measurement. The vendors should come back with more details about that.

Thank you for for your answers!

 

[ilan7]

I would vote for the Guided Radar.
The Radar Instruments to day are totally different from those 2 or 3 years ago. Guided Radar has an antena inside the tank and works better then the old radars.
Contact Magnetrol, Rosemount (Emerson), K-Tek,

ilan

 

[cbarn24050]

Hi Guass, yes you use the load cell to measure weight. You can weigh the silo or you could put a false bottom inside it whith the cell underneath. All the external pipework doesnt make much difference normaly, and can be calibrated out quite easily.

 

[Gauss2k]

I've checked more thoroughly the guided radar systems on the market. I was also able to have more details about the probe that was tested and failed.

It was in fact only last year. It was guided radar.
The problem they faced is that the dielectric constant of the material is about 1.02 (almost the same as air).

The technical datas I saw on the internet said that the guided radar technology required a dielectric constant at least around 1.5 - 1.6
So I think we can forget that option

Measuring the weight of the silo could maybe be done, but the silo is going up to the ceiling, there is almost no room if we want to add someting under it. And i'm not sure if it would me precise enough anyway. The material in a full tank weight about 120-300 pounds (I learned that the density can be as low as 0.6 pound per cubic foot and as high as 2 ppcf!). I don't know how the tank itself weight, but it is probably considerably more. The measure would need an maximum error of a few pounds.

The vendors that I contacted seem to have difficulty solving that problem too. A couple of them asked for samples to make more tests.

 

[ScottyUK]

Another thread - thread248-98219 - discussed use of a laser-based system for distance measurement and posted a couple of links. I guess the polypropylene beads would reflect a reasonable amount of light for the sensor to detect, but I'm not sure what the effect of the granular surface would be. Anyone got any experience?



------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[IRstuff]

The spot diameter for the SICK DMS is less than 1 milliradian, so it will be affected by the bead distribution, etc.

Which raises an interesting question. How level is the sample? Is there some process by which the beads are actually leveled?

One possibility is to use a laser with some sort of scanning system so that more level samples are collected and averaged or otherwise processed.

Ultrasound should work, but you'll need to have a reference distance to calibrate against in the chamber

TTFN

 

[Gauss2k]

The material is exiting from the center bottom of the tank. So the beads are undoubtedly forming a cones as it empties (even though I haven't actually seen it since it's under pressure and we can't open it) Of course, this makes the readings more difficult, but I guess a precision of +/- 0.5 to 1 foot wouldn't be too bad.

The ultrasonic sensors I've seen up to now would be great if it wasn't their low resistance to pressure. This technology doesn't seem to be adapted to that kind of environment.

Someone suggested me an admittance sensor.

http://www.drexelbrook.com/products/Sku.cfm?sku_Id=12038&GO.x=16&GO.y=8

But since it's reading is based on the dielectric constant of the material, I wonder if it would be precise enough to distinguish the beads from the air.