dp dell level measurement

[sparky970]

I am trying to measure the liquid level in a vessel that is operated under a vacuum where NaOH is being evaporated from 32% to 50%. The low side leg is filled with Halocarbon, while the high side is tapped near the bottom. We are constantly battling accurate readings. Could this installation be improved to get better results, or would a bridle and radar probe be the best solution?

 

[JLSeagull]

I am a radar fan - both open and guided wave. I also consider magnetostrictive level measurement. Check the dielectric constant of the fluid for the radar instruments.

 

[danw2]

I curious. What sort of inaccuracies are you seeing? 0.5%? 5%? 15%?

What level are you working at? 1m? 5m? 20m?

Is the low leg (Halocarbon) a filled capillary or filled impulse tubing?

Is a smart transmitter (low drift) or an older one with zero/span pots?

Dan

 

[itsmoked]

You aren't using "head" are you?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[dcasto]

Hummmmm the density of the fluid is changing, so the precieved level is changing. Switch to a UT or capacitiance Level indicator.

 

[sparky970]

I am using a 3051 Rosemount,3m range, with filled impulse tubing. I think the specific gravity is about 1.4. Hopefully, I provided enough information for constuctive feedback.

 

[lacajun]

If your density is changing, no dp cell will work accurately and I am a huge Rosemount fan. I am a fan of radar, too.

 

[quark]

Density compensation is possible with 3051 types.

 

[sparky970]

Sorry, the density in this vessel is staying constant.

 

[danw2]

It's a vacuum process, and you've got a halocarbon fill in the low leg, so you must have a diaphragm seal on the vessel. The high side direct connected to the process.

Correct?

How subject to temperature changes is the low leg? Outdoors? sunlight/clouds? Indoors, radiant heat from a nearby process? Can you coorelate the accuracy changes to temperature changes the low leg sees?

 

[sparky970]

The filled low leg does not have a diaphragm seal. The low leg is exposed to ambient temperature. It does see radiant heat from the process.

 

[danw2]

What keeps the halocarbon level in the low leg at a constant level? How do you know?

Dan

 

[sparky970]

Would the dp cell setup work without any Halocarbon in the low side leg?

 

[quark]

I am still thinking about how the halocarbon would help in the vacuum process. I think you should get better results without HC (and also no problem to diaphragm due to high temperatures as this is vacuum).

 

[davsy]

Sparky970

A few thoughts and questions which may help understand the problem, excuse me if some of them sound simple or obvious.

What temperature and vacuum are you operating at?

How is the liquid level in the reference leg maintained. Is it possible that as it does not have a diaphragm seal and is exposed to ambient temperature you may have condensed vapors causing level or density changes in the reference leg.

What is the density of the Halocarbon that you are using? Is it drained and replaced on a regular basis or is it left without checking. Have you tried running it with a dry reference leg with a small drain pot at the bottom that is emptied on a regular basis.

We had a similar problem on a vacuum vessel which was greatly improved by changing to a dry reference leg, also tracing and insulating to stop condensation causing problems.

 

[sparky970]

The temperature runs between about 120 deg C at -14" Hg. The leg filled with Halocarbon is connected directly to the vapor portion of the vessel. It has a S.G. of about 4. The Halocarbon is not drained and replaced on a regular basis, and we have not tried it with a dry leg.

 

[dcasto]

how do you get a constant density in a system that has a changing density, do you have absolutely perfect homogenious mixing, could there be discontinous miing?

 

[davsy]

how do you get a constant density in a system that has a changing density, do you have absolutely perfect homogenious mixing, could there be discontinous miing?

additional to that what about variations in the concentration of the feed stream. Do you know that it is always exactly 32%, any variation here will affect the dP reading.

 

[HEC]

Evaporatting from 32% to 50% NaOH the SG will change from around 1.3 to 1.5, this will fundamentally affect the simple hydrostatic differential pressure level measurement. There are several options,

1. As suggested use a diferent type of measurement system, radar/ultrasonic/capacitance that is not affected by the density of the fluid being measured.

2. Use a three point diferential pressure measurement system. The points are, bottom (high), mid, mounted a known distance above the bottom, and vapour space (low). The mid point must remain submerged.
Fluid density is derived from rho=P/g.h where P is the diferential pressure between the bottom and mid sensors and h is height of the mid sensor above the bottom. Tank level is then h=P/rho.g
Cheers

Mark Hutton