level sensor 2

[spatel1765]

Hi friends,
I am looking for some level sensors for pharmaceutical reactors in which conditions will be as below.
from full vacuum to 100 psig pressure.
form -20 deg.C to 130 deg.C temperature.
liquid die electric constant may be from 1.4 to more than 70.
there will be agitation in the reactor with frequency drive.
vacuum distillation will be carried out in the reactor which will create vapor generation.
Liquid may be conductive or nonconductive.
There will be precipatation during process.
There will be foam generation sometimes.
Appreciated your advice.

Sanjay

 

[quark]

A load cell.

 

[FiltrationEngineer]

Look at Endress and Hauser (

http://www.us.endress.com/)

They have a pretty good online selector for the appropriate type of level sensor all kinds of applications.

 

[danw2]

Foam or vacuum eliminates ultrasonics
Varying dielectrics eliminates capacitance
Foam/precipitation might eliminate radar.
- Radar requires a relatively large antenna, which is fine on bulk storage, but in a reactor ? ? ?
- Precipitation on the antenna ? ? ? ? ?
Precipitation might hang the float on a mageto-restrictive sensor?
- is mounting a vertical sensor suitable in an agitated reactor?
Differential pressure might work
- Readings will oscillate during agitation.
- if the connection is through capillary impulse lines with chemical/sanitary seals, large area seals have distinct limitations at higher temperatures/vacuum. I once had the sanitary diaphragm seal bulge out (destroyed the seal) when the capillary fill fluid boiled and created pressure, rather than transmitting pressure. The transmitter sales guy gave me a chart for temp/vacuum working ranges, but I can't locate it.
- you don't mention specific gravity changes, but S.G. directly affects head pressure readings.
- you need intrusive access at the bottom of the tank, which might or might not be available.

I like Quark's advice, load cell, because it misses all the baddies, if you can rig flex fittings for piping.

I'm really impressed with the little known Global Weighing (now Sartorius) on agitated vessels, because of the unique mounting that zeroes-out side forces. Their matched load cells have eliminated periodic tweaking for drift correction. There is no drift. Period. Calibration takes about 1 minute - add a Cal weight and check the output. It's always dead nuts on. Well worth the premium for German engineering.

Dan

 

[25362]

Question: are load cells compensated for varying pressures as spatel1865 is pointing out ?

 

[danw2]

Are load cells compensated for varying pressures?

Answer: No. The load cells weigh the vessel, its contents and any anomaly that appears to be "weight" as a force contributed by piping or connections to the vessel that do not allow the vessel to free float on the load cells. The pressure in a contained reactor would be a component of content weight.

Vessel weight is typically eliminated by electronic subtraction.

If the pressure is applied from outside the reactor, then the weight of its gas contributes an additional component to the total weight.

If the pressure results from a reaction in the vessel then total mass would not have changed and the content's weight would remain the same.

If compensation for the weight of the gas is required, it could be calculated, knowing the volume of the vessel at a given level and the pressure of the gas; and then subtracted. The weight of 6 atmospheres of gas might be insignificant in a production sized reactor operating near capacity.

 

[bud3]

Forgot one technology (because I've never used it) that might work - Guided microwave or "radar on a rope" that shoots an RF signal down a probe, where a reflection bounces back off the medium.

Never having used it, I'm not sure what would limit its application, whether foam or precipitation might give false readings.

Correction to previous post above - the technology is magneto-strictive; not restrictive.

 

[spatel1765]

Friends,

Thanks for your response,
The density of the liquid is also going to change form time to time. so load cell will be a problem. I am planning to try Radar or Guided radar. What's your opinion.

 

[Ashereng]

spatel1765

Precipitates may affect the guided radar.
The foam may affect the radar.

If you know what the new density is, the load cell is the way to go. You can change the density to recalculate the level.

The only other solution not mentioned that I can think of is nuclear radiation level measurement.

 

[Ashereng]

quark, I also liked your suggestion.