Level Transmitters vs. Leve Gauge Readings 1

[Bal.]

Hi All,

Hope you're all doing well.

I'm a chemical engineer and am new in this forum.

Currently, I am calculating the corrosion inhibitor availability using the difference in the tank levels that are measured using level transmitters (magnetic level indicators). I tried to compare the level transmitter values with the level gauge values and they showed a difference ranging between 28 L and 81 L. I'm not sure which one is more accurate. However, high corrosion inhibitor availability is required in this system and inaccuracies in the measurements are really undesired.

Have you encountered such differences between level transmitter and level gauge readings? which one is more reliable? what are the possible reasons? and how to rectify?


I appreciate your responses.

Many thanks,
Bal.

 

[danw2]

What do you mean by magnetic? Magnetostrictive ?

Or a float whose internal magnet flips the colored flags (with a crude chain of resistors to create a stepped continuous output)?

What technology is the ’transmitter’ you refer to?

Level is inherently ’distance’ from an arbitrary datum bottom point. Where/how is the volume (L) calculated from?

Roughly, what are dimensions and maximum volume capacity of the vessel?

What are the characteristics of corrosion inhibitor? Self cleaning like alcohol? Viscous? Made from hydrocarbon stock?

 

[roydm]

It could well be due to the different principle of measurement.
The level gauge is probably using a float which will be partially immersed whereas the transmitter could be bouncing a signal off the surface.
If you could let us know the make and model no we could make a more accurate guess.

 

[Bal.]

Thank you. Following are the inputs that you've requested:

* There are two CI tanks; the volume of tank A is: 1800 Liter and the volume of tank B is: 8250 Liters.

* Level transmitters are Magnetostrictive. Make: Magnetrol, Model:Jupiter 200 (242-AH00-100-1M).

* For the level gauge, Made is Penberthy and Model is Multiview.

* The corrosion inhibitor is water-based, composed of Quaternary Ammonium compounds with a viscosity of 3.70 cPs.

Awaiting your responses.


Thanks,

Bal.

 

[danw2]

Magnetostrictive is on the higher end of level measurement accuracy instruments. typically ±1.5-2mm, enough so that some sites take the level measurements as HART primary variables or Modbus values, in order to retain the accuracy of the measurement and to avoid the 4-20mA conversions on each end.

The Penberthy one is a magnetic float flag flipper. The float is designed with custom weights for a specified fluid's specific gravity.

Typically, a chain of resistors is magnetically coupled for a continuous 4-20mA output. I do not know the accuracy of the 4-20mA off the top of my head, but I'm betting it's on the lower end of level measurement accuracy AND resolution.

I'm guessing that your errors of 28L in 1800L = 1.5%, or 84L in 8250L = 1.0% are likely due to combinations of

- lack of resolution in the Penberthy. I'm sure you can research how many points of resolution the 4-20mA has over the vertical range, but it's not likely to be as resolved a signal as the magnetostrictive. I have seen 'jumps' (due to lack of resolution) in the 4-20mA outputs of float flippers when the level is continuously changed in one direction and one watches the transmitter output.

- the match of float's bouyancy (for your liquid) and the point at which it flips the flag or couples to the resistor chain. Although that will be constant over its range, it will be an offset with respect to the datum.

- a different datum (reference point) for zero between the two technologies which affect the volume calculations

- 4-20mA conversions

- lack of precision in the implemention of the volume calculations

- error in volume calculation when the vessel is a horizontal cylinder, particulary with domed/dished ends.

- diminished level change in a low, wide tank where a volume change does not produce a large level change. The accuracy spec affects small level changes more than large level changes. The higher the ratio of vertical elevation to horizontal width dimensions produces better level measurment because a given volume change creates a larger change in the vertical level, which is what you're measuring.

I would tend to trust the magnetostrictive for level reporting repeatability. You'll only know its volume accuracy if you test it with a known volume.