Maybe you can use a sort of a level measuring devices that is (was) used in silo for powders: a wall mounted aggitator with a horizontal shaft where the motor current is measured. It increases when aggitator is surrounded by rag.
m777182
I'm looking forward to seeing some good answers here. I was involved in this sort of work a long time ago, but I have another design coming up soon.
Back in the 80's we looked at static pressure, ultrasonics, conductivity, etc - but we never found anything truly reliable. We even built calibrated floats that were supposed to float between the two layers, but the densities of the layers were too close and varied too much.
I really hope the instrument guys have made some progress on this one - they have done some amazing things in other fields. The biggest changes in process engineering during the time I have been involved have been in the area of control.
We have used Guided Wave Radar interface level transmitters, however there suitability depends a lot on what fluids you are looking at and their dielectric constants.
Some very valuable advice above. I would make one comment that dont rule out sight glasses we have full length sgs on coalessor/seperators - we do see a rag layer and purge off as required.
Now it all depends on MAWP limits, risks, materials .. if this is acceptable - always use a guard on your sight glasses scafolders have a difficult job and these are easy to hit.
I have worked on a plant that only use video cameras and sight glasses to measure interface but that was an old plant.
Depending on vessel size/ fluid density someone like tru-tech offer scanning services which can be effective. Not i've tried this on brick linings and it doesnt work for IF level.
I appreciate the suggestions I've received. Some had suggested not ruling out the use of a sight glass. While I agree that sight glasses can be specified and installed so as to minimize safety and loss prevention concerns, I don't have this option in this case. My client forbids the use of any sight glasses, unless they are required by law, such as on boilers.
One suggestion that I hve received elsewhere is to simply pump out the the rag layer periodically (once per day?) and recycle it through a filter (cartridge?) back into the feed to the decanter, thereby removing the rag layer. This is likely the way we will go as the the feed is fairly clean, and the rag layer should build quite slowly.
You may want to look into Neutron Backscatter technologies. They are based on the hydrogen content of the process fluids and can determine interfaces such as oil/water/vapour as well as emulsion layers. The one drawback is the depth of investigation is limited to about 4 inches from the outside of the vessel wall for accurate measurements.
