Guided Wave Radar Interface

[roydm]

Hi everyone, I have a couple of questions on GWR.
Has anyone used Guided Wave Radar to measure interface where the Upper level is acidic water and the lower level Hydrocarbon. My intuition tells me the radar will not be able to penetrate the highly conductive water layer, am I correct?
Where you have an emulsion at the interface would you get a good echo?
We normally use bubhle tubes for this application but a new customer is asking for radar.
Personaly I think displacer type transmitters would work better, any thoughts on that type?

Thanks & Regards
Roy

 

[itsmoked]

Is the water layer always less than some amount? Like a few inches? That might allow good results with GWR.

BTW why is the water floating ON TOP of the oil? Isn't that backwards?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[roydm]

Hi Keith,
No the water could be up to 1 meter deep (4 meter vessel).
This particular hydrocarbon has an SG over 1.1, so no it's below the acidic water. Occasional stickiness rules out anything in contact e.g. magnrtorestrictive float.
thanks
Roy

 

[itsmoked]

Okay ..

Also sad as I was waiting for Roy to jump in...

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Cat3347]

Most TDR/GWR require the bottom fluid to have at least 10 times higher dielectric constant than top fluid.

Stick w/ displacer or magnetostrictive.

 

[roydm]

Keith, Cat,
Thanks for your response. I am leaning towards a displacer, I have ruled out magneto-restrictive as that relies on a float sliding on a rod. Our process can be sticky at times.
We currently use bubble tubes but client has asked for GWR, I have my doubts but haven't radar enough to get a good feel for it's limitation.
Any more suggestions?
Regards
Roy

 

[Unotec]

Talk to the supplier. They'll have to warranty it anyway

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[roydm]

Unotec, Yes but a warranty isn't much good when you are in a startup situation and everyone is asking you why your level transmitter is holding everyone up.
Agreed, if it's a one off it's OK but many times on a large project it could be 6 months between purchase and startup.
Thanks
Roy

 

[Unotec]

Yes but, unless you're an expert on these units, your best bet is to talk to a couple suppliers, get their technical input and then do your own evaluation.
That's what I usually do with equipment I know what it does and how it works but I am not an expert in (mainly all instrumentation)

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[roydm]

Hi Guys,
Thanks for the response so far. I did some digging and found out the following.
Upper Fluid
Acidic Water Dielelctric constant 55 - 80, SG 0.997 Conductivity > 100 mS/cm3
Lower Fluid
Nitrobenzene Dielectric constant 20 - 25 SG 1.2 mS/cm. Conductivity 0.005 ms/cm3

The fluid is very nasty, Glass, Teflon & Tantalum are about the only things that stand up.
Regards
Roy

 

[itsmoked]

Yeah indeed but it smells really cool.

With those radically different conductivities I wonder if you couldn't use two conductive rods and a top float to accurately calculate both depths.

Be a science project though.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[roydm]

Keith,
Yes, I have been promoting the idea of using electrode-less conductivity for a couple of the interface point level applications.
Hope to try it out in our lab.
Roy

 

[danw2]

"Electrode-less conductivity for interface"

What's that look like?
Dan

 

[roydm]

Danw2,
I'm not sure if you were asking how you measure conductivity see attached

http://www.omega.com/GREEN/pdf/CONDUCTIVITY_REF.pdf

OR
how you would measure interface.
Obviously as the interface from a highly conductive water based solution is replaced by an organic low conductive solution the output will change accordingly. I assume that when the solution is half way or in an emulsion the probe output should also reflect that. I haven't had the chance to try it out.
Roy

 

[danw2]

Either exposed wetted electrode or torroidal conductivity is a 'point' measurement, as opposed to a continuous measurement.

I'm trying to figure out how one uses one or two conductivity measurements, presumably at some fixed elevation 'point', to see or determine a change in interface level.

I'm assuming that the conductivities remain constant, even if the interface level changes. Is that correct?

Or is the conductivity of one of the media changing which results in a change of interface level?

 

[roydm]

No, the probe would be located at the interface level. In this case if the level went up the conductive acidic water would be replaced by low conductivity organic.
I'm just not sure what happens when the interface is half way. I don't expect a liner relationship.
Likewise i'm not sure what the response to an emulsion would be.
Regards
Roy

 

[99loki99]

I've not heard good things about using GWR when there's an emulsion pad of any great thickness.

To get round the problem of the fluids being the wrong way up you could always stick the probe in from the bottom! Only problem would be draining the entire tank contents if you needed to change it...

I'm looking at retrofitting a capactive probe into a separator at the moment. E & H do one and so do Siemens. Anyone heard any reports - good or bad?

Fluids are sour gas, 35deg API oil and brackish produced water at about 60C. Glass lined vessel so would have to include a ground tube.

 

[JLSeagull]

In my refinery apprentice years both engineers and technicians held capacitance level instruments in low esteem. I am a fan of GWR and avoid capacitace level instruments. However, I would assure that the GWR application was solid before going that direction. And don't forget about the business that salemen are in.

 

[roydm]

Thank-youi all for your response'
I have concluded that GWR should work provided there is a clean interface (no emulsion layer) and the bottom layer dielectric constant is significantly higher than the upper level.
For interface where the upper fluid has the higher dielectric constant I will recommend a magnetorestrictive type.
As I mentioned earlier we normally do this with differential bubble tubes, that works well but the client wants to use something a bit more sexy.
Regards
Roy

 

[99loki99]

I have a couple of questions about the bubble tubes roydm was talking about...

What gas do you use? (N2? Process Hydrocarbon Gas? Air would be a bad idea!)

How do you compensate for a pressurised tank?

Thx - Loki.