dear control colleagues,

i would like to understand, even though people nowadays changing their displacer to GWR for measuring the interface measurement, there are still reliability issue in term of GWR that is facing the maintenance people. any idea how to solve and sharing of lesson learned on this?

 

dear control colleagues,

GWR Technology is reliable and effective. We have replaced one of our very problematic displacer with this technology transmitter and it is reliable...

 

GWR should work ok provided you have a good clean interface and two fluids with a significant difference in dielectric constant.

The lower constant fluid needs to be on top e.g. Oil on Water. We have an instance where the opposite is the case, the organic is heavier than the highly conductive aqueous layer, radar wouldn't work there unless it was mounted from the bottom.

If you have an emulsion layer I would be very cautious. Displacers on the other hand don't get effected by emulsion layer and don't care about the electrical properties.

Roy

 

My friend, these Radars are the best when it comes to I/F measurements but they need to be set up correctly. Below is a writeup I have done for someone else that needed to know how to find the right di-electric constant values. Hope it helps, but please contact me if you need more help.

mass44 [at] hotmail.co.uk

You said interface level so I must assume you are measuring oil and water. My experience is on the Khrone BM100A and C, but your radar might be similar since they both use TDR technology.

The di-electric constant of crude is about 2 to 4 and water is 80. The variance in di-electric constant WILL have a small effect, so in order to find the right DC value one quick way is to use your sight glass to set the di-electric constants so that the radar reads the same as the sight glass on the water and oil. a Good average is normally 2,5 and 80.

Unfortunately this "jippo" trick will not work unless you have setup your Z/S parameters correctly.

If you are still having problems you need to do a complete setup from scratch.

In order to do this you need to get hold of the design engineer's internal vessel drawings, and look what and where the calibrated span should be on your vessel. What you are interested in is the exact mm measurements from vessel bottom to Z/S positions on the vessel. From there it is just a matter of taking exact measurements in the field of your vessel AND your radar installation, and make yourself a neat, accurate (cut away)drawing to indicate the design spec positions and the actual internal position of your probe in relation to the vessel and the probe. Calculate the Z/S from this drawing,and put these values in the output 1 and output 2 parameters.

Zero value for the radar is measured from the probe bottom up and 100% is also from the probe bottom up and not from zero position up.

So if you can see where the design engineer have said zero should be and you can see in exactly what position your probe is in relation to those points it is a matter of calculating how high you need to measure up from probe bottom to get to those Z/S points marked by the design engineer on the vessel. It takes a bit of trigonometry to do but is easy enough.

Typically these parameters should look something similar to this, keeping in mind that these radars are just distance measuring devices and that total probe distance minus measured distance equals level.

Exp:
Output 1 (Top product level) 4mA = 150mm, 20mA = 2500mm,
Output 2 (Bottom product) 4mA = 150mm, 20mA = 2500mm.

Look strange I know but we have found it is better to set them both the same instead of trying to set each one to an individual span. Both spans are setup in reference to actual vessel level %. This is also as per vessel design spec since interface is also a actual level measurement in relation to the whole vessel and not to just half the vessel, to for instance to where your weir plate is.

In this example the 150mm might be the mm you needed to measure from the probe bottom up to get to the Zero position of the actual vessel as indicated by the design drawing, and the same with the 2500mm. Obviously just examples. This is where your exact measurements in the field comes in.

NB!!

Also make sure you have the right probe length in the "tank Height" parameter and not the real vessel height. This probe length is normally stamped on the little spec plate on the head by the supplier.

Also take note that this is a method I have developed myself and is not as per your radar's manual. I had to do some very strange and out of the ordinary setups before, like on constantly moving FPSO's, and this method worked for all them there as well.

Good luck!

 

Dear All,

GWR have many advantages compared to the mechanical types. However, there has to be a clean interface (be careful with the emulsion layer). As a rule, if the emulsion layer is greater than 50mm then do not use GWR since they will not work. In the case of greater emulsion layers you can use capacitive probes that they have not problems with emulsion layer. Both of them are widely used and don´t have issues with maintenance. With GWR you can measure the total level and the interface level with just one instrument (you have to ask the vendor to include the special feature of total level and interface level if you want to measure both at the same time). However, with capacitive probes you just can measure the interface level.

 

Hi Frank

Please look at this write up I have done for someone else and give some advice if you know how to setup these capacitance probes other that what I have done below. Anyone is also welcome to makes any comments or opinions. If you think I am wrong I would like to know why you disagree please.

Indecently the "Khrone BM100" GWR use a permanent 100mm as the emulsion layer, so even with your tank completely empty the radar will still show 100mm level. These radars even worked fine in the FPSO environment with the fast and high pressure product in comings and the constant turbulence due to fast flows and FPSO movements. Personally I would use them on just about any I/F application, but it does make sense like someone said earlier to use them only on products with distinct differences in DC as well as that the top product should be the lower DC.

Anyway look at my writeup below and send me some comments on it where you agree or disagree. Like I said if I am wrong please let me know.

Previous writeup where someone wanted to know how to measure two products in a vessel simultaneously. In other words how to do I/F level measurements and what instrument I would recommend.

In a case like this “how to measure a interface level?” is not a straight forward answer since I cannot know what the vessel looks like, know what type of tap-off points is available and so on so I will give you a generic rule of thumb explanation on interface levels and tell you what is the perfect solution to all interface level problems. I will also tell you what you must not do.

Since this is a very confusing subject to a lot of technicians and engineers, although very few will admit it, so it might be worth the effort to do a detail explanation on the subject. First of all regardless of what anyone tells you, you NEVER use a differential pressure transmitter for a oil/water interface level application. Neither a piped nor a capillary type DPT is suitable to measure interface level. Believe me I have tried as well, and have done the calculations front to back and back to front and saw that it is theoretically possible to do, but believe me in practice it just doesn’t work. The only application where it might have a small chance of success is if you work on a vessel with a very stable and constant top product overflow. In other words your top product is always at exactly the same level and you only measure the variance of the bottom product. I don’t thing a application like this exist but anyway this is the theory of it. The moment the top product level changes, even by a couple of mm, your whole calibration is invalid and therefore your level control becomes unstable. So if you have a application like this you can try it but other than, that stay away from the DPT to measure interface levels.

Further more this is whole installation is dependent on product density so you can image for yourself what will happen if the top or bottom product density changes so this is just not the way to do it, it’s to unstable. It is like standing on a knife edge. Yes you might be able to stand straight and upright for a couple of seconds, but not for long.

a Lot of people will tell you the capacitance probe it the answer, it’s not. The capacitance probe can only measure one product. There is a special interface capacitance probe available on the market but most people are not even aware that the probe they are working with is not a interface measuring probe and that it is just a single measuring probe. 100% of the bottom product is 0% of the top product and then 100% of the top product is 0% of the bottom product they would say is the way to do it. In theory that makes sense but in practice it is not so easy to do. Image you are on a live plant and you now want the production to fill up the vessel to 100% with only the top product and then ask them to drain that and fill the vessel up with the bottom product again. They will most probably start laughing at you, thinking you are joking. I have tried this the one time since the cap probes I was working on was installed in a separate stand pipe on the side of the vessel so I thought if I can fill this standpipe I will not interfere with operations but eventually this didn’t work either since production will not allow you to open a pressurized vessel that is online with single isolation only. So in theory something might sound easy to do but in practice it is not always so easy. The normal capacitance probe calculations are also just for measuring a single product in a vessel under perfect conditions so they are useless as well.

Don’t trust your sight classes either since with interface levels most sight glasses do not work accurately all the time. These sight glasses were designed to measure a single product and not the variance of two products in a vessel. Draw yourself a vessel and vary the levels of the two products and see for yourself how it will reflect on the sight glasses. You will find that at various points the sight glass cannot reflect the levels accurately due to the sight glass tap-off points. Yes I know about the overlapping type of sight glasses but even they do not reflect the product levels accurately all the time.

But even if they do, you still sit with a problem of blockage when working with crude and other high viscous products and big vessels where the level changes very slowly. Is this now a accurate level indication or is the sight glass blocked again?

So in theory, if you can trust your sight glass and your sight glass is installed in a stand pipe attached to your vessel and you can open this stand pipe up to fill and drain it as you please, you can set these probes up like that, but this is a lot of if’s.

The best result I eventually got with these Cap probes was to setup my TDR radars (radars and cap probes installed on same vessel for control and ESD) perfectly according to the design specs and then calibrate the capacitance probes to the same zero and span positions as the radars. This eventually resolved the whole problem. If I had my way I would have thrown them out and installed a TDR in their place as well.