Electromagnetic Flow Meters

How many cubic meters flow through the pipe in one day?

What is the accuracy range of the meters?

Take the average of the two meters and don't worry.

1. Check the specs and piping arraignment for the proper upstream and downstream parameters.

2. Check for proper grounding on both meters.

Turn one of them off.

What did the flowmeter manufacturer say on the phone (apart from thanks for buying two instead of just one) about this?

That's why one of them needs to be turned off!

It is truly unlikely that quality meters were incorrectly calibrated.

What is more probable is that at least one of the meters has an instalation anomally.

It might be too close to a bend, or a valve or a reducer or some other flow obstruction that is causing turbulence and so disturbing the reading. It could even be the rubber seal is projecting inside the pipe.

Depending on configuration, there might be air inside the pipe etc.

We have some lines with two flowmeters. One at each end, to monitor that what we pump into the pipe arrives at the destination.

One other option (unlikely) is that in your software, the scaling parameter for mV to ML is incorrect. We spent some time chasing an 8ML "change" in one of our plants. It was a SCADA scale error, not the MAG. If that is the case, then the error you see will be an exact ratio to volume measured.

We have some lines with two flowmeters. One at each end, to monitor that what we pump into the pipe arrives at the destination.

Great idea!

Yes,

Some of those lines are over 35km long across various landscapes and cover and since we are handling water, there is very little environmental "signal" of leaks, especially in wet times.

We also have some sewer lines (around 1km long) that go through environmentally sensitive locations where a "Spill" requires significant rectification and a lot of bother!! The use of flowmeters here provides som assurance that we are containing the material.

If you are using Krone mag flow meters, check that the GK numbers are the same on the sensor and on the controller.

"... there were installed in series because of the location and need for that setup...."

.

Following the word 'because' with unrelated info or circular reasoning doesn't really qualify as an explanation and isn't actually helpful. It does sound like the salesman was having a good day though. "why do you need two? we'll because it is going over there, of course....as well as the need to set it up that way"

.

.

"...two different readings (ranging in the range of approximately 50cubic meters per day) and there is no leakage in between...."

.

That may not be a problem at all. If the total daily is 10,000 m³ or greater, the meters could be accurate within ±0.25% and still show a 50 m³ discrepancy. What accuracy are the meters supposed to provide? What is the typical total daily flow?

.

"...I have already accounted the amount of water in the pipe between the two valves(6 cubic meters)..."

.

Okay, now you have my attention. How exactly are you accounting for the water in between? Can you describe the process by which the pipe volume between changes the flow past the two points? There isn't a point when the pipe it less than full, is there?

All liquid flow meters should ideally be installed in vertical pipe with an upwards flowstream. This arrangement helps to guarentee a solid column of water so the meter is not seeing partially filled pipes and registering it as full bore flow. If one or both are mounted in horizontal pipe you are likely to see differences in the output.

thanks jhhassociates, have realised that its one area which was overlooked in terms of installation. One of the meter is installed in a vertical pipe while the other is installed in a horizontal pipe. Apparently the one which is installed in a vertical pipe is giving us the smaller reading. Under this scenario is it that which is supposed to be expected.

What is the distance between the flow sensors (flowtubes)?

You must have uninterrupted flow (no disruptions) for a minimum of 5 x pipe ID upstream of the first sensor as well as in between the meters with 2 x pipe ID on the downstream side of the last sensor. (no bends, no fittings, no penetrations, no dents, no pipe flexibility, no connections,etc.)

Any/all cavitation and/or disruption in the flow will cause offset and/or error(s).

If the distances are not acheivable, my advice is to use an four-hole orifice plate meter with temperature compenasation. ER, EH, ABB, & Siemens all make this type of meter.

i need clarification on this, whats the effect of air in the system, how does this affect the overall flowrate considering that the medium being measured is water and how does turbulance affect this. Want to reiterate that the one connected in the vertical column is giving me the lesser reading.

To avoid turbulence affecting (lowering) the reading you need straight pipe with a minimum of 10 pipe diameters upstream of the meter (20 is better if you can achieve it) and 5 pipe diameters downstream of the meter. Air in the system will give you a false reading. The meter is reading the magnetic field strength and velocity of the liquid. As more air is entrained the field strength goes down but the velocity goes up. With a small amount of air the results are unpredictable, but with a large amount of air (10% +) the reading will always go down because field strength is the dominant factor.

Flow=area of pipe X velocity. Mag flows measures velocity and always assumes it is a full pipe, so if the pipe is not full , your reading will be wrong.Some magflows works with a level meter installed besides it.

Using double instruments to measure the same thing is asking for a headache. Unless they are measuring exactly the same point (e.g. two temperature sensors within an inch of each other) you're most certainly going to get a difference.

In our process, we have two flow meters that are essentially connected in series though there are vessels in between them. A difference usually signals a leak somewhere. This is the only reason I can think of to have this kind of setup.

Are your pipes completely filled? No bubbles? No debris? Is there enough straight run upstream of the flow tubes? If you're using remote heads, are the excitation cables of the same length and size? Do you have the same voltage for their supplies?

If your answers to the above are all "yes" maybe you can try connecting the flow tubes in series within a few feet of each other and comparing their readings. It's possible there is a slight difference in calibration. Or you can send them to a certified calibration facility.

regards,

Vulcan (",)