Flow meter records water pumped during operation but when we are shut down over-night, the meter continues to register/record during the night. Any info would be appreciated.

Marsh-McBirney, Inc. is the meter manufacture

 

Either there's a delta P and flow path so the meter is indicating true flow or there's true zero flow and the meter is falsely indicating flow.

Since I've seen numerous examples of true flow when the pump is off with low flow sensitive clamp-on transit time ultrasonic flow meters, I tend to believe claims of true zero flow only for meter isolation with a couple new ball valves.

The causes for flow when the pump has been turned off have been traced to
- Fluid runs downhill (gravity drain)
- Valve leaks: check valve, shutoff valve
- "I didn't know that line was Tee'd off there"
- flange leaks
- thermal syphon
- pressure syphon

The causes for a flow meter to indicate flow when it is truly isolated (ball valve on either end) are

- flow meter is not zeroed; for instance, Coriolis meters need to be zeroed

- electrical noise from VFDs can cause a magmeter to show false flow (non-factory cable or incorrect shielding on remote mounted magmeter converters)

- If a magmeter electrode cable goes open (gets disconnected)

- an empty pipe can cause a false flow indication on some mags

- I've heard of vibration causing false flow indication in a vortex meter, but haven't personally seen it.

- electrical noise from lack of magmeter grounding ring or unconnected grounding ring

- cross talk between magmeter electrodes and coil drives when cabling for multiple remote meters installed in common wire tray

- wet connections in JB anywhere along the cabling path

- electrolytic or cathodic protection system contributing electrical signal error to magmeter

 

Dear Mate,

You must be sure, there is no flow through the flow meter. After being sure, If the flow meter is tube type (not capillary) may be the one of the connecting tubes is empty because of leakage. This condition makes a differential pressure across the flow meter and it sense an unreal flow. So you must fill the tubes and repair any leakage.

 

Dear Mate,

Please give me more information, If the air you mentioned is instrument air or turbine air intake. If you performed cold wash (offline washing) before start up.

 

Well, it would be interesting to know which Marsh-McBirney meter it is but hearing none, let's proceed. I am going to assume that you have one of their two models of insertion magnetic flow meter. If that's not the case, please say so in your next post.

It would also be nice to know what the orientation of the flow meter is, horizontal, vertical or other. This could have a bearing.

If your flow meter is installed in a vertical riser (flow going up) and there is a check valve above the meter, the meter may be registering reverse flow caused by a leaky seat on the check valve. Same if your flow meter is installed in a vertical riser (flow going down) with the check valve located below the meter.

Do please let us have some more information so we maybe can really help you.

Walt

Walt Boyes, FInstMC, Chartered Measurement and Control Technologist
Life Fellow, International Society of Automation
Editor in Chief, Control and ControlGlobal.com
[email protected]

 

This may be too simplistic but can you configure the meter to log only when the pump is running?

Regards

icenta Controls
www.icenta.co.uk

 

> This may be too simplistic but can you configure the meter to log only when the pump is running?

That's the common fix for this problem. And its a REALLY BAD idea. The meter is telling you something. Its either mis-configured and reading wrongly, probably all the time, or you really do have a flow there and something is wrong in the plant.

Ive seen this done many times. And EVERY time it was covering up a real problem in the plant. In one case the valve actuators were wrongly installed and would go past closed and open back up again. The true closed point was a 14% output on the valve. In another case somebody had incorrectly set the zero flow on several mag tubes and all read over 1000l/min errors.

Rob
www(.)lymac.co.nz

 

I have resolved a similar problem in the past on a E&H Magflow by increasing the empty pipe detection value on the flow meter. It happens sometimes that when the pump is stopped that the flow meter still keeps on reading the last bit of flow while the pipe is draining. You don't want that since that will screw up your total flow reading for the day and the operators in the control room will still see a flow indication after the pumps are stopped. They don't like that either. By increasing the empty pipe detection value you tell the flow meter to show zero flow if the flow goes below a certain percentage. I cannot say for sure that this is the same problem in your case but that is something you can look into. In my case I knew exactly why the flow meter was still giving a reading but like some of the other replies said, we should not use these features on a flow meter to hide a real problem somewhere else. You should give the pipe enough time to drain after a pump stop and what you should see is that the flow keeps on going lower and lower until it eventually comes to the empty pipe detection setting at which time the reading will go directly to zero. If the flow meter is not doing that and just keeps on giving a constant flow reading after the pump is stopped then you have a leaking valve somewhere or a zero shift on the meter.

 

Is this a flume type meter?

 

I'm with the posters above who are recommending that you look for a leaking check valve first. I've seen this happen numerous times.

 

> I'm with the posters above who are recommending that you look for a leaking
> check valve first. I've seen this happen numerous times.

if this were a weir or flume type meter that measures the depth of water in a given cross sectional flume then a build up of debris can give a false high level and therefore indication of flow.

nothing to do with check valves at all.

 

hi,

very clear explanation, but it seems you are speaking about "low flow cut-off" parameter instead "empty pipe detection" (anyway both are useful for meters measuring when the flow is going off)