Hello,

Within the last 3 months I had in our company a strange behavior of the gas line. There is a main input turbine flowmeter that belong to the supplier and another 5 flowmeter inside our factory that belong to us.

Until a few months ago the main flowmeter reading was very close to the sum of the other 5.

Within the last months the difference increase something like 40%, 55% and 90% in the last month. We did a lot of verifications and also many checks in which we let only one line open in our factory that showed the same difference between the main flowmeter and our flowmeter.

(one detail that you may thing about: there is no error caused by the pressure, I used it for flow calculation in our test - also there are flow compensation devices on every sensor)

In the next day after the main flowmeter sensor was changed by our supplier, the reading comes true: the main flowmeter equal the sum of our 5 flowmeters, ... and the same in the last 2 weeks.

The flowmeter was sent to laboratory for testing and the result was that it is good, the error is within the 1% limits. I was there and I don't have any suspicion. One detail that may count: a lot of dust in the flowmeter that was cleaned before testing.

The only thing I could thing now is that maybe there was something wrong with the mechanical transmission between the turbine and the total flow indicator.

I kindly ask you for any clue or any of your similar experience that I may use for further investigations.

Best regards,
Petre P.

 

i think -10v when the rotor in middle.

you will have approx - 8, 12 if you put the rotor on both end.

best practice is to check it before opening the machine for old setting.

 

> maybe there was something wrong with the mechanical transmission between the turbine and the total flow indicator.

The turbine meters that I'm familiar with are mechanical turbines, but electronic pick-ups and electronic flow indicators.

Do you have a meter that is strictly mechanical? Mechanical turbine and mechanical counter?

 

The first mistake made was to not test the flow meter "as is" and to have a thorough visual inspection, which lost important troubleshooting information (e.g. doesn't work with dust, works good without dust). Dust appears to be a potential culprit. Some turbine meters have open bearings and dust on the turbine blades can cause low or varying readings (e.g. increased drag, or varying drag/lift). Don't suppose the dust was analyzed to see were it was coming from? Since it was the supplier's meter that caused the problem and not yours, it might be useful to determine what the difference between your turbine meters and the supplier's meters is and whether they are continuing to see dust or you are seeing any dust in your system (potentially future problems). If this happens again, I would recommend that the meter be inspected, tested as is, as cleaned, and the dust be analyzed to see where it might be coming from. A recommendation of upstream conditioning and/or a change in the turbine manufacturer or model might be in order

William (Bill) L. Mostia, Jr. PE Fellow

SIS-TECH Solutions, LP

Any information is provided on Caveat Emptor basis.

 

Hi,

Thanks David for your feedback.
The flow meter is a turbine flowmeter with mechanical counter (flow totalizer). The transmission between is mechanical and maybe with some magnetic link. There is also an impulse type connection to an electronic flow indicator. With the electronic indicator was no problem as the uncompensated flow totalizer on the mechanical counter matched the uncompensated value on the electronic indicator.

I am so frustrated as I had deeply investigated and found more clues that prove that the flowmeter had error. I have received also some history files from the electronic indicator (actually is a data logger/flow calculator special for fiscal measurements). I have compared the data in this files with our internal SCADA system where I have all our gas flowmeters historized. Without any doubts the error is the same comparing with any of our gas flowmeters.
And in the last weeks from the moment when the fiscal flowmeter was replaced there is a difference of no more than 2%.

Now, I am thinking to send the flowmeter to the manufacturer for investigations of mechanical part (transmission, counter), but before doing that, I am looking for some similar situation.

Best regards,
Petre

 

> The turbine meters that I'm familiar with are mechanical turbines, but
> electronic pick-ups and electronic flow indicators.

> Do you have a meter that is strictly mechanical? Mechanical turbine and
> mechanical counter?

I think you refer to the turbines where an inductive or standard reluctance pick off coils senses when blade tips pass the sensor. e.g. http://www.c-a-m.com/Forms/Product.aspx?prodID=0fb193eb-0330-45f8-9707-ea428d60c830

but for gas flow mechanical turbines are common e.g. http://www.stream-measurement.com/5471/categories/546/turbine-gas-meters.aspx

Most such meters will under-record.
I can't think that the mechanical drive would result in an over-reading. More likely is a that the rotor is over-speeding for some reason. That suggests swirl or similar.
But either this is due to an installation fault which either exists or doesn't or someone has done something and then undone it. I don't see how dust can account for it.

But, when these error readings occurred do we assume the on site meters collectively showed normal flow patterns or was there something abnormal? i.e was there unusually high gas usage at the time?