Water Meter Design

[gomirage]

I need to install a water meter in a system composed of the water tank, a pumping station providing pressure to the water main (8 inch pipe). I want to know the best place, type of water meter to install on this system. Possible locations are between tank and pump station or on the water main after the pumping station. Anyone has experience on this? Any suggestion, resources on design of water meter are welcome.

chris

 

[semo]

Are you trying to meter the amount of water the pump station puts into the system?

Do you want the meter installed outside or inside a building/vault?

How accurate of a reading do you need (what is the meter's purpose)?

 

[gomirage]

I am trying to meter the amount of water coming in my distribution system. I am looking for the most cost-effective design (inside or outside building/vault). Right now i am prospecting all possible scenario. I will choose the best later on.

I would like to have it as accurate as possible. But can give up on some imperfections if necessary.

chris

 

[hydrae]

The best place to install a larger (magnetic, turbine, propellor, insert pinwheel) meter in with a long straight (5 to 10 pipe diameters) pipe upstream and no obstructions within one pipe diameter downstream, many inaccuracties come from installing the meter just downstream of an elbow, control valve, or pump discharge. other inaccuracies come from using the wrong size meter for the flow. The widest ranges are in magnetic type, having about a 30 to 1 turndown, most other meters will have a 10 to 1 turndown in the high accuracy range. If that 8 inch pipe is serving just a few houses even a mag meter will not be accurate below 50 gpm, so if you want accurancy in the low end, a compound meter will be needed, compounds are not as sensitive to upstream objects. the drawback to compounds is they have a lot of moving parts.

All meters will require maintenance at some point so it need to be in a place it can be worked on.

Hydrae

 

[KRSServices]

Is the meter to be manually read or integrated into a SCADA system? The meter will have to be properly sized to ensure that it is not too large, and therefore lost revenues.

Also, is the meter to be used for billing purposes or as an instrument for billing to your community from a provider (as in a regional system). In this case, you should also have an airtight agreement for the provision of water, including rates, deliverability and emergencies.

KRS Services

http://www.krs-services.com

 

[PUMPDESIGNER]

There are so very many considerations.
I can state my opinion though without full justification.

If you can afford to, reduce pipe size to increase velocity, meters just love velocity for accuracy. In this reduction go ahead and install your flow run (10 or 15 times diameter upstream, 5 downstream is close, ask manufacturer of meter). How much you reduce is the question. If you have a lot of flow time below .5 feet per second then squeeze that sucker down some more if you have the head to spare.

I would use mag meter or insertion mag meter. Some would feel comfortable with other types that are less expensive if they have experience with those, (I do not), but the mag will get it done for sure.

With most meters you can get any output you want.

PUMPDESIGNER

 

[quark]

PumpDesigner is right about the straight lengths and the velocity. If permanent pressure loss is not a concern for you, you can go as high as 8 m/s velocity with mags. You can get 0.5% accuracy with a mag if your intended turndown is not more than 30-35 (some manufacturers claim 40). Beyond that the inaccuracy rises asymptotically.

Endress+Hauser, Yokogawa, Fischer Rosemount and Foxboro all have free software for meter selection. You can lay your hands on them and they are very interesting.

DP meters have low turndowns and high inaccuracies comparitively. However, it's budget that all matters.

You can also check reverse flow with a mag (not all models)

Regards,

 

[gomirage]

I am trying to use the meter below. Some colleagues had used it before and seem to be satisfied. It is a turbine meter. I will probably a 8" on a 10" watermain.

Chris

 

[gomirage]

http://www.neptunetg.com/emplibrary/PSTURB112-10.pdf

 

[BRIS]

As to the location it doesn't matter whether it is upstream or downstream of the pump station. Pressures are less upstream and the lower pressure rating might give a nominal cost saving in the meter cost.

 

[semo]

As all have said, the length of straight pipe before and after the flowmeter is critical to the accuracy.

The Schlumberger/Neptune meters typically have a +/-5% accuracy. Chosing a meter in your flow range can reduce that to around +/-1.5% (so the manufacturers say). I've never compared them side by side to a mag meter.

I have used their as well as similar meters on many systems with varying results. You need to look at your flows and select a meter based upon them, not based upon the pipe size. You will more than likely have to reduce the pipe size as mentioned previously too. Schlumberger has charts outlining the pressure loss and accuracy of the various sized meters at various flows.

If your flow varies, you may need a compound meter which basically has a turbine meter for high flows and a disc meter for low flows.

If your waters have a high hardness and tend to scale, the meter will collect the scale and have to be replaced over time. That happens in my area on a regular basis. Typically the meters last 2-5 years; but, they are less expensive than say a mag meter.

Also as was mentioned, pay attention to where you locate the meter. You will have to work on it and keeping it above ground in a dry environment makes repair/replacement easier and will extend the life of the meter.

I particularly like mag meters; but, sometimes power is not available and in all reality, if you are using turbine/disc meters at the user connections, a higher accuracy main meter doesn't give you any better results. For treatment plant applications it can be different, you are basing flows on what the treatment equipment is designed for and you want higher accuracy for control purposes.