Flowmeter Selection

[AWloo]

Hi all,

This is my first post here, and I am at the beginning stages of replacing a currently installed flowmeter. We are looking at expanding the flowrate in the system and are needing to change our current installed flowmeter. Some background information on our requirements:

Fluid Temperature=150 Deg Celsius
Flowrate= 0-18 L/min (preffered) 0-12 L/min (mininum)
Fluid is water and or ethylene glycol at various mixture ratios. (ie non constant density)
Line size: 0.5" NPT or possibility for 3/4" NPT

We currently have a turbine meter installed that has a range of 0.38-11 l/min. I understand that turbine meters are not accurate at the lower ends and thus is a driving factor in replacing it.

I am currently considering magnetic flowmeters, positive displacement flowmeters, and other turbine meters. Due to the nature of the high temperature glycol we are trying to minimize pressure drop if possible.

Thanks for your assistance everyone!

 

[bimr]

The selection of the flow meter will be guided by the accuracy that you want. What accuracy are you trying to get?

 

[LittleInch]

Also flow range. Apart from start up, what is your lowest and highest steady state flow? does flow change rapidly or smoothly?

Pressure rating?

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[AWloo]

The accuracy I would be looking for would be +/- 1%.
The flow rates must be close to 0 and upwards to 18 l/min or 4.75 USGPM for my imperial friends.
-The flow rate would me maintained at the desired setpoint with minimal fluctuation. The response time of the sensor is not critical.

Cheers,

AWloo

 

[AWloo]

Sorry I forgot to include pressure, but that would be below 150 psig.

 

[LittleInch]

what are you measuring - volumetric flowrate or mass flow rate?

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[AWloo]

The application calls for volumetric flowrate.

Cheers,

AWloo

 

[zdas04]

You are simply not going to get 1% accuracy with a variable mixture. Even something like a Coriollas meter that purports to be a "mass flow" meter (because the frequency of the vibration of the pipe is proportional to the density in a constant composition flow regime) can't give you that kind of uncertainty.

With this kind of problem I tend towards simpler. I'd use a V-Cone and know the when the fluid composition matched the input parameter I'd get better than 1% uncertainty and other than that I wouldn't.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[AWloo]

Thanks for the reply zdas04,

Currently we are measuring turbine flowmeter which meausures volume not mass flow. Attached is a linearizer which provides a 0.1% accuracy. The current manufacturer of the valve does not have a size that matches are new requirement and we are looking elsewhere.

I will look into the V-Cone.

Cheers,
AWloo

 

[zdas04]

Nothing actually "measures" either volume flow rate or mass flow rate. They all infer a volume (or a mass) from something they can measure. The turbine meter infers flow from the count of revolutions of a wheel (since momentum imparted on the wheel is related to the angular velocity of the wheel, momentum needs a density, the electronics takes your composition and turns it into a density and uses that to get back to a volume, a lot of assumptions go into that conversion).

Talking about "accuracy" is quite imprecise. Typically we talk about "uncertainty", "repeatability" and "hysteresis". An equipment manufacturer that boils those three (measurable) components into "accuracy" (not measurable) would cause me to run screaming from their equipment. It is just smoke and mirrors. I've used turbine meters a lot and they definitely have a place. I spec them when I can tolerate an uncertainty of around 5%, a repeatability around 95%, and considerable hysteresis (it takes a LOT more force to start a turbine from a dead stop than to change it's speed once running).

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[AWloo]

Thanks for the response everyone and for explaining the inherit inaccuracies and sources of error in flow meters. For now I am looking for which type of flowmeter would work in this application, especially with the non constant density of the fluid.