We have a Foxboro CFS30 coriolis meter. We have calibrated the meter for density and it works well with a velocity of 2.5 to 3.5 m/s (8-11 feet/sec). When the velocity gets above about 4 m/s (about 13 feet/sec) the density reading shifts and no longer follows the calibration. Have others seen this happen? Is there anything we can do other than changing to a larger meter?

 

What is the material you're measuring? If it is a slurry, you might be
experiencing de-watering effects.

Walt

Spitzer and Boyes LLC
837 Suncrest Drive
Aurora IL 60506
[email protected]
www.spitzerandboyes.com

 

It sounds like cavitation.

As velocity increases the pressure decreases inside the sensor.
Typical effect would be low reading in density to very unstable at high cavitation. Increased back pressure could help to avoid cavitation.

 

Might very well be. Still would like to know what the product is.

Walt

Spitzer and Boyes LLC
837 Suncrest Drive
Aurora IL 60506
[email protected]
www.spitzerandboyes.com

 

The simple answer is that there ought not to be a direct effect of flow on the density measurement.

Simply, resonant frequency changes with the mass of the system and the mass changes with the density of the fluid contained.

But there are some other causes of changes in resonant frequency that have to be compensated for. These include temperature and pressure effects and, for low density fluids, velocity of sound effects. Viscosity can also theoretically have an effect (it is noticeable with some density sensors such as the tuning fork and open tube types sensors) but it is usually very insignificant for this type of sensor.

The Solartron density transducers, also used for slurries, do not however have any correction for flowrate. It might be tempting to suggest that this is why there is a comparatively low flow rate limit for these sensors but the flowrate limit is not related to its effect on the measurement - the MassMaster 40, a coriolis meter derivative (it had an extra pick-up to detect phase shift) was rated for up to 40m3/hr and even at these flowrates there was no significant effect of flow on density.

So the probability is that we are looking for a secondary effect.

This might be a flow split problem... if the slurry is not fully homogeneous at higher flowrates, for example. Possibly there are some pressure effects (optimised for flow rate measurement the temperature and pressure effects on density in coriolis meters may be quite significant and some coriolis meters do suffer from pressure pulsations... perhaps the pump pulsations are more significant at higher flow-rates?

Or maybe it is something to do with the non-Newtonian behaviour of slurries.

 

I need to eat some humble pie here.

I have just come across an old manual for the 7860 2" single straight tube coriolis meter and it appears it did suffer a flow effect on density.

The 7860 is rated for 150ton/hr (I spoke earlier about the 1" rated for 40ton/hr) and there apparently is a small flow effect on density. This is 0.1kg/m3 at around 30ton/hr rising more steeply at higher flow rates to 2kg/m3 at 140-150ton/hr.
The electronics automatically compensated for this effect.

To what extent this is a function of the coriolis effect on the resonant frequency, or to what extent it is sensor technology dependent, I couldn't say.

We might best assume there is an effect to be found in coriolis meters that may be a function of the coriolis effect and/or it may be technology dependent.

Taking your 4m/s flow velocity, in the 7860 this would equate to around 25m3/hr at which flowrate it would suffer a density effect (offset) of around 0.05kg/m3.

In a smaller meter the effect may be more significant but in a smaller meter the density accuracy of the CFS30 is not so good; 9kg/m3 in the 1/4" and 3kg/m3 in the 3/4".

For 25mm to 80mm the density accuracy is then 2kg/m3 so just how significant will a flow effect be?

I would suggest checking the manual and seeing just what corrections they do apply and their magnitude....

But where there is any effect on the measured parameter encountered in normal operation, there ought to be, as in this case, a factory calibrated correction that is automatically applied.
Hence , if there is an effect and it has been calibrated at the factory and configured for auto-compensation the user should not see any flow effect on density via this mechanism.

Note: the stated density accuracy of the 7860 is +/-0.2kg/m3. Hence at 4m/s the flow effect on density is only 0.05kg/m3 and is relatively minor. But as flow increases, it does start to become more significant (compared to the density accuracy declared) until a point at which it is significant enough for the correction to be applied.
It may be that the flow effect on density is not significantly greater with the Foxboro and in which case, compared to their density accuracy, it may be considered insignificant at your flow velocity.