Isokinetic Sampling of Sand in a Liquid Stream

[Nosey]

Ladies & Gents,

I'm currently working on a system for removing the sand from water. The pipes are all flowing fully flooded with a small gradient. When looking at sampling the stream, all the references talk about isokinetic sampling being required for gas streams with no reference to it being required for liquid streams.

Do I need to design for isokinetic sampling in a liquid stream? If not, why not?
Will simply sticking in a quill to the centre of the pipe be sufficient due to the relative momentums of the fluid and the particulate?

Thanks in advance,

Nosey.

 

[bimr]

What you want is a representative sample. How you obtain that is a matter of conjecture. For example, the low gradient will allow the solids to settle to the bottom of the pipe.

There are a number of parameters involved such as fluid velocity, particle size and shape, particle weight, fluid viscosity, temperature, velocity distribution, etc.

Sticking in a quill to the centre of the pipe is probably the best option. Take a number of samples at different times and blend the samples to try to get a more representative sample.

 

[cvg]

consider a bulk sample, capture the entire flow for a short period otherwise with low flow rates you may not get adequate samples.

 

[Nosey]

Good point bimr.

So, if we managed to configure a vertical section and aim at the centre, it should be pretty close to representative.

I suppose my real question is, does the velocity of the sample stream relative to the main stream make a significant difference to the sample accuracy. It appears to be significant for gases (see

http://www.photometer.com/en/abc/show.html?q=Isokinetic sample

for an example).

 

[bimr]

Yes, the velocity makes a difference. Unless the velocity is less than 3-4 ft/sec (water applications) you will have material settling out.

 

[SepiaTones]

I think not only is the velocity important, but so is density. Recall that isokineticity is important for stack testing because of the density difference between the particulate and the gas. If there is not much difference, there is no need for isokinetic sampling. For example, you don't need to do isokinetic stack tests for gaseous compounds such as CO, NOx etc.

So if the sand is not much heavier than the water, the need for isokinetic sampling is diminished. Or so it seems to me. I'm only used to stack sampling.

 

[ccfowler]

What stage(s) of your cleaning process is (are) of concern?

I expect that you probably have a much greater proportion of larger particle sizes in the "dirty water" going into your process and only the smallest particle sizes surviving into the "clean water" leaving your process. It seems likely that there may be a significant proportion (on a total mass flow basis) of the sand rolling or dragging along the bottom of the pipe in the "dirty water," so sampling that doesn't include the lower portions of the flow may miss important proportions of the sand flow. If your process really does clean the sand from the water very effectively, then the very fine particles that remain in the "clean water" may behave almost as though they are dissolved rather than suspended in the water.

I suspect that you may need to do some serious experimenting to determine which sampling techniques will provide the most meaningful information for your needs. Most likely, isokinetic withdrawal of the same will not be a dominant concern.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.