Sample pipe for water coming off of high point of process line?

[m9dave]

Hello all,

I have encountered a situation to which I am not sure what equations apply.

I have a pump discharging to a 4" stainless pipe with several elbows to the top of a tank which is vented atmospheric pressure, and provides suctions to other pumps in downstream processes.

The construction group has installed 1/4" tubing to our sampler via a tap coming out of the top of the 4" stainless pipe immediately before the elbow where it dumps down into the tank. I have attached a rudimentary P&ID, and have isometrics available so I can calculate friction losses in the piping.

How can I tell if I will have flow to the sampler with this arrangement? I do not even know how to calculate if there will be water at the top of the pipe in this arrangement given that it is dumping to the tank immediately downstream; for all I know there will be an air gap here. Should the sample line be installed either at the bottom of the pipe or at a lower point, instead of taking off the highest point in the system? Sampler requires 0.5 gpm.

Thank you!

 

[bimr]

1. You will want to minimize the length of the sample pipe as much as possible. The sample pipe may be subjected to the elements (freezing for example). Don't see why the takeoff has to be at the tank top, why not after the pump?

2. There has to be some type of drain at the sample collection point.

3. You will need some type of backpressure at the tank. Otherwise, the elbow will drain into the tank when the flow stops or you may not get flow through the sample tubing. You can use an automatic valve to close on pump shutoff.

4. Are there any concerns with chemical reactions in the tubing with the sampled materials?

http://sentry-equip.com/Public/BrochuresandDataShee/GuidetoGasLiquidSlurry19.1.0rev1web.pdf

 

[m9dave]

Hello bimr, let me clarify please. This is being constructed by an EPC at a new facility.

1. Great question and one we are asking now. It does not seem correct to us that the sample would come off the top of the pipe at the absolute high point in the system, rather than lower down where it is guaranteed water flow. I do not know how to calculate whether there will be water up to the top of that horizontal pipe or if instead there will be an air gap at the top as the water is pouring down into the tank immediately downstream.

2. The flow goes through a turbidity meter and then to a floor drain, not shown on my basic drawing.

3. Will not need flow when the unit is not in service.

4. No. This is city water being pumped through an ultrafiltration unit and then to the tank in the drawing. The tank provides suction to pumps for reverse osmosis and then electro deionization after a further downstream break tank.

 

[bimr]

For a water treatment plant, you don't need to take the sample from the top. A sample taken after the pumps or off the discharge pipe (if the pipe is close to the sampler) are adequate.

The sample take off point will need periodic maintenance and it will be easier to service on the ground.

There should be some type of solenoid valve on the take off point to stop the water flow when the system shuts down.

A turbidity sampler at this point in the process is somewhat of an overkill and a waste of resources. After an ultrafilter, there will be essentially be no turbidity in the water.

 

[m9dave]

Hey all, thanks for the answers so far.

The system is currently already constructed as stated and we are preparing to assist the manufacturer commission it within the next week. I am wondering how to calculate whether the current system will work as installed or if there will not even be water at that sample point as it is dumping into the tank and I believe this portion of the pipe may be partially air and partially water.

Thank you

 

[bimr]

Obviously it will not work, the tubing needs to be full.

Install an orifice plate or backpressure valve on the discharge to provide the backpressure for the sampler.

 

[LittleInch]

I've never used it, but looking at similar threads here, look up Froude number to see whether your pipe will be full or not. What is the flow rate / velocity of the water in this pipe?

As bimr says, the simple solution without changing anything is install an orifice plate, but you are correct - this is a really stupid place to put this and lower down just after pump outlet or much closer to the sampler to avoid getting incorrect results.

Your other option might be to install a pitot tube to get more pressure.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.