Location of differential pressure sensor

[Takket3]

Hi… question…….

I’m looking to locate a differential pressure sensor for a secondary pump VFD in a chilled water system. The chilled water system serves process equipment and runs 365 days a year.

I know that typically you want the sensor across your most remote piece of equipment to make sure you have enough DP to move water though the control valve, coil, etc. In this case my most remote unit is one that only uses chilled water in the summer. In the winter time this unit’s chilled water valve is closed.

Father upstream there are other units the serve a process and use chilled water all year long.

I know in the summertime I’ll be fine but in the winter time, when my remote AHU chilled water valve is closed, and my process equipment still wants chilled water, will I run into problems? I would think I’ll be okay because the pressure in my supply and return headers is still going to fluctuate at my remote AHU even if its chilled water valve is closed, right?

Basically my question is, is it okay to leave the sensor at my remote AHU or should I locate it farther upstream before the unit that uses chilled water all year?

And also, at the "remote AHU" there are taps off the main header with shutoff valves. Is it okay to put the sensor after the shutoff valves instead of in the main header? If I do this I only have to drain the piping to the remote AHU instead of the whole header to make the pipe tap. The pressure should more or less be the same in each place, right?

 

[mk3223]

As you described, in the winter time, the pressure point at the "remote AHU" will be the pressure at the dead end, which is equivalent to the point where the pipe is branched off on the loop. In this case, the system may have an issue of the pressure control in the winter.
It's a good practice to have an isolation valve for the pressure sensor or gauge to facilitate the installation and maintenance.

 

[LittleInch]

can you sketch a schematic of the system so we can understand what you mean.

Is there any height difference between the rest of the system and the remote AHU?

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

 

[Takket3]

Sketch attached as requested. There are a couple of other branches to AHUs between the process equipment and the last AHU which I didn't show, but you get the idea.

The entire system is at the same elevation on one floor of the building.

 

[quark]

If you don't have any sub branches that cater to a group of equipment, having DP sensor at the end of headers is good (not at AHU, after the valves). If there are sub branches, then this location is not right. Generally, if you install at 2/3rd distance from pump to first branch off and set the point corresponding to full flow condition, it serves the purpose.

 

[SAK9]

Leave the existing DP as is and install an additional DPT near the process load. Poll the two DPs and choose the greater of the two to control the pump speed. If that is not possible use the thumb rule of 2/3rd distance from the pump on the main header to locate the DPT

 

[georgeverghese]

In a new plant, the ideal location would be just upstream of the isolation valves leading to this summer only operating AHU. But in this brownfield modification, I can see why you'd want to have this dp cell downstream of the isolation valves leading to this AHU - this avoids having to shutdown the entire plant, and drain these main headers just to make this taps

This now means that you'd have to install another set of maintenance isolation block valves further down from where you propose to have these taps now.

Agreed, the location of the dp cell in the area close to the remote summer only operating AHU will serve for both summer and winter.

 

[Takket3]

Thanks all for your help. There is a process branch just a few feet away from the most remote AHU branch so we are just going to put it in the process branch, then open the chilled water valves, ramp up the pumps, and measure the DP to determine our setpoint.