Sizing a booster pump in a closed hydronic system

[BronYrAur]

I need some help sizing a booster pump for a closed-loop chilled water system. 3 sketches are attached. I have a large chiller water plant that has 6 secondary pumps is parallel. The pumps each have a VFD and the program maintains DP settings out in the system.
Originally there were only 2 loads and there were never any problems. A third load was recently added with an extremely long piping run to get to the load. This third load is causing problems on design days. On other typical days when flows are not high, there are no problems. Load #1 is the “driving” force for the secondary pumps. Maintaining the 30 psi setpoint at load #1 provides more than enough pressure at loads #2 and #3. Only 3 of the 6 secondary pumps need to run with VFDs in the 60% range. See first sketch.

The second sketch shows what happens on a design day. Since the piping run to Load #3 is so long, in order to provide 6 psig at DP3, my pressures elsewhere are much higher than setpoint. And to make matters worse, all 6 pumps are needed at 100% VFD to meet the DP at load #3. The total combined flow of all loads is not close to the combined pumping capability at reasonable heads. However, I am at the limit of pump head, so additional pumps in parallel essentially do me no good. It is the large pressure drop to get to Load #3 that is killing me, not the flow.

I would like to put a booster pump in the pipeline serving Load #3, but I am not exactly sure how to size it (sketch 3). I know if will have to meet the design flow of Load #3, and it should be variable speed. I assume the best point of control for this pump would be DP3. Physically, the best place to put this pump would be near the Load #3 location. So essentially half of the piping run would be on the booster pump suction and half on the discharge. The load itself, which consist of a heat exchanger bank, control valve, and circuit setter would also be on the discharge side of the booster pump.

My biggest concern would be that this booster pump not interfere hydronically with the secondary pumps and create issues at Load #1 or load #2. So my thought is to size it just for the additional pressure loss in the long piping run. In other words, on a design day in order to maintain 6 psi at Load #3, Load #1 is seeing 50 psi. That is 20 psi more that its setpoint. If I remove that 20 psi from the secondary pumps, I would need to pick it up with the booster pump. So my initial thought is to size the new booster pump for design flow @ 20 psi head. If I include the actual load #3 itself, 26 psi. That way the booster pump is only helping me overcome the additional friction loss in the long piping run and is not affecting the rest of the system …… I think.

Thoughts?

 

[LittleInch]

That looks a pretty good assessment to me.

I would be much happier with the pump located just after the branch to load 2 and controlling on the DP3 set point.

If you have it at the end of the initial run and for whatever reason load 3 is running at a higher load than 1 and 2, then you run the risk of starving the pump at the end of you long run, because the system is only supplying enough pressure for the controlling input (what you refer to as the "driving" input).

The system should just balance itself. As your draw flow at the load 2 branch connection with your pump this will lower the pressure at load 1 or 2 and the pumps then speed up to increase pressure and flow.

The input pressure to your pump should be a fairly constant 28-30 psi according to your sketch so needs to be sized for max flow for load 3 at about 25ps differential and control it on DP 3.

You could add a controller and non return bypass so that if your new pump at min rpm your DP3 is > say 10 psi, then the pump stops and you revert to your current system and your new pump only cuts in when DP3< 10 psi AND DP1 > say 33 psi?





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

 

[BronYrAur]

LittleInch,
Yes I was wondering about the case when the existing secondary pumps can easily Supply load 3. In that case I don't really need the booster pump. So a bypass around that pump with a check valve that would only allow flow in the proper direction would allow the secondary pumps to supply the load and when the booster pump is not running. I was thinking of putting the booster pump on a vfd anyway but this might allow me to completely shut it off at times.

Regarding your other comment, I really can't put the pump in the branch line where the other loads take off. I actually didn't draw it very well but all three loads branch off at approximately the same Junction location. The only place where I really have physical room for this booster pump would be almost all the way down near load 3. Do you see that as being a problem?

 

[LittleInch]

Well as I said it's not ideal simply because of the risk that if the load on the other 2 is low compared to the load 3 then there might not be enough pressure to get to the pump, but you can probably work out what the pressure drop along the line to the pump is at say 80% max flow and an inlet into that branch at 30 psi. If its still positive pressure then your new pump is probably Ok, just not the best location to put it.

should still work and if its VFD will be able to get you at least part of the way down from your max head on the main pumps if you run into low inlet pressure problems.

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

 

[lilliput1]

Set the 6 pumps to just satisfy both DP2 and DP1. Control the booster pump to run if DP-3 is less than 6 psig and control it to maintain minimum 9 psig at DP3. Size booster pump to boost DP3 gpm 3 psig.

 

[lilliput1]

Correction - size the VFD booster pump for 50-30-6 = 14 psig at full DP3 branch total gpm. Shut booster pump off if DP-3 is less than 6 psi. Reset DP3 set point to 9 psig. To be sure calculate the branch total pressure drop using total design flows & actual pipe sizes, lengths & fittings.

 

[lilliput1]

No - size the VFD booster pump for 50-30 +(9-6) = 23 psig at full DP3 branch total gpm. Shut booster pump off if DP-3 is less than 6 psi. Reset DP3 set point to 9 psig. To be sure calculate the branch total pressure drop using total design flows & actual pipe sizes, lengths & fittings