Hydronic Pump VFD Differential Pressure Sensor location when large pressure drop is near the pump

[BronYrAur]

I am obligated to put a VFD on a hot water pump in a situation where the application is bad. I personally don't think it is going to reduce speed very much at all, but like I said, I am obligated. Common sense is not part of this equation. Can you help me choose the best location for the DP sensor.

I have a hot water system pump that serves an air handler in the boiler room and VAV coils throughout the building. The air handler is piped direct-return, and the VAV's are piped reverse-return.

When I calculate head in the AHU loop alone, I come up with 33'. If I calculate the VAV loop alone (even with full flow through the AHU), I get only 15'. The the pump is sized for the total flow of 72 GPM at a head of 33'. Approximately 75% of the flow goes through the air handler right in the boiler room.

The pump curve is pretty flat. Since the air handler coil and immediately piping constitute almost all of the pressure drop, I was going to place the DP sensor in the boiler room somewhere around the AHU coil. With the flat curve, the pump won't slow down much at all.

The control contractor on the job has the "rule of thumb mentality" that the DP sensor needs to be placed 2/3 of the way down the piping. I don't see this as an option. If I placed the sensor out in the system and tried to maintain 10' (or whatever) across the VAV boxes, I would have no way to guarantee flow through the AHU coil, would I?

Any suggestions on DP sensor placement and setpoint would be appreciated. See attached.

 

[MTES93]

Short answer is you are correct. In order to guarantee flow in the ahu coil you would have to set the remote dp transducer to 33 feet out in the system. Which would cause your balancing valve set lower (closed) to 'eat up' this excess head at the boxes. And I would see the pump not modulating much, sort of like putting a VFD on an open loop well pump where all the 'head' is static lift. The DP transducer should be placed at the hydraulically most remote coil, not necessarily the furthest away. In this case your AHU and not your longest run.

There maybe different options of controlling the pump speed rather than a constant DP setting. You can do a reset based on AHU demand, or possibly use a sensorless type pump that can auto-adapt to the demand...maybe...that would require a bit of digging.

 

[EnergyProfessional]

use static pressure reset and you see how pump speed decreases. your design head is at design condition (plus safety). in real life under part load it is much lower.
i assume these are 2-way valves?

With static pressure reset (based on valve position), you can measure pressure over the pump.

 

[LittleInch]

I really can't see how this system works now, never mind with a VFD.

Unless you have flow control on your loops, those flows will not happen. The system will just hydraulically balance itself and the flow through the high DP section ( your AHU in the mech room) won't be 51, but something lower and the flow through the VAVs more than 21.

If your pump curve is flat, why does anyone think a VFD is going to help anything.

Does the system work well now? If not what is the problem?

If common sense isn't present then just place the DP sensor at the pump outlet and be done with it. What I really think you need are some flow regulators and a flow meter to use to control the VFD to your total of 72. Is the temperature control by on/off or by bypass of the heating elements?

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

 

[BronYrAur]

the scope is very rigid and limited. It includes a boiler replacement, pump replacement with the model that I attached, changing a 3-way valve to a 2-way valve at the air handler coil, and installing a vfd on the pump.

as much as I would like to redesign the whole concept, I don't have that possibility here. so I have to put in the vfd, even if it doesn't do a darn thing. There are circuit setters at each coil that I did not show, so we will be able to get an initial balance. I'm just trying to find the optimal place to control the vfd giving the circumstances

 

[LittleInch]

Ok, so you have variable flow if the units turn on and off, but flat out you're doing 72 gpm?

Your key pressure drop as far as I can figure out is the large unit, so make that the control point / DP. If you need 33ft to make it work, but the other units need less ( thought they will be more once you balance the system), then make that your control point.

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

 

[lilliput1]

The AHU only need heat for preheat whereas the reheat coils in the VAV boxes need heating whenever the AHU is on. So call for reset schedule of the DP set point based on outdoor air temperature. Control for: 33 ft wg differential when OAT is 50F (or less by taking credit of heat from return air)and below; control for 15 ft differential when OAT is greater. Make sure you maintain minimum flow through the boiler whenever it is on.

 

[BronYrAur]

I like that idea lilliput1. How about one step further? What about resetting the DP schedule in a linear fashion from 33' to 15' as outside air changes from 0F deg (my design air temp) to 50F deg? It might take some trial and error but that would at least allow the drive to perform its function.

 

[LittleInch]

But if you set the PD for only 15 ft, but your circuit setters are based on 33 feet then will still get the flow you want?? Not sure how these circuit setters are set up.

A full layout drawing showing where the flow control is and how it works is needed before anyone can say whether the differential pressure can change or not.

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

 

[MTES93]

LittleInch is revealing the reason (or at least one of them) that autoflow balancing valves were developed. To overcome the manual balancing valves 'fixed' head based at design flow. There are other considerations when using autoflow valves (like two-way modulating control valves and them fighting)... Are your current balancing auto or manual or non-existent like your sketch shows?

 

[lilliput1]

Omit the circuit setters in the VAV box reheat coil. The reverse return is self balancing and the small valves on the reheat coil can throttle to provide 33 ft differential when required while meeting the reheat coil heating duty.

 

[EnergyProfessional]

if DAT is controlling the valve, it will reduce water flow. This most likely is the case in an AHU, but less common in a VAV coil.

Of course flow will be reduced in 2-way valves when valves are not fully open (partial load). What else do you think a throttling valve does?

 

[Drazen]

how can it work? performance would mostly depend on terminal valve characteristics, and you did not tell us about them. pump cannot do anything alone.

 

[lilliput1]

The space temperature sensor would modulate the VAV box damper and hot eater reheat coil valve in sequence to maintain space cooling temperature set point. On a call for cooling the reheat coil flow control valve will close and the VAV damper will modulate open till maximum flow set point. On a call for heating the VAV damper will modulate close till minimum volume set point. After set dead band the reheat coil flow control valve will modulate open to maintain space heating temperature set point.