WSHP loop pressure diff.

[dapper]

On a water source heat pump loop with water-side solenoid valves and VFDs on the loop pumps, where is the preferred location for the differential pressure transmitter to control the loop pumps?
I have seen 'across the pumps' and also '2/3 distance from the pumps'. Any thoughts and experience with this?

[Tubes110]

My vote is 2/3 down the pipe. We just did a job at a school that had 7 HP pumps and about 15 heat pumps. The original install did not have VFDs or valves of any kind on the heat pumps. The pumps ran 24/7 as well. We put modulating valves on the heat pumps and VFDs on the loop pumps. The valves are indexed based on the temp of the water coming out of the heat pump and the loop pressure setpoint is indexed up and down based on the position of the valves. It works beautifully and the building's electric bill was nearly 50% lower than it was in previous years.

Reuben

[twisted pair]

2/3 sounds reasonable, however I am thinking closer to the last WSHP coil.

Is this a single story building or does it have multiple floors?

[dapper]

multiple, 3 floors and basement

[twisted pair]

dapper,

I think I would locate your DP sensor closer to the furthest WSHP, or the coil that requires the greatest differential pressure. 2/3 may not be enough in my opinion. From experience a 3 story building we did once had issues with units tripping on high head pressure. We were able to use software time delays once the isolation valve opened before the compressor would be enabled (I think 45 or 60 seconds) which ensured we had adequate and full flow through the condenser coil. We had to tweak the condenser water pump VFD PID for awhile. Also we used a fairly rapid scan rate on the AI for the loop DP so the VFD PID could get the process variable quick enough to ramp the PID.

[lonboy]

And years down the road how will you (or the next guy) find that failed pressure transducers somewhere out there on a remote piping loop?

We did a similiar 100 WSHP hotel years ago in Santa Monica with Landis & Gyr System 600 and put the pressure transducers at the supply/return piping mains on the roof. A VFD maintained minimum differential pressure by varying pump speed. Setpoint on the differential at 10PSI. Worked great. We had one PID maintaining water temperature and one PID controlling water pressure. We also had a hot water boiler and a VFD tower fan on the loop. I don't think the boiler ever ran.

Then Landis&Gyr came in and stole the account after they refused to sell upgrades and replacement parts direct to contractors.

Of course if it's a T&M job and you will have the service contract for life - put em at the most remote heatpump and hope that the water treatment guy keeps the sludge outa the system.

Cheers,
lb

[twisted pair]

Note to self: Tell all General Contractors not to conceal any ceilings. Don’t want some tech to have to get a ladder and look for something.

Where would you locate a supply duct static pressure sensor? On the roof

That’s what as-builts are for.

I would locate it where it will provide the best feedback to the system, not the most accessible location 15 years from now when it might fail. This is for condenser water. My experience is all the individual refrigerant circuits are more critical since this sensor will effect head pressure control than a convenient install spot. Just my opinion.

[lonboy]

Twisted,
Yes, I might install the static pressure transducer on the roof in a mechanical room control enclosure and pull a 1/4" pneumatic tube (in conduit) 2/3 of the way down the supply duct run connected to a pitot tube. Or, maybe I'd install the transducer in a panel in a phone closet somewhere on that floor. I don't think I do that with a water pressure transducer. Water leaks can be costly.

During building commissioning, how about trying different set points for differential pressure while you watch your gauges at that farthest heatpump? If there are Pete's plug's you should also be able to monitor system pressures out there. Get to know the system how it really operates - not how some engineer predicted it to run!

In the real world, the ceiling will be concealed hard-lid and the as-builts (if there are any) will be anything but. And, the system will go down on the hottest day of the year during the annual sales meeting while you're on call but out skiing on the lake and your backup tech has no ladder.

If you're worried about individual refrigerant circuits getting enough flow, maybe you should install head pressure transducers at the farthest heatpumps and use those values to modulate the VFD for the loop pump.

Cheers,
lb

[crab master]

In working with a balancer that I have a lot of respect for he told me the best location for a DP sensor is at your furthest run across the supply and return lines near a 3 way valve or the furthest 3 way valve with the least amount of differential pressure. Here you maintain the minimum differential pressure and flow required by that particular valve, by taking flow and pressure measurements at the valve and shutting down all the other valves. You then ramp down your pump to get just above the minimum pressure and flow required and that is now your system differential pressure setpoint to maintain. If you always have that minimum pressure and flow at your furthest valve/one with the least amount of pressure differential then you will have enough pressure for the rest of your valves and you don't have to overcompensate in your system differential pressure setpoint.

[osiyo]

In working with a balancer that I have a lot of respect for he told me the best location for a DP sensor is at your furthest run across the supply and return lines near a 3 way valve or the furthest 3 way valve with the least amount of differential pressure. Here you maintain the minimum differential pressure and flow required by that particular valve, by taking flow and pressure measurements at the valve and shutting down all the other valves. You then ramp down your pump to get just above the minimum pressure and flow required and that is now your system differential pressure setpoint to maintain. If you always have that minimum pressure and flow at your furthest valve/one with the least amount of pressure differential then you will have enough pressure for the rest of your valves and you don't have to overcompensate in your system differential pressure setpoint.

Agreed, probably the best tactic. If possible. Not always possible depending on time and money (what customer is willing to pay for).

[twisted pair]

lonboy,

I have no problem running remote tubing or wiring, but let’s be practical. Ceilings contain VAV boxes, terminal unit controllers, hydronic isolation valves, control valves, electrical junction boxes, fire sprinkler valves and tamper switches, smoke dampers, and so on.

So if I understand you correctly all of these things should be exposed so you can locate them. After all in a building with horizontal water source heat pumps aren’t they all above the ceiling?

I will stick with my original suggestion to locate the DP at the last WSHP. How it gets installed is a matter of preference.

Installing a static pressure sensor with remote poly tube is one thing, wet transducers should be installed in the piping at the location. That’s my preference. I think dapper can figure that part out.

Your making an assumption that there is even a penthouse. Every building does not have a penthouse. Most WSHP buildings around here are under 4 stories and some have the cooling towers on the ground.

So if I understand you we should mount all the WSHP controllers in the penthouse so someone can find them and just run a few thousand feet of I/O wire?. You can install field devices wherever it makes you happy and I will put them where it makes me happy so we can move on.

Honestly your point about the ladder is pretty stupid. What if the damn heat pump needs to be looked at? Maybe a filter change? Or should we stick with exposed ceilings. I have no idea why this a big deal to you. I mean you’re not suggestion that the heat pumps be installed in the penthouse so a tech can find them without a ladder?

Anyway someone asked for an opinion and I gave it, I have no time to get in a pissing match over where a transducer gets installed, because I really don’t care. Have a great day!!

[dapper]

Thanks for you input guys. My thoughts were this; 2/3 downstream makes sense in the same way that a static pressure sensor is mounted 2/3 downstream, but typically circuit setters are set at the pumps for a given flow so why couldn't you control the 'flow' at the pumps?
I already have taps in the supply and return lines at the end of the second floor loop but that is several hundred feet from the mechanical room in the basement. My thought was to wire the transducer to a local controller and share the data on the network with the device controlling the pumps. I am now wondering if the lag in the data transmitted to the pump controller will cause the VFD to hunt or otherwise not perform properly This network will be approx. 150 devices on MS/TP.

[MuddBuddha]

It can be done either way.

Most budget minded engineers prefer across the pumps for the sake of keeping the control wiring to a minimum.

It also helps when trying to locate the pressure switch once it has failed 8-10 years in the future.

[twisted pair]

I guess it’s a matter of preference. 2/3 for a regular hydronic loop would be okay by me, or the method crab master listed. The difference in my opinion is this; any other hydronic system (hot water or chilled water) if it takes a little longer for the pump to ramp to keep the DP setpoint the worst case scenario is the coil will not have the GPM it needs. On the airside with static pressure, some VAV boxes will modulate open, however with WSHP condenser water, my concern is these are solenoid valves or isolation valves not modulating devices or water regulating valves. If the WSHP’s at the end of the loop do not have sufficient flow through their heat exchanger you risk having the WSHP units trip on high head pressure which may result in a manual reset or having to cycle power.

The energy savings by having variable water volume is great, however protecting all those individual refrigeration circuits would be my 1st concern, energy savings 2nd. I would be more comfortable with the DP sensor at the farthest run or last WSHP. I am sure it will work out.

Dapper I am curious is this new construction or a retrofit of an existing WSHP system with constant speed pumps?

Good Luck

[dapper]

new construction

[crab master]

Dapper - I haven't done an MSTP network but almost all my Water DP sensors are located on local controllers nearby the actual location of the sensor that are on their own power circuit/shared with others that are very unlikely to ever be powered off. I am passing the DP over the LON and then picking it up at the corresponding vfd/controller. I have not seen an issue with the lag time with the data transmitted and I am not using acknowledged bindings, I am using repeated even on the DP output. It has worked very well.

On all the comments on the location and finding the sensor I have been there, but that is one reason proper as-builts need to be done. We also label the ceiling at the location of the DP transducer, but most are located near enough to reheat boxes/fan coils or other equipment that will need some kind of regular service that the service guy/other should be able to know where this is at. I have seen locations where it is off all by its lonesome and that makes it a real PIA to find. Also I quit using DP transmitters w/o display - it has saved me more troubleshooting time overall that it makes up for the extra cost. Here's another thought put a label on the vfd and/or pump to the location of the DP sensor.

IMO - for most energy savings and the to have the most confidence that you are going to have enough flow put it near/at the furthest point. Again I am sure many of you have seen overcompensated pressure setpoints to compensate for that last box/coil, that is very likely an energy waste, unless of course the individual has spent enough time ensuring that actually is the minimum setpoint they can get away with.

[osiyo]

On all the comments on the location and finding the sensor I have been there, but that is one reason proper as-builts need to be done. We also label the ceiling at the location of the DP transducer, but most are located near enough to reheat boxes/fan coils or other equipment that will need some kind of regular service that the service guy/other should be able to know where this is at. I have seen locations where it is off all by its lonesome and that makes it a real PIA to find. Also I quit using DP transmitters w/o display - it has saved me more troubleshooting time overall that it makes up for the extra cost. Here's another thought put a label on the vfd and/or pump to the location of the DP sensor.

IMO - for most energy savings and the to have the most confidence that you are going to have enough flow put it near/at the furthest point. Again I am sure many of you have seen overcompensated pressure setpoints to compensate for that last box/coil, that is very likely an energy waste, unless of course the individual has spent enough time ensuring that actually is the minimum setpoint they can get away with.

Correct. I've seen a LOT of installations where the DP is at or near the pumps and the setpoint used was higher than need be. Maybe wasn't set so high at first, but after some operational time and the discovery of problems associated with inadequate supply ... SP got bumped up.

Getting the DP 2/3's of the way down the loop, or even better ... locating it at the farthest critical demand point, and then having a balancer do his thing and tell you the lowest DP you can maintain and still satisfy demand at that point ... has always worked the best in the installations I've done. That is, to set up things so that one is meeting demand, but no more than that.

As concerns finding the DP at a later date. It really should be a non-issue if one is doing a professional installation job. We mark locations of such items on plans and drawings, a copy of which goes to the customer and we also keep file copies. Ceiling grids are marked and labeled with a professional grade labeler.

And as part of the BOM package we give the customer there is a list of all controllers, and remote items such as actuators and sensors which might be hard to locate later; which lists the item and then gives a written description of where the item is located. We just call it the "Equipment Location Schedule".

i.e. HW Loop DP Sensor - Overhead of hallway outside of the door to Room 205.

Or,
HW Loop DP Sensor - Near AHU 5 in the machinery room.

And on the points sheets for each controller, a copy of which is always left inside the controller cabinet plus a copy of each is included in the BOM, an annotation is made as to where such things as a DP sensor which is connected to that controller is located (if their location would not be obvious).

As concerns pulling long runs of poly or copper ... we try like the devil to avoid that. Too problematic. Particularly in the case of DP sensors.

[Chris_Worthington]

I'll take it another step further....

We have been adding the location/operations of, etc to a note posted on the wire sheet in many cases and/or a job note page contained within the front end.

Nothing to lose that way.

[mechdorn]

The T&B guy is correct. On the other hand, if he adjusted all the valve circuit setters to design flow, then cranked up the VFD to match full flow, and marked the DP setting at any location, what's the difference were it’s measured. As far as accessibility, there is probably one or a couple on each site and they usually need little care. Go find and fix it.