Variable Primary Flow CHW and chiller interaction

[EloyRD]

Dear community,

I'm in doubt in the following aspect of a Primary Variable Flow CHW system :

The minimum flow of CHW through the chiller is provided by the lowest speed of the pump. Also, the chiller has a minimum load protection.

My doubt is if below the minimum cooling load the chiller stops but the PVF keeps operating at the minimum flow.

I'll appreciate any comment or link to a guide to the sequence of operation of a PVF system.

Regards,
Eloy Ruiz

 

[cdxx139]

We typically have the chiller minimum flow be kept by a bypass valve in the system. And the pumps are controlled by pressure DP. So when flow drops close to chiller minimum, bypass valves begins to open, reducing pressure in the system, and the primary pumps begin to ramp up.

Check out
Bell & Gossett Variable Primary Flow Systems Technical manual TEH-910A
They are the leaders in the industry

knowledge is power

 

[EnergyProfessional]

what cdxx139 said.

In addition I'm thinking a flow meter may be better than a dP sensor since at low flow the pressures are low to measure accurately and I assume they run it at a bit higher dP for safety. A magnetic flow meter also may more durable and accurate over the long run and you can read flow directly as opposed to deriving vie dP (in some cases the dP sensor include strainer etc., and the chiller data on dP may not be necessarily accurate as they are for design, not operation)

 

[cdxx139]

Is the flow meter replacing the dp at the evaporator to control the bypass valve? (I can see that)

Or is it replacing the dP that controls the pumps? (I cant see that)

knowledge is power

 

[LowDeltaTSolver]

ELoyRD,

The B&G Technical Manual TEH-910A discusses many issue and or concerns associated with automatic operation of a Variable Primary Flow (VPF) system.

Are you concerned that the primary pump will continue to run when the chiller shuts down and the chiller automatic control isolation valve closes?

If this is the case, the sequence of controls need to be programed to prevent this from happening.


B&G Technical Manual TEH-910A indicates that:


"Minimum Chiller Flow"

"At very low loads, the pumps have slowed and de-staged.
It's possible that only one of the pumps in parallel may be able to provide the required part load flow.
But if all the evaporators are open to the reduced below its minimum.
Therefore, a decrease in demand for chilled water must also be accompanied by a reduction in the number of operating chillers. Turning off, or “de-staging” a chiller compressor is not as simple as opening a set of contacts.
Careful analysis is required in order to determine the part-load flow rate at which to de-stage a chiller, since isolating one chiller will increase the flow to the remaining chillers.
A rapid increase in flow during this transition must be avoided by using slow closing valves on the de-staging chiller.
Chiller plant operators should be involved in order to reduce the possibility and consequences of a failure in the automatic process.


"System Complexity"

"...the VPF system depends on accurate, timely, automatic sensor inputs at several points in normal operation. The selection, installation, periodic calibration and servicing of those sensors might be complex..."

VPF system control is extremely complex.

Your doubts are well founded.

If this is a new system, you may want to consider a Variable Primary Flow-Variable Secondary Flow (VPF-VSF) system.

See ASHRAE Journal October 2014 Article " Simplified Chiller Sequencing" for an alternate approach to a VPF System.

http://uwf.edu/media/university-of-...ctober-2014-Simplified-Chiller-Sequencing.pdf

 

[EnergyProfessional]

cdxx139: correct, the meter would control the bypass valve. Pumpspeed would be controlled by the system dP sensor.

My thinking is based on a similar application for boilers in variable primary flow. The problem (or good thing) with boilers is they have very low pressure drop compared to chillers. In my case 3 psi at max flow, and maybe 0.25 psi at minimum flow. So with a standard 10 psi sensor this would be really hard to measure. Flowmeters on the other hand are still accurate within 2-4% at low flow. I realize with chiller having higher dP and higher minimum flow than boilers this may not matter much.

 

[Drazen]

the main purpose of using bypass valve was to save on pumps in primary.

it's rarely the issue these days, though, so it's hard to justify against hydraulic decouplers.

 

[EnergyProfessional]

Drazen: are you suggesting primary/secondary is acceptable again?
I'm more moving to variable primary wherever i can. the smaller primary pumps steal space, don't last as long. And you can match system and chiller/boiler flowrate much better. Especially with boilers with high turndown the minimum flow rates are so low. In addition primary/secondary loops for boilers never match the flowrates, increasing return water temp unnecessarily, forcing boilers to go out of condensing mode.

 

[Drazen]

i'm just saying that there is much less need to save on primary pumps invesments than it was years ago.

when all pumps were fixed-speed and used much more energy than today, it was feasible to resolve some hydraulic issues with bypass valves. now situation is much different and i find and apply bypass valves only in small applications where simplicity is in focus.

type of primary regime has to be agreed with manufacturer. neither chillers nor boilers have the same issues for each product, and manufacturers' requirements are different.

 

[EloyRD]

Dear @cdxx139, @LowDeltaTSolver:
Thanks for the links. I've been reviewing them through the weekend to understand a system that has some issues with the minimum load. We have a VPF systems with two 325 tonR centrifugal chillers in parallel; the starting point of each chiller is 49 tonR (15%) but the minimum load at nights is around 40 tonR.
I understand now that the VPF is capable of maintaining the minimum flow of the chiller so now I'm evaluating 3 alternatives: adding a smaller chiller to the system; installing a Hot Gas Bypass in one of the existing chillers or adding an inertial deposit. I'll let you know how that works out.

Dear @HerrKaLeun, @Drazen:
Thanks for the comments. It seems like the installer isn't familiar with VPF and we're checking the control settings of the pumps and bypass valve.

Regards,
Eloy Ruiz

 

[Timothi]

Yes... If the minimum flow is 20% and ur building load is <20% your VPF system will be running in 20% load.

Sharing knowledge is the best way to learn

 

[cry22]

when the system is down to such a low load, not only the plant is down to one chiller but the chilled water temperature should be reset upward, thus more flow.
savings are better with higher chiller water supply temperature than with lower pump speed.

See Taylor engineering web site, they have have a good five series of chiller plant design written for ASHRAE by Mr. Taylor.

 

[nabeel3]

u can use a additional chiller with low capacity.

 

[EnergyProfessional]

Drazen: always good to hear your expert opinion. My experience is more with boilers. But I try to go more VPF whenever possible. Not only it saves space and place for pumps. Also maintenance and replacement savings with fewer pumps.

Also wanted to reiterate why I think flowmeters are better than dP sensors. dP sensors are rated for 50 psi or higher to withstand the pressure. But at minimum flow (at least for fire tube boilers) dP is a fraction of a psi. At that point the sensor is highly in accurate. (numbers may be different for chillers, though)

 

[cry22]

If I may add to lowDeltaT
"A rapid increase in flow during this transition must be avoided by using slow closing valves on the de-staging chiller"

I suggest that designers read the link below. Several chiller staging soltutions are presented.

http://www.bcxa.org/ncbc/2010/docum...-2010-chilled_water_commissioning-villani.pdf

 

[marcoh]

I would suggest always use mag flow meters to measure the primary and building flow rates. Use the actual building flow demand to stage the chillers, and the primary flow meter to control the bypass valve.

Pumps speed is controlled by a dP in the field near furtherest AHU to maintain th pressure set point.

Measuring dP across a chiller is very crude, especially as flow change with variable speed primary pumps. In multi chiller installation, the by pass valve would likely on be required when 1 or 2 chillers are operational.

 

[LowDeltaTSolver]

marcoh

See 16-Jan-15 comments

If the approach indicated in the ASHRAE Journal October 2014 Article " Simplified Chiller Sequencing" is used,

No costly flow meters or differential pressure sensors are required to control staging of chillers.

The indicated approach only requires four temperature sensors.

See ASHRAE Journal October 2014 Article " Simplified Chiller Sequencing"

http://uwf.edu/media/university-of-...ctober-2014-Simplified-Chiller-Sequencing.pdf