How to Estimate Chiller GPM Using Pressure Drop

A field guide for HVAC technicians

The Problem

When you are logging or troubleshooting a chiller, flow rate is one of the first things that has to be verified. In most cases, if you do not know the flow, you cannot judge chiller performance with confidence.

It is the backbone for understanding load, heat transfer, and whether the machine is performing the way it should.

And this is not just a yes-or-no question. The real question is whether you have enough flow.

So how do you know if the chiller is actually moving enough water?

One of the most practical field methods is to use pressure drop across the heat exchanger. It uses a well-established pressure-drop relationship to estimate flow from the design point.

The image below shows the basic idea: measure pressure at the entering and leaving water connections, calculate the pressure drop, then use that pressure drop against the design pressure drop to estimate flow.

Diagram showing how to estimate chiller water flow using entering and leaving pressure on a two-pass chiller barrel
Pressure drop is measured across the heat exchanger. On a common two-pass barrel, the entering and leaving water connections are usually on the same end of the machine.

The Formula

GPM2 = GPM1 × √(PD2 ÷ PD1)

Where:

  • GPM1 = the design flow rate from the chiller's selection sheet, in gallons per minute
  • GPM2 = the estimated actual flow rate, in gallons per minute
  • PD1 = the design pressure drop across the heat exchanger, in feet of water
  • PD2 = the pressure drop you measure in the field, converted to feet of water
  • Quick conversion: Field gauges usually read in PSI, while chiller design pressure drop is usually listed in feet of water. To convert PSI to feet of water, multiply by 2.31. To convert feet of water to PSI, divide by 2.31.

Where the Numbers Come From

Design GPM and design pressure drop come from the chiller's original selection sheet or submittal data. If you don't have the submittal, check the O&M manual, the nameplate, or call the manufacturer with the model and serial number. These are fixed values. They don't change over the life of the machine.

There are estimates you can use if you do not have design data to give you a rough idea to work from. That will be a different post.

Actual pressure drop is what you measure in the field. You need two pressure gauges, one on the entering water side and one on the leaving water side of the heat exchanger, whether evaporator or condenser. Subtract leaving pressure from entering pressure to get the pressure drop in PSI, then multiply by 2.31 to convert to feet of water.

Example: If your entering water gauge reads 42.3 PSI and your leaving water gauge reads 30.1 PSI:

  • Pressure drop = 42.3 - 30.1 = 12.2 PSI
  • Convert to feet of water: 12.2 × 2.31 = 28.2 ft H2O

Why This Matters

Flow is one of the biggest variables affecting chiller interpretation. If flow is wrong, everything downstream gets harder to trust.

  • Tonnage calculation depends on flow. Tonnage = GPM × ΔT / 24. If your GPM estimate is off, your tonnage is off, and your kW/ton number is not defensible.
  • Approach temperatures shift when flow changes. Low flow makes the evaporator approach look worse than the tubes actually are. High flow makes it look better.
  • Chiller protection matters. Most centrifugal chillers will trip on low evaporator pressure or high condenser pressure if flow drops far enough. Understanding where flow is before it becomes a trip can help prevent a callback.
  • Flow can also help explain fault patterns. If a flow proving device is failed or misreporting, the machine may behave like it has adequate flow when it does not, which can push you toward low-pressure or unstable operating conditions.

Common Mistakes

  • RULE #1. Do not blindly trust the gauges mounted on the piping. If flow matters, use your own known gauges or trusted digital instruments. Keep at least two and compare them against each other regularly so you know what you can trust.
  • Using the wrong design values. Make sure your design pressure drop is in feet of water, not PSI. Selection sheets may list both. Use the ft H2O value, or convert PSI to ft H2O by multiplying by 2.31.
  • Using the wrong pressure locations. Take the reading across the heat exchanger path, as close to the chiller as practical. The farther the reading location is from the chiller head, the more external piping loss can get mixed into the number.
  • Ignoring gauge accuracy. A $20 gauge with ±2% accuracy on a 40 PSI reading gives you ±0.8 PSI of possible error. That propagates through the square root and can swing your GPM estimate. Use calibrated gauges or digital sensors when accuracy matters.
  • Comparing to the wrong baseline. The design pressure drop assumes clean tubes. If the chiller has years of fouling, the actual pressure drop at design flow can be higher than the selection sheet says. This means your GPM estimate may read high because the formula thinks you have more flow than you do. Factor in tube condition.

When to Use This

Use it whenever chiller performance matters and direct flow measurement is not available. Once you have the design values, it takes two pressure readings and quick math, or you can use our free GPM calculator.

If you're documenting chiller performance without knowing flow, your tonnage and kW/ton numbers are not defensible.

Get the flow right first. Everything else follows.

Disclaimer

This method has limitations and should be treated as a field estimation tool, not a guarantee of exact flow. Confidence depends heavily on where the pressure ports are located, the quality of the pressure readings, and how much piping and how many fittings sit between the ports and the chiller.

In general, the farther the pressure ports are from the chiller, the lower the confidence in the result, because more external piping losses can be included in the reading.

This method becomes most useful when you are logging the same chiller repeatedly or checking it again on return trips. Under a consistent process, it gives you a practical way to track change over time, even when direct flow measurement is not available. We use it primarily to measure movement under repeatable conditions, not to guarantee absolute accuracy in all situations.