An industrial evaporator factory acceptance test (FAT) is the structured verification step where you prove a wastewater evaporator performs to contract before it ships. It happens at the manufacturer’s shop, with your engineers present, and it is your last clean shot at catching design or fabrication problems while the unit is still on their floor. I have sat through dozens of these tests, and the projects that go smoothly are almost always the ones where the buyer showed up with a written protocol instead of a handshake. This guide walks through what to check, how to run the test, and where I have seen good equipment fail a bad FAT.
Why the Factory Acceptance Test Matters More Than You Think
By the time an evaporator reaches your site, the cost of fixing a mistake multiplies. A pump that runs hot in the shop is a warranty claim. The same pump running hot six thousand miles away is a crane rental, a shutdown, and a month of finger-pointing. The FAT compresses that risk into a controlled window.
I treat the FAT as a design review that finally has real water in it. Drawings tell you what someone intended. A running machine tells you what actually happens when scaling starts, when the compressor loads up, and when the control loop tries to hold level on a foaming liquid.
There is also a commercial reason. Most contracts tie final payment and shipping release to a signed test report. If the protocol is vague, you lose leverage. If it is specific, you get a clear pass or fail and a documented path to correction.
What to Verify Before the Unit Even Starts
Documentation and Material Traceability
Start with paper. It is boring and it catches real problems. Confirm that wetted parts match the specified alloy. Heat exchangers in evaporator service often run in duplex stainless, titanium, or higher grades depending on chloride levels and pH. A mill certificate that does not match the drawing is a red flag you want to find now.
Check the pressure vessel documentation for the shells and any separators. Confirm relief valve set points and certifications. Verify that the electrical panel matches the hazardous area classification your site requires, not the one that was convenient to build.
Mechanical Checks
- Rotation direction on every pump and the compressor or blower
- Coupling alignment and guard installation
- Torque on flange bolts, especially on high-temperature and vacuum service
- Instrument calibration certificates for flow, level, temperature, pressure, and conductivity
- Insulation continuity and heat tracing where freeze protection is required
I once watched a commissioning team discover that a level transmitter had been calibrated for water density when the process fluid was nearly twice as dense. The FAT caught it in an afternoon. On site it would have flooded a separator.
Building the Test Protocol
A useful protocol has four parts: a scope statement, a list of measured parameters with acceptance criteria, a data recording method, and a punch list process. Keep it short enough that people actually use it.
| Test Category | What Gets Measured | Typical Acceptance Basis |
|---|---|---|
| Mechanical run | Vibration, bearing temperature, noise | Manufacturer limits, project-dependent |
| Hydraulic performance | Feed rate, circulation rate, pressure drop | Design values with stated tolerance |
| Thermal performance | Evaporation rate, steam or power consumption | Indicative range, corrected for test conditions |
| Control logic | Interlocks, alarms, shutdowns, restart | Functional pass or fail |
| Product quality | Concentrate density, distillate conductivity | Project specification |
| Leak and vacuum | Pressure hold, vacuum decay rate | Defined hold time and allowable drop |
Notice that I did not put hard numbers in the thermal row. They do not belong there in a generic guide. Your evaporator’s expected performance depends on the fluid, the concentration ratio, the available heat, and the ambient conditions in the shop. What matters is that the protocol states how the manufacturer will correct measured results to design conditions, and that you agree on that method before the test starts.
Running the Actual Test
Water Test Versus Real Fluid Test
Most shop tests run on water, sometimes with salt added to simulate conductivity. A water test proves mechanical integrity, control function, and basic heat balance. It does not prove you can handle scaling, foaming, or fouling. Be honest with yourself about what a water test can and cannot tell you.
If your wastewater has high hardness, organics, or oil, ask whether the manufacturer can run a small batch of representative fluid or a synthetic surrogate. Even a short run with the real stuff surfaces problems that water never will. For oily streams, the behavior of the unit is different enough that it deserves its own review — see our notes on evaporation with oily wastewater.
Instrumentation and Data Logging
Insist on independent measurement where it matters. If the manufacturer’s own flow meter is the only source for the performance number, you are trusting the same party that benefits from a pass. Bring your own clamp-on ultrasonic flow meter and a calibrated temperature probe. It takes an hour to set up and it changes the conversation.
Log data continuously, not by hand every fifteen minutes. Trends show you whether the unit is stable or slowly drifting. A climbing compressor discharge temperature over a two-hour run tells a story that a single snapshot never will.
Performance Testing by Evaporator Type
Different configurations put different demands on the test. What you watch depends on what you bought.
Mechanical Vapor Compression and MVR Units
For an MVC or MVR system, the compressor is the heart of the test. Watch suction and discharge pressure, discharge temperature, current draw, and vibration across the load range. The unit should hold a stable evaporating temperature without hunting. If the compressor surges or the control loop oscillates, that is a control tuning problem you want solved in the shop.
Heat exchanger performance deserves its own attention. Fouling that develops during the test is a preview of your cleaning frequency. Our page on MVC evaporator heat exchangers covers the design details that drive this behavior. For a deeper look at the technology itself, the MVR technology overview is a useful reference.
Multi-Effect Evaporators
With multiple effects, the test focuses on inter-effect balance. Each stage should hit its design temperature and concentration. Steam economy — the ratio of water evaporated to steam consumed — is the headline number, but it only means something when the feed conditions match the design basis. If the shop cannot supply design steam pressure, the manufacturer should document the correction.
Single, double, and triple effect designs each carry different tradeoffs between capital cost and energy use. A single-effect unit is simpler to test and control. A triple-effect unit saves steam but adds stages, instrumentation, and failure points to verify. More stages mean more to check, and the FAT is where you find out whether the control scheme actually coordinates them.
Hybrid and ZLD Trains
If your evaporator is the back end of a membrane and zero liquid discharge system, the FAT often covers only the evaporator skid. Make sure the interface conditions — feed quality from the upstream RO or DTRO stage, flow rate, and turndown — are written into the protocol. A skid that passes on clean water but receives a brine twice as concentrated on site has not really been tested.
Safety and Interlock Verification
Do not let the safety checks get squeezed into the last hour. Test every interlock deliberately by forcing the condition. High level, low level, high temperature, high pressure, low flow, and loss of utilities should each trigger the correct response. Record the actual action, not just that the alarm lit up.
On one project I insisted on tripping the main power mid-run to verify the safe shutdown sequence. The unit dumped hot concentrate back into a cold feed tank and cracked a fitting. Better in the shop than in a plant where someone could have been standing nearby.
Confirm that emergency stops work from every station. Verify that the control system fails safe on instrument signal loss. If the unit runs unattended, confirm the remote alarm and shutdown paths.
Common Problems I See During FATs
- Foaming. Water tests hide it. Real fluids with surfactants or high dissolved solids foam badly and carry over into the distillate.
- Scaling on the heat transfer surface. Even a short run can show a falling evaporation rate if the fluid is hard.
- Control loop oscillation. Level and temperature loops that were never tuned under load.
- Pump cavitation. Often a suction-side design issue that only shows at temperature.
- Instrument drift. Uncalibrated or wrong-range transmitters.
- Leaks at flanges and seals that only appear when the unit reaches operating temperature.
None of these are unusual. The point of the test is to find them while the unit is still in the shop and the fix is cheap.
Cost and Schedule Realities
A proper FAT costs money. You pay for travel, for the manufacturer’s test time, for the fluids, and for the labor. I have seen buyers try to skip it to save a few thousand dollars and then spend ten times that on site. The math rarely works in favor of skipping.
Schedule matters too. A test that gets rushed because the shipping date is fixed will miss things. Build contingency into the schedule so a failed test does not force a bad decision. If the unit fails, you want the option to fix it properly rather than ship it with a promise.
For context on how the capital and operating numbers move across configurations, the discussion of MVR evaporator cost drivers is worth reading before you finalize the acceptance criteria.
What a Good Punch List Looks Like
End the test with a written punch list. Each item should have a description, a responsible party, a target date, and a definition of done. Separate items that block shipment from items that can be resolved during commissioning.
- Critical: prevents safe or functional operation — must be fixed before shipment
- Major: affects performance or reliability — fix before shipment where practical
- Minor: cosmetic or documentation — can close out during commissioning
Get signatures. A punch list that nobody signed is a suggestion, not a commitment.
A Practical Checklist You Can Reuse
| Phase | Key Actions |
|---|---|
| Before the test | Approve the protocol, confirm fluid, agree on correction method, arrange independent instruments |
| Documentation | Verify materials, certifications, calibration, electrical classification |
| Cold checks | Rotation, alignment, torque, insulation, instrument loop checks |
| Mechanical run | Vibration, bearing temperature, noise, leak check |
| Performance run | Evaporation rate, energy use, product quality, stability over time |
| Safety and controls | Interlocks, alarms, E-stops, fail-safe behavior, restart sequence |
| Closeout | Punch list, signatures, as-built documentation, shipping release |
If you are planning a test and want a second set of eyes on the protocol before you commit, our team reviews these regularly and can flag the gaps that tend to cause trouble later.
Have Your FAT Protocol Reviewed by Our Engineers →
Frequently Asked Questions
How long does a factory acceptance test usually take?
For a skid-mounted evaporator, plan on two to four days on site at the manufacturer. A simple single-effect unit with a water test can finish in a day. A multi-effect or MVC system with full performance and interlock testing typically needs three days or more, plus travel. Add buffer time if the unit has never been run before.
Can I accept a unit based on a water test alone?
Only if your feed is clean and close to water in behavior. For most industrial wastewater, a water test proves mechanical and control integrity but says nothing about scaling, foaming, or fouling. If you cannot test with representative fluid, document the limitation and plan for a performance test during commissioning.
What should I do if the unit fails the performance test?
Do not sign a shipping release. Document the failure, agree on the root cause, and set a corrective action with a re-test date. If the manufacturer wants to ship and fix on site, get that in writing with clear responsibility for costs, schedule, and performance guarantees. In my experience, fixing in the shop is almost always cheaper.
Who pays for the FAT?
It depends on the contract. Usually the buyer covers travel and their own labor, while the manufacturer covers shop time and utilities. Larger projects sometimes bundle the test into the purchase price. Whatever the arrangement, put it in writing before the test is scheduled so nobody is negotiating in the middle of a run.
How do I verify energy consumption during the test?
Use a power analyzer on the main feed, not the manufacturer’s panel readout. Log current, voltage, and power factor continuously across the load range. Correct the result to design conditions using the agreed method, and record the correction in the test report. Energy is often the largest lifecycle cost, so it deserves independent measurement.
What if the shop cannot supply the design steam or power conditions?
This is common. The manufacturer should provide a documented correction method so measured performance can be translated to design conditions. If no credible correction exists, treat the test as a partial verification and plan a full performance test at your site. Do not accept a verbal assurance that the unit “will perform better on site.”
Should I bring my own instruments?
Yes, at least for flow, temperature, and power. Independent measurement removes doubt and speeds up any disagreement. A clamp-on ultrasonic flow meter, a calibrated temperature probe, and a portable power analyzer cover most of what you need and fit in a carry-on case.
The Bottom Line
A factory acceptance test is not a formality. It is the moment when drawings, promises, and real equipment meet, and it is the cheapest place to find out whether they agree. Bring a written protocol, measure independently, test the interlocks on purpose, and leave with a signed punch list. Do that, and the unit that arrives at your site will be one you already understand.