FAT vs SAT: what to test and when

Who this is for

Manufacturers and project teams importing a production line or major machine — particularly from China — who want to understand exactly what FAT and SAT are for, what to test at each, and how to structure them so they actually protect the project. Also useful as a brief for an outsourced FAT representative.

The two tests, in one sentence each

FAT proves the machine meets the agreed specification at the supplier's factory, before it ships. SAT proves the installed machine meets the agreed specification on your site, in your real operating conditions, before final payment and warranty start.

They are not duplicates. FAT catches build and design faults while you still have leverage and the supplier's full workshop around the machine. SAT catches installation, integration, utility, and real-feedstock issues that only appear on your floor.

Why FAT matters more than buyers expect

The moment the equipment leaves the supplier's factory, your leverage collapses. At FAT you have: the machine fully assembled, the supplier's engineers and workshop on hand, money still owed, and the legal position that the machine has not been accepted. After shipment you have: a machine in pieces in a container or on your floor, the supplier 12 000 km away, most of the money paid, and the burden of proving a fault is theirs and not your installation.

This is why "we'll fix it at SAT in South Africa" is almost always a bad trade. Fixing in China at FAT uses the supplier's resources; fixing in South Africa uses yours, on your time, with the supplier's cooperation now optional. Our FAT rescue case study is a worked example of what happens when FAT discipline is missing.

What to test at FAT

FAT runs the machine against the agreed protocol at the factory. Good FAT covers:

• Throughput / rate — the machine runs at or above its specified rate, sustained, not just in a brief burst.
• Product-realistic testing — run with feedstock as close to your real raw material as possible. A filler tested with water when you make juice has not been meaningfully tested.
• Controls behaviour — start/stop sequences, recipe handling, changeovers, alarms, interlocks.
• Safety systems — guards, E-stops, light curtains, pressure relief — actually triggered and verified, not just present.
• Quality output — measurable acceptance criteria on the product the machine makes (fill accuracy, dimensional tolerance, seal integrity, etc.).
• Component verification — the brands and specs match the bill of materials (catching silent substitution).
• Documentation completeness — schematics, BOM, spare-parts list, manuals present and correct.
• Scope completeness — every item on the contract scope is physically present.

Failures are documented, rework is agreed in writing, and FAT is re-run on the failed items. A signed FAT certificate is issued before the machine leaves.

In-person vs video FAT

In-person FAT — your engineer (or a competent representative) physically present — is the gold standard and is strongly recommended above roughly USD 500 000 contract value or for any first-time supplier. Video FAT is the absolute floor: a live, interactive video walk-through where you direct the tests, not a pre-recorded highlights reel. A pre-recorded video is marketing, not a FAT.

Costs, 2026: an in-person engineer trip from South Africa runs roughly USD 5 000–15 000 (flights, hotel, time); an outsourced FAT representative in China runs roughly USD 1 500–4 000. Against the cost of discovering a fault after shipment, both are cheap.

What to test at SAT

SAT runs the installed line on your floor, against the same (or a site-adapted) protocol, in real conditions:

• Real raw material — your actual feedstock, not a clean test sample.
• Real utilities — your power, water, steam, compressed air quality, with their real fluctuations.
• Your operators — running the line as they will in production, not the supplier's experts.
• Sustained rate — the line holds its specified rate over a meaningful production run, not a brief demo.
• Integration — interfaces with upstream/downstream equipment, conveyors, utilities, and existing plant systems.
• Quality at rate — product meets spec while running at production speed, not just slowly.
• Cold → hot → production sequence — dry run, then with product, then sustained production.

SAT is where the import meets reality: the spec that passed in a Chinese factory now has to pass on your floor, with your maize, your borehole water, and your two newly trained operators.

FAT vs SAT at a glance

Tie payment to the tests

The acceptance tests only protect you if money is tied to them. A typical structure for an imported line:

• ~30% deposit on order.
• ~50% on successful FAT — not on a calendar date, on a signed FAT pass.
• ~20% on successful SAT — final payment and warranty start on a signed SAT pass.

Payment milestones tied to physical, verified deliverables — not invoice dates — are what give the tests teeth. A milestone that pays on "machine ready to ship" instead of "FAT passed" hands your leverage to the supplier. This connects directly to the import sequence in How to import a production line from China to South Africa.

Write the protocol before the build

The single most important discipline: the FAT and SAT protocols — test sequence, measurement methodology, acceptance criteria, pass/fail logic, hold points — are agreed in writing before the build starts. If you write acceptance criteria after the machine is built, the supplier has already built to their own interpretation, and you cannot meaningfully fail it.

Deferring items from FAT to SAT should be a deliberate, written decision (e.g. a sensor tolerance that genuinely can only be verified on site), not an accident. "We'll check that at SAT" should never be a way to avoid a hard FAT result.

Who should attend

• FAT: your engineer or a competent representative who knows the machine class and can read the workmanship — ideally the same person who will own the SAT. For a complex line, a representative with FAT-specific experience is worth the cost.
• SAT: your engineering and production leads, your operators, and the supplier's commissioning team. The people who will run and maintain the line should be present, not just sign off.

Common mistakes

1. Skipping FAT to save a trip. The most expensive false economy in the import. A video FAT is the floor; in-person is recommended for high-value or first-time contracts.

2. No written protocol. A test with no pre-agreed criteria cannot be failed.

3. Testing with the wrong feedstock. A machine tested on water when you make a viscous product has not been tested for your application.

4. Booking the commissioning team before the site is ready. SAT cannot happen on an unready site, and idle commissioning engineers cost money daily.

5. Paying the FAT milestone on a date, not a pass. This quietly removes the leverage FAT exists to provide.

What CISH does in this part of the process

We write bilingual FAT and SAT protocols, represent buyers at FAT in China, and run SAT on site — under China Procurement & Sourcing or as part of a Turnkey Production Lines delivery where CISH holds acceptance authority. See the supplier audit checklist for the step before FAT, and the FAT rescue case study for what recovery looks like when the discipline was missing.

Frequently asked questions