TR19 & standards
TR19 Grease turns 'clean enough' into a measured number. This is the test, the thresholds, and where the readings are taken.
The short answer
TR19 Grease sets the bar as a thickness of grease inside the ductwork. The mean must stay under 200 microns between cleans; any spot at 500 microns or more needs immediate cleaning; and after a clean the surface should read down under about 50 microns. The reading is taken with a wet film or deposit thickness test, at set points through the system.
The two tests
Two measurement methods sit behind the number. The Wet Film Thickness Test (WFTT) - a method pioneered in the UK duct-cleaning industry and written into the specification - gauges the grease film on the duct surface. The Deposit Thickness Test (DTT) measures the depth of accumulated deposit. Both give a reading in microns that can be compared against the TR19 Grease thresholds and recorded in the post-clean report.
Readings are not taken at one convenient spot. The specification points to a spread of locations so the picture is representative: behind the filters in the canopy plenum, roughly one metre and three metres along the duct from the canopy, midway between canopy and fan, just upstream of the fan, and on the discharge duct after it. Grease does not settle evenly, so a single reading proves very little.
Reading the thresholds
The headline limit is a mean average under 200 microns - about half the thickness of a business card - across the system between scheduled cleans. Cross that and the extract is treated as a fire risk. A localised reading of 500 microns or more is a red flag that needs spot-cleaning without waiting for the next scheduled visit.
After a compliant clean, the same points should read far lower - down near or under 50 microns, roughly half the width of a human hair - which is what a good post-clean report demonstrates with paired before-and-after figures. The gap between those two sets of numbers is the evidence the clean actually worked.
The thresholds also shape the schedule, not just the pass mark. Where there is no history to work from, the specification uses the readings and the days between them to set an initial cleaning frequency - the aim being to catch the system well before it drifts toward the 200-micron mean rather than after. Because grease loads unevenly, a section near the cookline or the fan can reach the spot-clean trigger long before the rest of the run, which is exactly why the readings are spread along the ductwork and reviewed each visit rather than assumed to hold from last time.
The limit
A reading only proves the point it was taken at. That is why the spread of locations matters, and why an inaccessible section - one no probe could reach - leaves a genuine blind spot rather than a pass. A clean that measures beautifully at the hatch but never reached the riser is not a compliant clean; it is a well-documented partial one. Access, covered in the sibling guides, is what lets the numbers speak for the whole system.
Questions
A mean average under 200 microns between cleans. A localised area at 500 microns or more needs immediate spot-cleaning, and after a clean the surface should measure down near or under 50 microns.
The Wet Film Thickness Test gauges the grease film on the surface; the Deposit Thickness Test measures the depth of accumulated deposit. Both give a micron reading used to check compliance.
At representative points: behind the canopy filters, about 1 m and 3 m along the duct, midway to the fan, just upstream of the fan, and on the discharge duct - because grease does not settle evenly.
About half the thickness of a business card. It is small - which is the point. By the time grease is visibly caked, the system is well past the limit.
No. One reading proves one spot. Compliance is judged across the whole accessible system, which is why access to every section matters.
Phoenix Duct Clean · by the numbers
We take grease-thickness readings through your whole system and show you exactly where you sit against the 200-micron limit - before and after we clean.