The fluid-loss test: filtrate, cake & what it costs you
Fluid loss reads like a mud-chemistry number, but the cake it builds is where solids control shows up in the hole. A thin, slick cake keeps the pipe free and the formation intact; a thick, sticky one is how a string gets stuck and a zone gets damaged — and the difference is usually how many drilled solids you left in the mud.
What the test measures
The filtration test measures two things: the filtrate — how much liquid the mud gives up through a permeable filter under pressure — and the filter cake it deposits while doing so. A low, controlled filtrate with a thin, tough cake is what you want. The mud is pressed against filter paper; the liquid that passes is collected and read, and the cake left on the paper is measured.
API (low-temperature, low-pressure)
The standard API filter-press test runs at 100 psi for 30 minutes at ambient temperature, through a filter area of about 7.1 in². You report the API filtrate in mL at 30 minutes and the cake thickness in 32nds of an inch (or mm). It’s the routine bench check that tracks day-to-day filtration control.
HPHT (high-temperature, high-pressure)
Surface conditions don’t represent the bottom of a hot, deep hole, so the HPHT cell runs at elevated temperature and a 500 psi differential. Because the HPHT cell uses about half the filter area (~3.5 in²), the measured filtrate is doubled to report it on the same 7.1 in² basis as the API test. HPHT loss is almost always higher than the API number — that gap is the part of the picture the bench test hides.
Why it matters to solids control
Filtration additives get the credit, but the cake is built by particles — and a big share of those particles are drilled solids. The right distribution of fine solids bridges the pore throats and lays down a thin, low-permeability cake. Too many fine, low-gravity drilled solids and the cake turns thick, soft and sticky: higher filtrate, more differential-sticking risk, and more formation damage.
That’s the link crews miss: when fluid loss creeps up and the cake thickens, the first question isn’t “add more polymer” — it’s “what is my drilled-solids load doing?” Clean the solids out and the cake often fixes itself.
Reading the numbers
Read filtrate as a trend against your mud’s target, not a single pass/fail. A climbing API filtrate with a thickening cake points at solids or a depleted additive system; a normal API number with a high HPHT number points at temperature-sensitive filtration that won’t show on the bench. Always log cake thickness next to filtrate — a low filtrate with a thick cake is still a problem.
Limitations to keep in mind
The bench API test is ambient — it can look fine while downhole loss runs high, which is exactly why HPHT exists for hot wells. Cake measurement is a feel-and-eye read, so consistency of operator matters. And the test characterises the fluid; it doesn’t tell you which solids are driving the cake — pair it with the retort, PSD and methylene blue tests to see the solids behind the number.
Quick reference
| Standard | API RP 13B-1 / 13B-2 |
| API pressure / time | 100 psi · 30 min |
| API filter area | ~7.1 in² · ambient |
| HPHT differential | 500 psi · elevated temp |
| HPHT area / reporting | ~3.5 in² → filtrate × 2 |
| Reported | Filtrate (mL) + cake (1/32″) |
A thin, slick filter cake is a solids-control achievement as much as a chemistry one. Keep the drilled-solids load down and the right fine fraction in the system, and the cake stays thin, the pipe stays free, and the formation stays intact. Let the solids run and no amount of filtration additive fully buys it back. Measured, not guessed.
Put it to work
If fluid loss and cake are creeping while drilled solids climb, a remote evaluation can show whether it’s the solids load or the fluid system that needs the attention.
Request a remote evaluation More field articlesRelated reading
Grounded in API RP 13B-1 / 13B-2 and field solids-control practice. Test conditions and limits vary with the system and rig — treat values as engineering guidance and follow your standard’s current edition.
