Two numbers do most of the work of specifying the sand or anthracite in a water filter: how big the grains are and how much their sizes vary. Those are the effective size and the uniformity coefficient. Together they capture the gradation of the media, and gradation is what decides how the filter clogs, how deep floc penetrates before it is caught, and how well the bed re-sorts after backwash. This guide defines effective size and uniformity coefficient, explains how they influence head-loss buildup and floc penetration depth, and why operators track media loss and gradation to plan media replacement.
Effective Size / UC in one line: The effective size of granular filter media is the sieve size through which ten percent of the media by weight will pass, written as d10, and it characterises the fine end of the media that does much of the filtering. The uniformity coefficient is the ratio of the size through which sixty percent passes to the effective size, d60 divided by d10, and it describes how spread out the grain sizes are; a value near one means very uniform grains, while a larger value means a wider spread. Together they specify the media's gradation, which controls head-loss buildup, how deeply floc penetrates the bed, and how the media re-stratifies after backwash.
Both figures come from a sieve analysis, the standard way of measuring how a granular material is distributed by size. The media is passed through a stack of sieves and the fraction passing each is plotted to build a gradation curve. The effective size, d10, is read from that curve as the mesh size that lets the smallest ten percent of the material pass; it is called effective because those finer grains have an outsized effect on how the bed filters and resists flow, so the fine end is a good single-number descriptor of behaviour.
The uniformity coefficient is a second reading from the same curve, the size at which sixty percent passes divided by the effective size at which ten percent passes. A uniformity coefficient close to one describes media whose grains are nearly all the same size, while a larger coefficient describes a wider range from fine to coarse in the same batch. Filter media is usually specified to have a fairly low uniformity coefficient, meaning reasonably uniform grains, because a very wide spread brings problems in how the bed packs and re-sorts.
Specifying media is therefore a matter of naming a target effective size and a maximum uniformity coefficient. Those two values tell a supplier what to deliver and tell an engineer what to expect from the bed, and they appear on the media specification for a filter alongside the depth and type. Because they are simple, measurable, and tightly linked to performance, effective size and uniformity coefficient are the language in which filter media is bought, checked, and compared.
Effective size drives a direct trade-off between clean-bed resistance and how the filter clogs. Finer media, a smaller effective size, has smaller passages between grains, so it catches floc closer to the surface and produces very clear water, but those small passages fill and resist flow sooner, so head loss climbs faster and runs are shorter. Coarser media, a larger effective size, lets floc penetrate deeper into the bed before being caught, which uses more of the depth and builds head loss more slowly for longer runs, at the cost of catching the finest particles less aggressively near the top.
This is exactly why dual-media and graded beds are designed the way they are. Putting a coarser layer on top lets floc penetrate into that upper layer rather than blinding the surface, spreading the solids load through the depth, while a finer layer below polishes the water. The effective sizes of the two layers are chosen so the coarse layer captures the bulk of the load and the fine layer finishes the job, giving both long runs and low turbidity. The whole art of grading a filter bed is really the art of choosing effective sizes that place the solids where the bed can hold the most of them.
The uniformity coefficient matters because it governs how cleanly the bed re-sorts after backwash and how it packs. When the media is fairly uniform, the bed fluidises and re-stratifies predictably after each wash, keeping its intended coarse-to-fine profile. When the spread is too wide, the finest grains can migrate and blind the surface, the bed stratifies less cleanly, and the smallest grains are also the ones most easily washed out during backwash. A controlled uniformity coefficient keeps the bed behaving consistently wash after wash, which is why it is specified as tightly as the effective size.
Filter media does not last forever. Every backwash risks carrying off some of the finest grains, grains grind and round over years of fluidisation, and beds slowly lose depth and drift in gradation. Over time this changes how the filter behaves: a bed that has lost media runs shorter and may allow more penetration, while a bed whose fines have washed away shifts its effective size and uniformity coefficient away from the original specification. Left unaddressed, an aged bed quietly loses the performance the media specification was meant to guarantee.
Operators therefore keep an eye on the physical bed, not just the water leaving it. Measuring the depth of media to detect loss, and periodically taking core samples for a sieve analysis to check that the effective size and uniformity coefficient are still near specification, are the direct ways to tell how the media has aged. Comparing today's gradation to the original tells whether the fines have been lost, whether the media needs topping up, or whether a full media replacement is due. Scheduling that replacement from evidence, rather than guessing, avoids both running a degraded bed too long and replacing good media too soon.
Trends in the water and the cycle give the earliest hints that the media is changing, and this is where SCADA connects to the physical bed. In a platform such as Merobix, filter run length, head-loss behaviour, and filtered-water turbidity are trended over months, and a bed that is steadily giving shorter runs, building head loss differently, or losing its turbidity margin points to media that has changed. Those trends flag which filters to inspect and sample, so operators track the physical gradation and media loss where the SCADA trends suggest, and plan media replacement before performance falls out of specification.
Effective size, written as d10, is the sieve size through which ten percent of the media by weight will pass, and it characterises the fine end of the grains that do much of the filtering. A smaller effective size means finer media that catches floc near the surface, gives very clear water, but clogs and builds head loss faster. A larger effective size lets floc penetrate deeper, using more of the bed depth and giving longer runs, at the cost of catching the finest particles less aggressively at the top.
The uniformity coefficient is the ratio of the size through which sixty percent of the media passes to the effective size, d60 divided by d10, and it describes how spread out the grain sizes are. A value near one means very uniform grains; a larger value means a wider range. It matters because uniform media re-sorts cleanly after each backwash and keeps its intended coarse-to-fine layering, while a wide spread causes the fines to migrate, blind the surface, and wash out more easily, so it is specified as tightly as the effective size.
You tell from the physical bed and from long-term trends. Measuring media depth reveals loss, and periodic core samples put through a sieve analysis show whether the effective size and uniformity coefficient have drifted from the original specification. Alongside this, SCADA trends of shortening filter runs, changing head-loss behaviour, and a shrinking turbidity margin flag which filters have aged. Replacement is scheduled from that evidence so a degraded bed is not run too long nor good media replaced too soon.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.