Skip to main content

Fuel Filter Micron Ratings Explained

Not long ago I received the following note from a reader; it echoes questions I’ve been asked on this subject literally hundreds of times, and as such, bears sharing with as wide an audience as possible. The subject is both vitally important, yet frequently misunderstood.

“I have a Sabreline 36 Sedan with twin 6LYA-STP Yanmar 370hp engines.
The on-engine filters are standard Yanmar. I have Racor 900MA turbine primary filters. I currently use 2 micron elements and have a pressure drop of 2­–3 inches of mercury [a standard vacuum unit of measure] across the Racors at wide open throttle. My mechanic says that I should only be using 10- or even 30-micron filters in the Racors and that by using the 2 micron ones, I am damaging my engine fuel lift pump and perhaps the injector pump. Is he correct?
I maintain that the filters are rated for my engine's flow rate (about 66gph max) and as long as the pressure drop across them is less than 15 inches of mercury and I can reach rated rpm (3,300) at wide open throttle (W.O.T.), I am OK. Is there any way to tell if the lift pump or the injector pump is performing normally or if they have been damaged?”
Mike Long

This is a great question. The 2 vs. 10/30 micron primary filter element debate is one that has raged for several years now and I suspect the end is nowhere in sight, and it applies to all diesel engines, from 10 hp sail auxiliaries to engines like yours. There's good reason for the confusion. Many folks prefer to use a 2 micron filter because they don't want to deal with the on-engine, primary filter, and I can't say I blame them. This filter is often difficult to change without introducing air into the system; it's usually expensive and sometimes difficult to access.

Putting the Yanmar mechanic's advice aside for the moment, I'll come back to it, using a 2 micron element in the primary filter effectively reduces the area of the filter media by at least half, typically a disproportionately small quality of debris makes its way to the second filter, so it's not being spread out over the two elements. This means, if you encounter gross contamination, you have that much less filter media with which to capture it. Additionally, sequential filtration is scientifically proven to be more effective than single stage, fine micron filtration, but you don't have take my word for this, it's standard procedure for most engine manufacturers, including Yanmar, as well as filter manufacturers such as Racor. Additionally, this approach is also used by the semi-conductor and pharmaceuticals industries for obtaining exceptionally clean water, a necessary ingredient for their processes.

Now, back to your mechanic’s comments. Technically, he's right in that this is Yanmar's advice, however, I don't necessarily agree with Yanmar's reasoning as to why they recommend 10 or 30 micron primary filter elements. Indeed, chronically high vacuum on the intake side of the lift pump will cause diaphragm-style pumps to wear out more quickly, however, this would occur regardless of whether the element was 2 or 30 microns, the vacuum, and the accumulation of debris that causes it is the culprit, rather than the micron rating. Their logic, however, is, I presume, that a 2 micron element will accumulate debris, i.e. clog, more quickly, which will lead to higher vacuum (anything over about 7 inches of mercury is cause for concern by the way, and 15 inches of mercury is critically high) and thus pump damage.

Contrary to popular belief, a clean 2 micron element presents no greater resistance or vacuum than a clean 10 micron element (this is based on my personal experience and is confirmed by the folks at Parker/Racor). Just to be clear, if the lift pump is experiencing high fuel vacuum it's possible that its output will be hindered, which could lead to fuel starvation at the injection pump. And, because the fuel is the lubrication within the injection pump, this in turn could lead to cavitation and injection pump damage.

Thus, in the end, I agree with your mechanic and Yanmar, just not for the same reasons. If the vacuum is low, regardless of the element's micron rating, then the pump will not be damaged. I'm not suggesting you violate their instructions; that may be grounds for denial of warranty coverage, but that really doesn't matter because I'm a firm believer in sequential filtration, which would call for the use of a 10 or 30 micron primary filter element, which is in keeping with the manufacturer's guidelines. This, in combination with the on-engine filter supplied by the manufacturer (or a high quality equivalent) will provide you with just that, excellent sequential filtration.

If you don't want to deal with service of the on-engine secondary filter, you could always add another in line Racor 900, equipped with a vacuum gauge, after the existing primary Racor and before the on-engine lift pump and filter. This new filter would become your secondary and the on-engine filter would become the tertiary filter (it could be changed seasonally, at your leisure). Technically, this would violate Yanmar guidelines, but the fuel would be very clean (you'd have two opportunities to capture dirt and water), and you wouldn't have to deal with the on-engine filter element. If the vacuum experienced by the lift pump remained low, within the spec's for your engine, then it's unlikely that any damage would ever occur. Remember, any severely clogged filter, regardless of micron rating, will cause a high vacuum scenario and possible pump damage.

It doesn't sound as if you've done any damage to the pumps and the pressure drop across your filter sounds normal, although I wouldn’t wait until it reaches 15 inches of mercury before changing the elements. Finally, I’m a firm believer in the use of high quality, drag needle-equipped vacuum gauges (the kind with a needle that can be reset) and water sensors/alarms on all primary filters.

-The Newsletter entry you just read has been excerpted from Steve D’Antonio’s Marine Systems Excellence Blog. To access more entries like this, simply visit Steve’s blog at Sign up is quick and easy, it only requires an e-mail address.