I've been toying with the idea of replacing the fuel filtration system in my 1985 45' Hatteras Sportfish for quite some time. The "old technology" consisted of a pair of Racor 1000 MA filter units, in parallel with isolation and selector valves, thereby permitting the transfer of fuel flow from one to the other while running if necessary. The on-engine "last chance" filter is the standard paper-element cannister on the pressure side of the fuel pump.
This system has worked quite well. However, it did have its warts. The RACORs can be messy to change, as you must extract an element from a fuel-filled cannister. Also, any large particulate matter that was extracted remains in the bowl and almost never drains out even when the "water" drains are opened. Finally, and perhaps most importantly, maintenance is a real pain in the butt, since you have to tear them down evey couple of years to clean them and, of course, that requires draining a considerable amount of fuel.
In some boats this is ok, but in a reasonably-tight engine room its a recipe for a mess. I had almost no clearance under the drains, which complicated things significantly and required the use of a "sucker" pump to get the fuel out for maintenance. In addition they stuck out into the centerline, which made them a "catch risk" when stepping down into the enginer oom.
Finally, filtration technology has simply improved. There are now filter systems designed for common-rail injection engines that cannot tolerate any sort of contamination. These are clearly superior, but it appeared to be a real hassle to make up a retrofit kit from scratch for them.
Enter Tony; he has come up with a very nice dual filter system consisting of two filter heads and both a fuel strainer (nominally 30 micron) and a 10 micron "Stratopore" secondary. The latter is rated for use on common-rail engines, and provides superior one-pass filtration to the RACOR systems.
I decided to take the radical step of completely re-engineering the fuel delivery system from the copper lines that emit from the fuel-selector valves in the lazarette to the engine. This involved stripping out all of the old cruft, including a restrictive check valve (intended to prevent drainback when the engine is not running) and a lot of old hoses and three-way valves of various descriptions.
The smaller size of the new system also allowed me to install a Walbro continuous-duty rated fuel pump at the same time. This pump provides the system with positive-pressure priming and, with the addition of one valve and one "T" for the return line, fuel polishing. A 12-hour spring-wound timer provides for an up to 12-hour cycle for polishing purposes. The Walbro is rated to deliver up to 1 gallon per minute at 7 psi, or 60gpm. Each of my tanks holds approximately 290 gallons of fuel, so to turn it all over twice I need to run the pump for approximately 10 hours when in "polishing" mode. In addition it can also serve as a fuel transfer pump if the selectors are set to different supply and return tanks (care must be exercised to prevent an overflow!) Power for the Walbros were taken from the main house panel in the engine room via a spare fuse slot; a 20A fuse was selected as each pump requires a 10A individually, and 12 gauge wire was used for the primary runs.
Here are a couple of pictures:
This was taken during the installation; the startobar side is complete other than the temporary wiring for the pump (to check that it works!) while the port side is the original. Note the tremendous size difference. The little black cylinder with the copper top is the Walbro pump, and the ball valve that is closed just under the vacuum gauge is the fuel recirculation valve (open to polish, closed to run.) The bronze check valve just behind the restriction gauge allows the fuel pump to be completely bypassed when the engine is started (necessary, since the on-engine pump can draw radically more fuel than the little Walbro can deliver) yet closes so that the Walbro can pressurize the system for priming or polishing purposes. That check valve must be mounted in a "straight" position, since it relies on gravity to close the flapper. I decided on this over a spring-loaded valve as the spring-loaded ones present more restriction to forward fuel flow, and absolute sealing is not a requirement in this application.
This next image is the completed installation on both sides. Note the radically cleaner layout of the new system compared to the old, even with teh greater functionality. This is not a minor difference! No more snags on the centerline. Note also the restriction gauges which permit instant "at a glance" reading from the top of the engine room hatch - very handy while underway. Just pull and look; they are angled to be "in your face" intentionally.
The final image is of the port-side unit. The inlet side isolation valve is mounted on the terminal end of the copper pipe, and is not immediately visible (its off-image to the left). The outlet isolation valve is right in front of you. These two make filter changes a snap - turn them both off, drain some of the fuel out the bottom ports (note plenty of clearance to get a cup or other receptacle under there), unscrew and replace, then re-open the valves.
To prime, you open the valve to the extreme right (shown closed) and energize the Walbro. When the pump stops cavitating, indicating that it has expelled all the air in the filters (it runs much quieter when there is no air in the system) you then close the recirculation valve. This will cause the pump to pressurize the remainder of the system up to the on-engine fuel pump. You may then start up; as soon as the engine starts the check valve will be pulled open, allowing full fuel flow, and the Walbro shut down at your convenience. If you run out of fuel, there's no longer any hassle at all in getting going again.
To polish the fuel, open the bypass and set the spring-wound timer (not in picture) to the desired number of hours of operation. The pump will run until the timer expires, pulling fuel from the tank, through the filters and returning it back to the tank.
Note that the bypass valve must be closed before starting the engine or you will draw air up the fuel return line and the engine will quit almost immediately. If this happens you simply close the valve and use the priming pump to re-fill the lines.
Fuel transfers can be accomplished with either pump by setting the supply and return tanks for each side to opposite values, then energizing the Walbro pump. Be aware that this can result in overfull tanks and fuel spills, so this option is only to be used with extreme care. However, if there is some reason you need to drain a tank entirely, and the other has sufficient capacity to hold the fuel you intend to drain (e.g. for service to the fuel pickups, etc) you now have a simple on-board way to do it.
The fuel filter block assembly is available from Tony @ Seaboard Marine, email "tony@sbmar.com". The pump is a Walbro diesel fuel pump and is available from Depco Pumps. Miscellaneous fittings and hose are required to complete the installation, and of course mounting arrangements must be made. In my case there were existing aluminum risers on the engine bed caps that the RACORs were previously bolted to - these only had to be drilled to accept the new bolt pattern. I also used a block of 1/2" starboard behind each mounting bracket to afford greater clearance between the spin-on elements and the stringers, so as to make the use of a strap wrench for element removal simpler.
Results? I now cruise with total vacuum restriction of approximately 4-5" Hg with clean filters, and 7" Hg with a day's worth of polishing on one side. Detroit's recommended maximum is 12" Hg, so there's plenty of margin. For comparison the Racors these replaced used to show approximately 4-5" Hg of restriction on their gauges, however this was not a true reading, as there was a 1/4" pipe thread check valve upstream of the filters which was not part of the measured restriction. I suspect, but cannot prove, that the total restriction as seen by the fuel pump was dangerously close to Detroit's recommended limits, even with clean filters.
It will be interesting to see how the on-engine "last chance" filters look after a couple hundred hours with this setup. Previously they had to be changed on 100 hour intervals (with the oil) or they would load up and eventually restrict power output. I will be most interested in the long-term testing to see if they remain clean and restriction-free for a significantly longer period of time.