System1 Fuel Filter on custom made bracket

With the fuel tank guts taken care of, next is the plumbing from the tank to the injectors. The first hurdle is the fuel filter. Anyone who has changed the OE fuel filter knows it is royal PITA. There is nothing like the feeling of cold gasoline running down your arm into your armpit as you struggle to pull the fuel hose off the fuel filter nozzle…blind. Since I planned to run all braided SS lines, the OE fuel filter was history. I had picked up a HUGE 11” System1 fuel filter off eBay (w/-10 fittings, 30 micron filtering, and flowing 28gpm) for a song-and-a-dance. Now I just had to figure out how to mount it. Some things to consider; 1) it had to be protected from road debris and crush in a significant rear end collision, 2) it had to be accessible for service (unlike the OEM filter), 3) it had to be placed such that the large -10 lines could be routed easily to and from it, and 4) it had to be in a place where it could actually be mounted properly – I don’t “do” zip ties…PERIOD. I eventually settled on a location just in front of the fuel tank mounted to the rear subframe. I custom made a bracket to hang the filter and used rubber to isolate the filter from wear and tear of a rigid mount.

Fuel filter mounted to rear subframe. I used the existing holes in the subframe for the mounting points = no drilling.

Since I had success with fabricating the braided SS oil cooler lines with AN fittings for my ITS RX-7 HERE I decided I’d make all the braided lines myself. This really isn’t a difficult proposition, just rather time consuming. There was considerable forethought to ensure all the fuel lines were the correct length prior to assembly. Ensuring the right type of hose end was used also caught me out a few times. For example a line that I had used a 90degree fitting on after trial fitment ended up needing a 120 degree fitting. But that’s the beauty of reusable hose ends, simply disassemble and clean up the hose itself and slide on another fitting. I used a Dremel tool to cut the hose cleanly and used an anaerobic thread sealant on each connection, even though it may not have been necessary everywhere. AN fittings do not/should not need thread sealant.

Cross section of Aeroquip reusable hose end fitting.

The feed line from the fuel tank cover bulkhead fitting (fed by the dual pumps) is a AN-10 line. Overkill I know but I had an inventory of fittings and hose to use so why not? Routing the short -10 line from fuel tank to the filter was a piece of cake. The next hose was the long run from the fuel filter to the splitter on the engine block. It was tricky to route the hose up and over the rear subframe but there is room to squeeze it through and in a manner that the hose isn’t kinked or would get pinched. The hose is completely sheathed in ¾” rubber heater hose to avoid any chafing and wear over its entire length. The rear section is fully hidden/protected by the rear plastic undertray. The hose was run down the length of the car on the inner side of the chassis frame rail with adel hose clamps every 12inches. It is well protected by the frame rail and as safe as the OE fuel lines mounted just to the other side of the frame rail protected by a flimsy plastic cover. Some argue that running the braided SS hose under the car is very dangerous because it may get snagged on road debris or an object and get ripped from the car with fuel being pumped out by the still running pumps. I have two responses to the: 1) if there is something large enough to rip the fuel hose from under my car then the OE return line will go to…as well as the brake lines. And then I will have much larger concerns since I won’t have any brakes! And 2) there is nothing that should justify running pressurized fuel lines on the interior of a street car. I cringe whenever I see surge cans mounted in the hatch area of a street car with no bulkhead separating fuel and driver. The alternative…run the lines outside the chassis...which is what I did. Let’s move on to the rest of the fuel delivery.

AN -10 line from the filter to the splitter mounted to the block. Note rubber hose sleeving for the section running alongside the OE plastic fuel line shield/guard.

The AN-10 delivery line feeds an aluminum splitter block mounted to the topside of the engine block. The block feeds two short AN-6 lines to each fuel rail – primary and secondary separately. These two short fuel lines were wrapped in a thermal sleeve since they run directly over the hot engine block. A custom made aluminum plate mounts the splitter block. This was possible because the ignition coils were moved and the majority of the hardware used to locate the solenoid rack (and the solenoid rack itself) was cut and removed such that only the UIM support leg remains. This left plenty of room for the lines and fuel hardware. The fuel rails are fed in parallel directly from the splitter.

Splitter block mounted atop the engine block (right of oil filler neck) feeding two AN -06 lines to each fuel rail. Coil pack and solenoid rack have been relocated. The fitting on top of the splitter feeds a fuel pressure sensor.

The OE fuel pulsation dampener was removed. The aftermarket fuel pressure regulator performs this function so no dedicated dampener is necessary. The AN -6 lines feed modified OE fuel rails. Why OE fuel rails? In the state where I live, emissions testing requirements are still in place. To have a truly streetable AND legal car I wanted to build an emissions compliant car that could meet the state emissions requirements and as such felt the easiest approach was to keep the OE air control valve (ACV) in place. Aftermarket secondary fuel rails will not clear the ACV. There are other band-aids to get through emissions but I decided to stick with the tried and true OE air pump and ACV with a stock main cat. Modified stock rails it was. Therefore I VERY carefully drilled and tapped the OE rails to accept AN fitting adapters. This was a very scary and delicate operation. Particularly so for the primary rail where there isn’t much aluminum material to deal with and the holes must be tapped precisely in order for the AN fittings to clear the neighboring oil injectors.

Stock rails fitted with AN fittings. Note on the primary rail there is very little clearance to oil injectors.

The fuel rails then discharge into a pair of AN -6 lines plumbed to an Aeromotive (A1000 series) adjustable fuel pressure regulator. With two pumps working to move fuel the ability to manage fuel pressure was imperative, hence the need for an adjustable regulator. A fuel pressure gauge was fitted to the FPR for ease of setting fuel pressure at idle even though I have an in-car fuel pressure gauge also. My in-car gauge is fed by a sensor plumbed into the top of the aluminum splitter block upstream of the fuel rails with its own AN -4 SS line. Idle fuel pressure was set to 40psi base pressure at the FPR gauge. The FPR vacuum line is simply plumbed to the UIM. The FPR discharges to a single AN -6 line back to the OE hard line at the firewall. I found a suitable AN to metric hard-line adapter fitting. It required I cut and flare the end of the OE hard line but the fitting fit perfectly and with no leaks. It is a low pressure return line so keeping it leak free was not difficult. Had it been a high pressure feed line connection I would have had second thoughts but this is low pressure so no worries. What I can share is that it took a lot of work and several attempts to figure out the best routing for the SS lines under the manifold. The lengths of each hose had to be perfect and the bend radius of each line maximized as SS hose doesn’t bend like a simple rubber hose does. To make room for the hoses, the coil packs were moved, as mentioned above, to the drivers’ fender, and all the control solenoids were removed or relocated. There is NOTHING under the UIM except fuel lines and electrical wiring.

Fuel Pressure Regulator on custom mounting bracket. Fuel pressure sensor (feeds in-car gauge) routing and mount to brake booster.

Fuel return hard line adapter for AN fitting. No hokey electrical tape and hose clamps on a SS hose here to make the connection.

All this fuel delivery and to feed what? I installed 2 stock 850cc secondary injectors in the primary rail and a pair of modified 850cc secondary rails bored out to 1300cc. The secondary injectors when tested were actually flowing about 1450cc/min each! (1424cc/min and 1460cc/min to be exact - Witchhunter flow tested at 43.5psi) That’s a lot of fuel! Some may argue the benefits of using 850cc primaries. Hard tuning at idle and poor off idle throttle response are common problems. Weigh those concerns over tunability during the transition between smaller 550cc OE primary injectors and larger 1450cc modified Denso (or larger 1600cc Bosch) secondary injectors. I chose the former approach for better control during moderate driving conditions and decided to forego the idle issues. Yes my idle is set at about 900-1000 rpms, still exhibits a rough idle, and has a tricky/sensitive off idle response. However, even with the big primaries, I was still able to pass emissions with flying colors. The larger injector combination is enough fuel to feed ~600hp. So no more issues of exceeding that 90% fuel injector duty cycle limit. At 15-16psi of boost and about 420rwhp I still only see about 80% duty cycle on the fuel injectors. Also there is not so much as a twitch in fuel pressure under full boost.

For more specifcs on the reconstruction of Princess go to the following pages:

Rebuild and porting of the 13B short-block to include port matching the intake side.
Custom baffled deep well aluminum oil pan.
Installation of dedicated 2 stroke oil feed system.
Installation of a dual fuel pump arrangement with fuel tank baffle box cover.
Relocation of the ignition coil packs to the drivers’ side fender wall.
Simplification and rebuild of the engine wiring harness and removal of all un-used control solenoids.
Porting and polishing of the throttle body and removal of the double throttle control assembly.
Turbo “improvements” to fix lower grade hardware and potential trouble areas.
Installation of 3 Bar MAP sensor and tuning for higher boost applications.