Dodge Cummins Diesel Forum banner
1 - 20 of 363 Posts

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #1 · (Edited)
After reading Cowboy303's fuel system improvement thread, I have been inspired to get the absolute most from my stock fuel system. Before you continue reading, go read his thread. There's lots of useful information on upgrading the stock fuel system in there. Additionally, you may want to read the TorkTek articles (here and here) on the stock pump, though I believe they oversimplify the issue in saying that the problem lies mostly in the suction line. The whole system needs a plumbing overhaul. Also, they miscalculated the pump displacement by not factoring in the rod diameter. Another good fuel system resource to be familiar with is Joe G's Fuel System Writeup. Though I will be recombobulating all of the fuel plumbing in the truck and most of the components, this article covers the stock plumbing and components in great detail.

I’ll warn you right now, this is a long one. The text alone of the initial write-up is over 8,000 words and would be 14 printed pages, so it might be a good idea to grab a beer and make some popcorn. Those of you who have read my other threads know how nutty-bananas meticulous I can be, and fluid dynamics compounds this issue. I have spared no “crazy” in this endeavor.

My assertion: There is a common consensus in the performance diesel world that high-horsepower trucks need high volume electric lift pumps like the FASS and Airdog. I disagree with this, and you will understand why after we examine the capabilities of the stock lift pump.

 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #2 · (Edited)
How much horsepower can the pump support?

Calculating the displacement of the pump to .255 cubic inches, we find that the pump itself can flow an absolute theoretical maximum of 33.157 gph per 1000 crankshaft rpm. Now, no pump is perfectly efficient and we will never see that full 33.157 gph per 1000 rpm. Additionally, efficiency in most pumps (especially piston pumps) will always drop as speed increases. Cowboy303 measured 1.029 gpm (93% efficiency) at 2000 rpm with his setup. This measurement was flow from the return line only, meaning that the fuel consumed by the engine during testing was not counted. My initial goal will be to carry this 60 gph out to 2500 rpm. Realizing that 60 gph is only 72.4% efficiency, my secondary goal will be to increase that efficiency back up to 90%, which would give me 74.643 gph.

We now have a flow value, how do we convert that into horsepower?

All fuels have a certain amount of energy content in them. A common measurement is the British thermal unit or BTU. A BTU is the amount of energy it takes to cool or heat one pound of water one degree Fahrenheit. A gallon of #2 diesel fuel has between 128,400 and 138,700 BTUs per gallon, depending on quality and seasonal blending. For our calculations we will average this to 133,550 BTU/gallon. Most units of energy can be easily converted between each other, and 1 horsepower is equivalent to 2544.434 BTUs. Simple division will tell us that 133,550 BTUs works out to 52.487 horsepower.

Now before you get giddy after mentally multiplying that horsepower number by the pump’s flow rate, we need to keep in mind that most of the fuel’s heat energy is lost in the combustion process. It’s a well-established fact that that only about 37% of the heat energy released during complete combustion — complete being the operative word here — is converted into crankshaft horsepower.

Where does that energy go?

The short answer is anywhere it can. Most of it escapes out the tailpipe or into the cooling system as heat. Mechanical friction between moving parts absorbs some of this energy. Normal blow-by past the rings? Wasted energy. Is fuel still burning when the exhaust valve opens? Wasted energy. Even the light and sound produced by combustion is energy that will not drive the crankshaft.

Finding crankshaft horsepower is helpful, but not entirely useful as the crankshaft does not propel the vehicle. We have to go another step, and realize that only about 21%-30% of the total heat energy, depending on what source you cite, makes it to the tires and is able to move the vehicle.

In addition to all the sources of energy loss I mentioned, we need to keep in mind that every single moving part of the powertrain is a small parasitic energy loss. Every engine accessory, every bearing, every gear, every shaft, and every seal between the piston crown and the tires either add friction or takes energy to rotate. The tires have a lot of mass and rolling resistance. Some vehicles have inherently more efficient drivelines than others (automatic vs manual transmissions is the elephant in the room here), but all bleed some energy.

So, what are the numbers?

Using the known value of 52.487 horsepower per gallon, and our minimum acceptable efficiency of 72.4%:

  • With 37% of potential horsepower making it to the crank, we can net 1165 hp at the flywheel.
  • With 21% to 30% of potential horsepower making it to the rear wheels, we can net between 661 and 945 hp.
So there we have it, a conservative 60 gph from the stock lift pump will support 650-950 rwhp. Hope you kept the receipt for that FASS 150.

If we can reach 90% efficiency:

  • With 37% of potential horsepower making it to the crank, we can net 1450 hp at the flywheel.
  • With 21% to 30% of potential horsepower making it to the rear wheels, we can net between 823 and 1175 hp.
Now, one thing the FASS and Airdog setups purportedly do that the stock Cummins fuel system does not is separate entrained air and cavitation from the fuel. Both types of phenomena reduce the fluid volume, and therefore the efficiency of any pump they pass through including the injection plungers, and it can also damage the components themselves. That 74.643 gph will only support the aforementioned power levels if it is free of cavitation and aeration, which we can accomplish so long as we pay adequate attention to the plumbing.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #3 · (Edited)
Understanding Aeration and Cavitation

Aeration and cavitation are sometimes conflated in the diesel world, so let me straighten it out:

Aeration is the entraining of air into the fuel flow. This can occur many places, and we generally see it caused by a loose fitting or pinhole in the suction line. This defect allows atmospheric pressure to push outside air into the fuel system, because the suction side of the system is under less absolute pressure than the atmosphere. When these comparatively large bubbles pass through the pump, the bubbles are whipped into a froth. If you have ever played with a bucket of diesel fuel (I had an interesting childhood), or looked down the filler neck as you top off your tank, you know this froth does not dissipate quickly. In a closed system, it will generally not coalesce back into a few larger bubbles before it passes through the filter and into the IP. The result is that the plungers in the IP are trying to compress and inject foamy, less-dense fuel instead of pure liquid fuel. As you can imagine, subpar injection pressure and the accompanying performance penalties result. The solution here is straight-forward: make sure all the plumbing is tight and leak-free.

Aeration can also be created by fuel splashing. We can’t do anything about fuel sloshing around in the unbaffled tank, but what we can do is keep the end of the fuel return below the waterline, and the further away from the pickup the better. This is to allow any aeration or cavitation in the return line to dissipate and larger bubbles to make their way to the surface.

Cavitation is a completely separate mechanism that also creates lots of problems. When a liquid’s fluid pressure drops below its vapor pressure (.2psi for diesel), small voids can form in the liquid. The voids are composed of vaporized liquid. This is the same mechanism as boiling, except instead of adding heat you are subtracting pressure.

In addition to forming in large low-pressure environments, cavitation can also form in highly localized low-pressure areas of a medium-pressure environment, such as vortexes created by fluid rapidly accelerating and/or changing direction through a restriction. Running the pressure side of the system as high as is practical will be very helpful in preventing cavitation from forming in those areas, but since we are changing all the fittings anyway, we should pick out new ones that will reduce flow restrictions and changes in direction as much as possible.

Cavitation can also be caused by mechanical vibration and sound waves physically separating the molecules in the fluid to the point that a low-pressure area forms.

Aeration and cavitation also both have a major capacity to damage components as well. The first and most obvious reason is that a void in the fuel between moving parts allows for direct part-to-part contact with reduced lubrication. Additionally, have you ever hooked up a fuel pressure gauge to the Cummins engine without a snubber or needle valve? It’s easy to see that the piston pump produces very rapid and violent pressure spikes. When aeration or cavitation is present on the surface of a fuel system component and the pressure spikes, a jet of liquid fuel will strike the surface of the part as the bubble collapses.



Over time, this can be very erosive. Cavitation can be prevented by choosing large line diameters to reduce the speed of the fuel traveling through them, and optimizing flow through each fitting and fuel system component.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #4 · (Edited)
Planning Stage

Okay, so you can do it, but why not just FASS up and call it a day? All the cool guys do it.

Because there's no reason to. The pump itself is not the problem, the plumbing is, and I'd have to redo all of my plumbing anyway for a FASS to operate efficiently, so that's a moot point. The mechanical pumps last an average of 300k miles. Show me one credible instance of a FASS, Airdog, Raptor, etc lasting 300k and I'll buy you a case of beer. The mechanical pumps are compact, lightweight, require no electricity to run, and the best part ... they are OEM parts! Your truck already has one, and if it’s at the end of its 300k mile lifespan, you can buy new ones all day long for $100! Also, I take no pride in being a lemming.

The first step is to plan it out, because fittings (and components) get expensive if you keep changing your mind about what you want to do. I’ll start by telling you how my fuel system was set up before this upgrade.

The fuel module was stock but modified in accordance with the DrawStraw installation instructions, with a new hose that extended all the way into the screen. 3/8” rubber fuel hose was clamped directly to the pickup nipple on the canister, and ran directly to the prefilter inlet without passing through any of the stock 5/16” hard line. The heater was deleted. The rest was stock except for the filter, which had been upgraded to the Fleetguard FS19519. The OFV had been replaced with a TorkTek about a year and a half ago. I was running my fuel pressure up around 45-50 psi, and could easily send it into the 30s with a heavy mash on the fast pedal. My approximate rear-wheel horsepower at the time was ~350. Coasting in gear down a long hill would peg the 60 psi gauge. The return was stock, but patched with a fresh piece of 5/16” rubber hose.

My first real upgrade was to replace the prefilter inlet fitting with a 3/8 NPT x 1/2” barb, drill out the inside of the LP inlet elbow to 1/2”, install a new lift pump, and replace the factory spring with the 975-1 valve spring. This is all documented in greater detail in Cowboy303’s thread, and was done because I suspected I had killed my old lift pump by revving the engine to 3900 rpm.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #5 · (Edited)
Sending Unit

For my new system, I wanted a sump. Here is why:

As I was preparing to perform this fuel system upgrade, I ran my tank as low as I dared to make draining easier. Fuel pressure was all over the place before the low fuel light was constantly on, but I knew I was really low when I started feeling the truck hesitate coming off of stoplights. I got it home and filled a 5-gallon bucket with the fuel left in the tank. Essentially unusable fuel, mind you, and enough of it to travel a hundred miles. Modifying the fuel module did help what's commonly referred to as the 1/4 tank issue, but obviously I wasn’t happy about having 5 unusable gallons in there. With a sump feeding from the bottom, you would get to use significantly more of this fuel before air infiltration from the sloshing caused your motor to die.

A little thing called head pressure. There are two pickup configurations, a straw and a sump. The sump is superior from a fluid dynamics standpoint because of head pressure. Head pressure can be referred to in simple terms as "gravity feed." The weight of the fuel above the pickup helps to push the fuel into and through the suction line. Head pressure is what makes a glass heavier than normal if you submerge it and try to pull it bottom-up through the surface of the water. With a straw, head pressure actually works against you because the weight of the fuel in the straw above the level in the tank is actually fighting the atmospheric pressure pushing the fuel up the straw. Though minor, this increased pressure differential helps create cavitation.

Which sump do I want? I went with the Deviant for a couple reasons. The clamp plate has more uninterrupted circumference, allowing for an inherently more positive seal. There are also two bolts used, which further increases clamping force and the seal. It is a hex shape instead of a cylinder. This would make cutting additional ports into it easier if a person wanted to plumb an auxiliary tank or return into the sump.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #6 · (Edited)
Water Separator

Since a sump lives at the bottom of the tank, and will absolutely pick up any water and sediment that collects, I need a way to get that junk out of my fuel system before it goes through anything important. The class 6 Izusu trucks at work have a small water separator attached to the frame next to the tank that is simply a plastic bowl with a drain on it:



I was able to find a similar unit on eBay. I was unable to find the original application, but it holds a comparable volume to the Isuzu separator and has larger ports. They are an M14 thread, which is not ideal for my 1/2” plumbing, but it will have to work. This will be mounted to the frame just in front of the fuel tank.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #7 ·
Prefilter/Pump

As mentioned before, the prefilter and lift pump were freshly serviced and upgraded. The notable thing here is that the spring has been upgraded. With a stock spring, the LP pushrod can float on the cam when engine RPM is increased above 2500, or when the system pressure is increased over ~35psi. I ran my pressure in the mid-40s for a long time on a stock spring, but rarely revved above 2000 rpm. I want to run my operating pressure around 50psi to prevent cavitation and aid in plunger filling. My 160 pump seems to run better the higher the pressure is, and I want to be able to rev all the way to 3000 without pump float. There is another, heavier spring (973-1) that I can install if I find the 975-1 spring isn’t up to the job. The LP inlet fitting has also been drilled out to ½”.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #8 · (Edited)
Plumbing

Between the LP and the IP, there is a lot of restrictive plumbing. Most of the fittings (all but one to be exact) between the LP outlet and the IP outlet are banjos. This is a big no-no for efficient flow. Banjos have their place, but in passing through one, the fluid has to rapidly change direction and rapidly accelerate multiple times to make it through the small ports in the banjo bolt. To rectify that, I opted to go with the ready-made 1/2" 12VFS kit from Vulcan Performance. It's a good, solid, complete kit that clearly had some thought put into it, and I like supporting local businesses. I had a question about installing it and Eric replied to my email very promptly. I look forward to doing more business with them in the future. The one deviation I will make here is that I will modify my fuel filter mount similar to how Cowboy303 did, which will simply require changing the provided M12 to -8AN fittings to 3/8NPT to -8AN fittings.

Why did I choose 1/2” line? Because it’s the largest line I can run that’s practicable. The other choice would have been 3/8”, but 1/2” flows almost twice as much. But wait, 3/8” isn’t half of 1/2”…



Correct. Cross sectional area (and therefore flow, when pressure is kept the same) increases as a factor of the radius squared. This gives us a parabolic relationship between flow capacity and line diameter. Double a diameter, and you nearly quadruple the flow capacity. Inversely, if you halve a diameter, it quarters the flow capacity. If we want the best flow, and the best protection against cavitation, we need to be pricks about diameter. This is not so much a concern for the line size, but for the fittings. All of the fittings neck down in size, some down to as small as 1/4” ID, which creates a major flow restriction. We need to modify them for the largest possible ID to reduce that restriction.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #9 · (Edited)
Filter

I wanted a bigger filter without a WIF sensor provision. Here is why:

I have never gotten any water in my fuel filter. If I did, it would have already traveled through the LP which I wouldn't have liked. I want my water trapped before it travels through anything important. The few inches in height that the WIF sensor takes up over the lift pump is a few inches longer that the element itself could be. A larger element translates to more dirt-holding capacity (longer change interval) and less flow restriction. A side note was that both of the WIF sensors I've had in my truck have been chronic leakers. Not much, but enough that adjacent components are always damp. Given the other reasons, I was disinclined to buy another one.



Here is a chart showing all the filters that will fit in the stock location. I initially chose the Fleetguard FF5421 but I went the next size down (FF5612) for reasons I’ll discuss later. I was unable to find Fleetguard specs on filtration or flow, but the Wix cross-reference filters to 7 microns and is rated to flow 8-10 gpm. Since this system will be operating at a nominal 1.25 gpm, this should make for a very long change interval and very low restriction until it gets there.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #10 · (Edited)
Regulator

After the IP, I wanted to ditch the factory OFV style regulator. Now, I am not knocking the OFV. There are hundreds of thousands of Cummins engines out there that have logged millions of hours and millions of miles using an OFV. It is super simple and reliable, but it is also a major flow restriction when delivery volumes are increased. The inlet orifice on my TorkTek is only 1/8” in diameter (take another look at that graph):



Flow capacity through the regulating device is especially important, because just driving the truck down the highway, the vast majority of fuel supplied is returned to the tank. For example, if your truck gets 20mpg on the highway, and you're running 1700-1800 rpm, your delivery rate is around 50 gph while your consumption rate is 3 gallons per hour. This means only 6% of supplied fuel is consumed by the engine, with the other 94% (47 gph) needing to flow through the regulator and return to the tank. The pump trying to push most of the fuel it's delivering through this 1/8” pinhole is a major restriction that we need to do our best to get rid of.

The effects of this flow restriction can be effectively described by Cowboy303’s discovery that in a fuel system modified for better flow, the OFV actually opens well below the maintained system pressure. Let’s evaluate Cowboy303’s example in slow motion. His system maintains 45psi going down the highway with a constant load, and varies quite a bit with RPM (there’s the first clue). The cam lobe compresses the LP spring, allowing it to fill with fuel. The cam lobe falls away and allows the spring to pump fuel. Pressure builds in the system until 15 psi. The OFV opens, allowing fuel to leak past it. The remaining 30psi of pressure increase past this opening point is entirely caused by the restriction of the OFV itself in the return line. When the LP spring is fully uncompressed, fuel continues to bleed out the valve until pressure drops below 15psi, if it even can before the pump begins its next feed stroke. If it doesn’t, fuel is constantly flowing through the OFV while the pump delivers more fuel intermittently. The OFV doesn’t have the flow capacity to keep up with the pump’s short duration, high volume delivery, so it compensates by venting less volume over a longer duration.

Though the reality of the OFV’s pressure relief function does not make for stable fuel pressure under varying load conditions, it does not necessarily harm engine performance so long as reasonable pressure levels can be maintained under full load. That said, when we increase engine RPM and start delivering more fuel than the OFV can vent into the return line, it creates a problem in the form of ever-increasing pressure between the LP and IP. There is very little in the fuel system that could be harmed by the high pressure in and of itself, though this pressure can increase enough to cause the LP's pushrod to float on the cam, especially at high RPM. This can physically damage both components, and when the LP stops pumping, fuel delivery ceases.

It’s important to understand how the premature draining of fuel from the IP will hurt us when trying to support lots of horsies. Keep in mind that the IP is constantly sucking fuel out of the pressure side of the fuel system. That’s why, in a perfect world, the regulator would stay closed until just over the maintained system pressure, be able to let the excess fuel flow out at the same rate it’s flowing in, and close as soon as the delivery stroke ends. This way, inter-stroke pressure drop is kept to a minimum, and a minimal amount of fuel is wasted during the pump’s delivery stroke, as only what is truly “excess” fuel is returned to the tank. I mentioned before that the OFV setup is simple and reliable, and works for most people. In that regard, it is not an un-suitable way to regulate fuel pressure. However, this setup needs to be changed if we want to utilize the “stock” fuel system’s full potential.

My first thought was to delete the OFV and put a 14mm banjo to -6AN adapter in its place, and installing a 3/8” hose to a standalone regulator. However, after examining the part, the two ports on the banjo bolt were only .195" in diameter each, which gives a combined CSA of .061". This is 5x better than the OFV, but I wanted more. I ended up selecting an M14 to -6AN elbow adapter, which I was able to drill out the M14 portion of to 5/16" (.077” CSA or 6.4x better than the OFV). This is still a far cry from a 1/2" line (.196” CSA), but it's the best we can do without cutting larger threads in the IP outlet, which I might look into if/when the IP comes off for another reason.

Fuel pressure will be controlled with an Aeromotive regulator. The one I chose (13129) has two inlets and one outlet, all -6 o-ring bosses. I couldn't find an actual flow capacity, with the manufacturer simply stating vaguely that it “can support up to 1000hp.” It should do the job. A really cool feature of this and most Aeromotive regulators is that they can be boost referenced, meaning your fuel pressure is regulated relative to boost as opposed to atmospheric pressure. I understand this is mostly for the gasser crowd, but it’s a neat feature to have. Rebuild kits are available, and with one, a rebuild can be performed on the side of the road in a matter of minutes, should the diaphragm start leaking.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #11 · (Edited)
Return

There is one part of our fuel system that will create cavitation which we can do nothing about. This is the regulator, because when system pressure is low and the valve is mostly closed, the fuel flowing through it accelerates significantly in traveling through the valve. This high speed causes a pressure drop, which allows the fuel to cavitate. This does not pose an immediate problem to the engine, except that this cavitated fuel is returning to the reservoir that fresh fuel is being drawn out of. What we can do is keep the return fuel far away from the sump and segregated from the bulk of the fuel supply. This will be accomplished by making the old suction line the new return line. Fuel will be directed back into the basket, where the cavitation in the return fuel will have a chance to dissipate before the fuel is drawn back into the sump. This also means that the return discharge will be below the waterline almost all the time, which will avoid the problem of return fuel splashing and aerating the fuel.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #13 · (Edited)
Execution

Sump

Starting with the tank, the first order of business was to drain it. I initially tried disconnecting the return line and directing it into a fuel can with the engine running, but with the level this low, all I got was a couple gallons before I started getting a pathetic dribble of foam.



At that point, I just used the pilot bit for the sump’s hole saw and punched a hole in the tank. My truck was the perfect height to just fit a 5 gallon bucket under the tank.






Once the fuel was drained, I ran the hole saw through it. The required size was 3.25 inches.



It’s important for a good seal to clean off all the stringers the saw leaves behind.



I used a half-round file and an exacto knife until all the edges were smooth. Also scoop out as much of the the plastic chips from the inside of the tank as you can.



I was pleased to inspect the inside surface on the cut out portion, with no trace of sludge, slime, goop, sediment, or any other undesirables.

Next, I assembled the fittings on the sump. A 1/2x1/2 NPT hex nipple, ball valve, 1/2 NPT to -8AN adapter, and finally a -8AN push-lok.



Why did I valve it? So I can work on the fuel system without draining the tank. Also, it allows me to pressure-test the lines without pressurizing the tank. Pressurizing the tank is not a good idea once a sump is installed. Once I’m satisfied with the system, I will remove the handle and put it in the truck toolbox in case I need it later on.



I installed the sump but didn’t tighten it all the way, waiting until the water separator was installed so I could clock the sump appropriately.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #14 · (Edited)
Water Separator

Before installing, the water separator needed some work. The M14 ports on it are a major restriction point in the supply line, so we need to massage every little bit of flow area we can out of the housing. In as-received form, the galleries in the casting were a wimpy, measly, disgusting ¼”. This reduces flow by 75%, and that’s completely unacceptable.





I took some careful measurements from the inside of the mount, put it in the mill vise, got it perfectly centered and parallel to the chuck, and selected a drill bit that would leave about .040” of wall thickness on the outlet gallery. I couldn’t get inside the outlet “well” area to open up the port, but enlarging the hole itself opened up the port enough that it had more CSA than the gallery itself.





The inlet gallery was enlarged a little more to .375” and the gallery inside the bowl area was removed completely with a carbide burr, blending the opposing face into the “ceiling” to direct flow down into the bowl.





Here’s the water separator installed.





Two through-bolts were used, and this is outboard on the driver’s side frame rail, just even with the front of the tank. The fittings are M14 to -6AN adapters, then -6AN to -8AN, then -8AN push-loks. A quick note here, the little plastic drain valve proved to be a pesky air leak that I couldn’t get rid of. I couldn’t get it tight enough without breaking the plug, so I replaced it with a 1/2” long 5/16-18 stainless bolt, reusing the square cut o-ring.



This is how I initially routed the fuel line, but I didn’t like the hose rubbing on the cab mount so I later installed a 45 degree -8AN fitting to give some better clearance.



Here's the clearance between the cab mount and the line:



A quick note on the push-loks, I found coating the inside of the hose and outside of the fitting with diesel was a great way to lube it to make installing the hose easier. Also, clean out the insides of any hoses you cut with a suitable cleaner before installation, especially on the clean side of the filter!
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #15 · (Edited)
Engine Bay

After running a new 1/2” line from the separator to the barb already installed in the prefilter housing (pics), I removed the filter and mount to get started on the engine plumbing.





I ran the line on top of the frame rail.

I decided to get the filter mount dealt with first.



It was very simple, I put the mount in my milling vise, drilled and tapped both M12 ports into 3/8 NPT. I also took the inside of the filter stud out to 3/8” so there would be no further restriction there.



Here it is with those nice, shiny -8ANs in there.





Since drilling out the stud removed the hex that is used to install the stud, I pressed a T60 socket into the new hole to give me a way to install it.



After that was done, I installed the Vulcan outlet fitting on the pump. This was the only part of the install that was really a pain. The protrusion from the tappet cover interfered just enough that I couldn’t freely thread the fitting on. I had to try setting it on with each of the six “flats” facing the protrusion, and forcing it with a wrench to see if a thread had caught. Eventually one did and I just kept forcing each “peak” on the hex past the tappet cover until the fitting was tight. I aimed it to the rear so the line coming out goes behind the filter.



Next I installed the IP inlet fitting. Not much to write about. Here’s a pretty picture.



Next, I went to get the IP outlet and regulator mounted. I drilled the inside of the M14 to -6AN adapter out to 5/16” and found I would have to grind some reliefs into the head to get the thing to thread in.







The reliefs are on each side of the square slot pictured. I used a carbide burr and took off just enough to get the necessary clearance.



After that was in, I installed the regulator line. It’s 3/8” in diameter, 30” long, and has female -6AN ends. The local hose shop made it up for me. With the regulator going where it went, it would have been better to make it 28” long.



The regulator mount is very simple. It’s a piece of strap cut to fit, and it mounts on one of the master cylinder studs.







The hose runs over the reservoir but there’s plenty of room to get the lid on and off, and there’s still plenty of room between the hood and the regulator.







The fittings on the regulator are all -6 o-ring bosses to -6AN male. The bottom port is the drain to the tank, and the valve assembly on the driver’s side is a depressurization valve. Open the valve, and it allows the fuel to bypass the regulator and go back to the tank. This is mostly for aiding in priming the fuel system, but makes a convenient place to attach a transfer hose if I needed to move fuel into a container or another vehicle.



Next up was the reinstalling the filter mount. The mount needs to be raised up for the filter to clear the pump. Since the bolts that attach the mount are metric, I decided to work in millimeters so new bolts would be easy to figure out.



I made two little standoffs on the lathe, 15mm high with a 5/16” hole. I forget the OD. It was some leftover stainless stock, and the diameter worked so I just cut it. The new bolts are M8, 50mm long. The filter mount also needs the slots elongated to allow the filter to sit slightly outboard. This is because the filter rubs on the head when the mount is raised this far. I used an end mill and elongated the holes by about .100”, and this seems to be enough clearance.



I cut the LP to filter hose before installing since it would be a pain to get to with the filter installed. It was at this point I discovered the housing still needs to be loosened to install the FF5421. I wasn’t crazy about that hassle, so I decided to go with the slightly shorter FF5612 instead.

The injector drain manifold needs to have the banjo cut off.



I used a small file to do this, and cut it back close to where the bolt tab is. I made sure to sufficiently blend the edges before pushing the rubber grommet thing on. I screwed up here and cut the hose too short, but it doesn’t leak so it can stay that way for now. I’ll do it up proper next time I pull the injectors.

Vulcan put a -4AN fitting on top of the filter’s outlet fitting, where the bleed screw used to be. This is thoughtful, but it won’t be used as a bleed fitting because I don’t want to douche my engine bay with fuel when I have a fancy-dancy de-press valve in the return line. This does make an extremely handy spot to hook up regulated shop air to leak-test the fuel system and to hook up a pressure gauge, however:

 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #16 · (Edited)
The two return lines (main and de-press valve) were teed into the old 3/8” feed line still attached to the factory fuel module.





I may do something different with it if I have to drop the tank again, but it should be good for now. This puts the return line emptying in the basket which should do a good job of keeping aeration/cavitation away from the sump.

That does it for the install. The bucket o’ fuel was siphoned back into the tank.





I then became a full-fledged idiot and forgot to plug my FSS relay back in before nearly draining the batteries trying to start the truck. The good news is I was successful in frying the starter contacts before I figured out what was going on. Derp to the derpteenth power.



 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #17 · (Edited)
Road Test

I idled the truck long enough to set my idle pressure to 50psi and went for a drive down to the filling station. I didn’t expect it at all, but I noticed right away after filling up that the truck had more power. Not tons, but enough that I didn’t have to convince myself I felt it. Throttle response definitely improved. I’m not sure where this is coming from, perhaps I had a small air leak I didn’t know I had? That’s the best I can come up with.

On the highway, pressure was sitting around 52psi on level ground (4 psi boost, 60 mph, and 1800 rpm). Maintaining speed up hills didn’t take it lower than 50 (pushing 10 psi boost). Maintaining speed in-gear, downhill, it would climb to about 54. Coasting while in gear would bring it to 58. I wound it up to 70 (2100 rpm). Fuel pressure climbed to 57psi and didn’t fluctuate much under load, but would peg the 60psi gauge going downhill.

So, while definitely an improvement over the OFV, it appears this regulator doesn’t quite have the capacity to handle whatever my current flow is above 2000 RPM.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #18 · (Edited)
Regulator Mod

I won’t be buying another regulator, so I decided to take the 13129 apart to see if I could modify it for better flow. Since I want to rev to 3000, and the current port starts choking in the neighborhood of 2000, I need to increase the relief port’s CSA by at least 50%.

A note of caution, back the adjustment screw out all the way before disassembly. I tried taking it apart without messing with the adjustment, but that was a bad idea. It turns out the spring is under pretty good pressure, and when the last screw finally let go, the top body of the regulator sprung violently upward. Luckily, my cat-like reflexes allowed me to Matrix my way out of danger.

This is how the regulator is set up. It’s very simple and I like that.



The relief port in the regulator body in stock form is .188”:



The valve that closes against it appears to be a ball bearing pressed into the diaphragm. I measured the ball, and the largest protruding diameter was .240”. Oh goodie, we have some room!



I selected a 15/64” drill bit, because at .232” I think it’s the largest diameter we can get away with and still have the ball make a good seal against it. I made the .188” hole a .232” hole.



When I set the diaphragm on it, the ball still rolled a little bit in the hole before hitting the sides, so it should work nicely. Let’s do the math on this size increase. .188 = .028” CSA. .232 = .042” CSA. .042 / .028 = 1.523, or a 52.3% increase. Boom, baby.
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #19 · (Edited)
Flow Testing

Speaking of flow, what is my current flow? I dunno.

I was going to do flow tests by timing how long it took to fill a 5-gallon bucket at various RPMs. Though time consuming, this is workable. But, the flow numbers we really need to be concerned with are above 2000 rpm, and I have some reservations about repeatedly running my engine with no load on it up there for as long as it would take to fill a 5-gallon bucket.

I found a cheap rotameter on eBay and plumbed it into the return line as shown.



The inlet of the rotameter is connected to the bottom port on the regulator. The outlet of the rotameter is connected to the return line, visible on the right.

This allowed me to get instantaneous flow readings across a wider RPM range, ultimately giving us more data points to establish more detailed flow and efficiency curves.

I had about 6' of line so I could sit in the cab while I did the tests.



I thought I was done working on the fuel system, and I was wrong. At idle, the lines to the rotameter were jerking back and forth with each stroke, and I could see air bubbles in the rotameter’s fitting threads jittering around. This was caused by pressure spikes in the return line on each delivery stroke. Obviously, there’s too much restriction in the return line. We will talk more about this later.

Test Protocol

All tests where flow and pressure numbers are recorded will be conducted with a full tank of fuel, the truck on level ground, and the filler cap removed. The pressure setting on the regulator will not be altered until testing is concluded, though we are mostly concerned with the shape of the curve as opposed to what the pressure actually was. Tests will be conducted 3 times and the readings averaged.

Flow and pressure data will be plotted on the same chart. Readings from the rotameter will be corrected by a factor of 1.190 because diesel has a different specific gravity than the water which the rotameter is calibrated for (diesel is .84, water is 1).
 

·
Premium Member
Joined
·
7,254 Posts
Discussion Starter · #20 · (Edited)
Test 1 Results - Invalid data, details in post 26.



Those of you who are paying super close attention will notice that the first few flow readings are actually above 100% volumetric efficiency, which isn’t possible. This readout error is caused by the fact that rotameters are not accurate when the flow is rapidly changing as it does with a piston pump. A pulse of fuel comes in and drives the weight up the rod, making it read high because of the higher-than-average velocity of the fuel, and the weight’s upward momentum. Then the weight free falls during the lower-than-average flow between delivery strokes. There’s not a good way to compensate for this, except to do a volume over time test (bucket method). However, as engine RPM increases, the flow spikes and time between delivery strokes both decrease, so the rotameter becomes more accurate at higher RPM. We are primarily concerned with flow above 2000 rpm, and since mine and Cowboy303’s respective VEs at 2000 rpm are only .7% apart, I am not concerned about the veracity of the data at and beyond 2000 rpm.

Flow ramps up until 2000 rpm where it starts to level off, ultimately peaking at 66.4 gph at 2500 rpm. At 2750, the flow is the same. At 3000, it has started to drop. This indicates 2750 rpm is where a suction-side restriction or the limit of the pump spring starts come into play. If we continued the test to 3800 rpm, it would likely reveal a rather drastic drop-off in fuel flow.

The pressure curve is reasonably flat, which is what we want to see. The maximum spread is 12psi, which is not bad considering the spread of the fuel flow through the one regulator valve. We can see that pressure begins to drop slightly, verifying that the flow-limiting restriction is on the suction side of the system or perhaps in the pump itself. This also shows us that the regulator is no longer choking at the flow levels we are currently seeing.



The good news is I have blown my initial target out of the water on the first go, beating it by 6.4 gph. Efficiency at 2500 rpm calculates to 80.6%, which is a significant margin over the minimum target of 72.4%. Using numbers provided earlier, this flow will deliver between 731 and 1045 rwhp worth of fuel at 2500 rpm. Now, let’s see about that 90%.
 
1 - 20 of 363 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top