Auto Service World
Feature   February 1, 2001   by CARS Magazine

Force Fed Fuel

The heart of fuel injection isn't under the hood, it's in the tank

Believe it or not, there was as time when engine fuel management was much more difficult to understand than it is now. That time was almost exactly 100 years ago, when inventors such as Otto Daimler, Ransom Olds and Henry Ford struggled with ways to atomize just the right amount of fuel and ignite it at just the right time in their primitive one-cylinder engines. Injection was always on their minds, but the invention of the carburetor by Wilhelm Maybach in 1892 changed everything, because it was a demand-feed instrument. Engine vacuum regulated the fuel flow from a simple float bowl, a system that worked so well that it wasn’t challenged in most vehicles until the ’80’s when computer control became cheap and reliable.

Computer control made fuel injection practical for mass-market cars and light trucks, but the key difference between FI and carburetors changed the architecture of the whole fuel system. Injection is a supply-push system, not demand-feed like carburetors, and that difference is most prominent in the way fuel is pumped to the engine. Carb-fed engines need pumps that deliver enough fuel to fill the float bowl and nothing else, while modern gasoline engines need both adequate volume AND pressure to ensure both adequate fuel quantities in each cylinder and efficient atomization of the fuel.

Why put the pump in the tank?

Locating a component as critical as the fuel pump in a place where the sun literally “don’t shine” has been a source of frustration for both the tech repairing the system, and the consumer paying the bill. It is buried in there for good reasons, although ease of repair isn’t one of them. The first is inherent to all pump designs: immersion pumps are more efficient, which is why high-end household water well and sump pumps are immersion types. The second is that the liquid fuel surrounding the pump is an excellent cooling medium, simplifying the design of the drive motor, and the third is probably the most compelling if you’re an auto maker: it’s cheaper that way. With the pump integrated into the hanger assembly, much of the fuel system becomes modular and easier to install on the assembly line.

Disadvantages flow mostly to the repair side, with inaccessible pumps demanding a tank drop in most cases. Add the safety issues in fuel handling (the tank drain was also deleted for cost reasons) and vulnerable high-pressure lines from the tank to the fuel rail and there are multiple places where trouble can begin. Short of a leak, however, the fuel pump is still the place to begin a fuel injection diagnostic.

Most modern FI tank-mounted pumps depend on external pressure regulation, according to DeWayne Rector, engineering technician for the Carter Fuel Systems division of Federal Mogul: “In the old days with the mechanical pumps, the fuel pump pressure was limited by the fuel pump, (whose) static pressure was controlled by a spring within the pump. On fuel injected vehicles, the pump does not control that pressure. It’s the regulator on the rail that maintains system pressure. The only control of maximum pressure that the pump has is by a relief valve within the pump. Higher-pressure pumps have that relief valve just in case there’s a blockage in he line. That keeps the pump from going to its maximum capable pressure, which may be enough to blow a line.”

Failures usually give little warning

Injection fuel pumps do last a long time, but when they fail, the vehicle owner rarely reports any warning of the impending failure. Hard starting and hesitation may be noticed over the last hundred miles or so of pump life, but most technicians will encounter pump failures as the vehicles arrive “on the hook”. Interrogating the owner will rarely point at the pump directly, but fortunately there are indirect ways to test pump performance. Can hard start and drivability problems be traceable to the pump? It’s possible, and good tools for checking pump performance aren’t limited to gauges and a voltmeter. Dewayne Rector describes the procedure: “When the pump nears failure it’s possible to see it on an oscilloscope. As the commutator passes the brushes, there’s a current change; it looks like a sine wave form. If that sine wave form has flat spots or spikes, that can be a sign of a pump that’s about to fail. You can determine that the pump is “on the edge”. You could take the reading just about anywhere between the fuse block and the pump. We do it with a precision clamp-on current probe.”

The average shop may or may not have a clamp-on probe attachment for the shop ‘scope, but precision is important when using this diagnostic technique. That’s because the current fluctuations will typically be in the range of plus or minus one-half to one amp peak-to-peak inside a ten to 12 amp supply current. Without a precision clamp-on probe it might be possible to break into the system to series-connect an ammeter, but few instruments have the resolution to detect the small current fluctuations.

Voltage is much easier to ‘scope and can be picked up almost anywhere in the pump’s electrical feed between the fuse block and the pump. According to Carter Fuel System’s Rector, the voltage trace can reveal a weak pump: “It can help if the brushes are worn, and there’s very little contact force between the brushes and the commutator. The brushes can bounce off the comm, causing voltage spikes. You can see carbon tracking on the commutator too.”

Is ‘scoping the feed worthwhile before a pressure test? Whether it’s voltage or current (with the right probe) the hook-up is fast and will validate both the pump’s electrical feed and the quality of the load.

Modern fuel injection has changed pump service from two bolts and two fittings to a serious procedure involving the toxicity and safety issues around handling significant fuel quantities. For the foreseeable future, we’re going to have to deal with in-tank pumps; the “hassle factor” alone warrants as much external diagnosis as possible. SSBM