After reading about my Electronic Fuel Injection project several people have cautioned me about “charge robbing” so I have spent many hours researching this issue and the following are my conclusions. I apologize for this “wall of Words” but here goes.
The cylinder head design of the Austin Healey 100 is over 80 years old and incorporates Siamese ports, a configuration never used in later designs. In this type of head, the forward and the rearward pairs of cylinders each share an individual intake port and the inner pair of cylinders share one exhaust port.
Additionally, in the interests of simplicity, these engines use a firing order of 1-3-4-2 or, expressed differently, 2-1-3-4, with the result that the inlet ports “see” a port flow pattern of “flow, flow, wait, wait, flow, flow, wait, wait’.
This design results in a phenomenon called “charge robbing” wherein a cylinder served by an individual port which draws the incoming air/fuel charge first gets “robbed” of part of this charge by the cylinder which draws its charge immediately thereafter.
Reputedly, this design results in the inner two cylinders (2 &3) receiving a “leaner” intake charge than the outer two cylinders (1 & 4) and we were always told that exhaust valve failure, a frequent occurrence which was invariably confined to cylinders #2 or #3 in these BMC “A” and “B” series engines, was a consequence of this “charge robbing”.
My 55 years of experience has been limited almost exclusively to carburetted engines of this type, both single and twin carburettor versions, and during that time I have regularly used spark plug colour as an indication of the air/fuel ratio being burned in any cylinder. I’m aware that this test is nowhere near as accurate as modern oxygen sensor systems, but these weren’t available at the time however, for me, this test has never indicated that the inner cylinders (2 & 3) consistently run leaner than 1 & 4.
Some decades ago, it was determined that in these engines the gas temperature in the Siamese exhaust port of cylinders 2 & 3 can be as much as 100° C higher than in the outer two ports and the resulting higher operating temperature of the exhaust valves in those inner cylinders is a much more likely explanation for their premature failure.
For “charge robbing” to affect the air/fuel ratio the “charge” must contain fuel droplets. To express this differently if the incoming charge consists entirely of an homogenized gas mixture of air and vaporized fuel, whatever enters either port will have the same air/fuel ratio. This is almost certainly the case with a carbureted engine wherein the fuel is introduced into the very turbulent incoming air stream as microscopic droplets well before that stream reaches the point in the intake port where the port splits, which allows plenty of time for any droplets to entirely vaporize. In such a situation the volume of the charge entering the inner cylinders may be smaller than that entering the outer cylinders, but the air/fuel ratio will be the same for both.
The same situation does not apply in modern multi-port fuel-injected engines where each port has its own injector. To minimize emissions during throttle transitions, an absolute minimum of fuel spray is allowed to hit the port walls, so the fuel is injected into the intake ports just upstream of each cylinder’s intake valve. As the intake valve opens backflow of combustion gases through it finalizes the vaporization of the fuel droplets within the incoming charge. In the very unlikely event that such a port injection system be used in an engine with Siamesed intake ports the smaller volume of air delivered to the inner cylinders would result in the mixture in cylinders 2 & 3 being richer unless some sort of specialized tuning is used to decrease the pulse width for those two cylinders.
So, how does this all play out in the case of the EFI system I’m developing for my 1956 Austin Healey 100?
The fuel injectors in this engine modification are installed inside the SU carburettor bodies upstream of the throttle butterfly and are therefore a significant distance from the point in the Siamesed intake port where the port splits. As throttle transition emissions are of no concern with this vehicle, this position is ideal in that it ensures that the fuel droplets are completely vaporized long before the intake charge reaches the point where the port splits.
For this reason, it is very unlikely that there will be any difference in the air/fuel ratio of the charge received by any of the cylinders.
Based upon this this conclusion I’ve decided to continue with my EFI project without making provisions to enable accurate mixture measurements in the individual exhaust ports.
Write A Comment