17 replies to this topic

[Ethel]

The siamese A Series is a bit of an oddity. Even before injection entered the tuning scene, it seemed happier in a lot of ways with an SU over more sophisticated carbs. It's a black art, but some have attributed it to the fuel atomisation (or lack of it) giving the inner & outer cylinders a more balanced share.

 

We were recently chewing over another MPI injection topic - waiting for buses etc. 

[IronmanG]

The siamese A Series is a bit of an oddity. Even before injection entered the tuning scene, it seemed happier in a lot of ways with an SU over more sophisticated carbs. It's a black art, but some have attributed it to the fuel atomisation (or lack of it) giving the inner & outer cylinders a more balanced share.

We were recently chewing over another MPI injection topic - waiting for buses etc.

Yes vizard says about the atomisation bit. So injection probably over atomises.

[Steve220]

I'll add my 2p's worth on this as i've been dealing with A series EFI for around 8 years now.

 

The MPi was born by Mike on a smoking break back in the mid-90s due to tighter upcoming EU emission regulations. It has no real performance benefits in the grand scheme over a well tuned carb A series, however it had to get the emissions as low as possible, and the only real way to achieve it was to have fully synchronous injection. To do that, you need a cam sensor. But why? Well, a full combustion cycle of a 4 stroke is 720 degrees, so the piston goes up and down twice. Using a crank sensor alone, the ecu doesn't know which is the compression stroke, and which is the exhaust stroke. The cam spins 360 degrees in a full combustion cycle (1:2 ratio vs the crank). The cam sensor tells the ecu which of the two 360 rotations is the combustion, then uses the teeth on the flywheel reluctor ring to time off its injection and spark.

The beauty of what Mike came up with for the siamese port was instead of just using injection pulse width to control fueling, he could use a 3D map to time WHEN the second injection event happened immediate after the first. So essentially giving the inner port what it wanted, but timed a second squirt of fuel almost immediately after when the outer needed it. What was born, was a way to make sure all 4 cylinders were fuelled correctly. This was a very special bit of coding in the ecu that has been difficult to replicate in an aftermarket application other than using MS3X.

 

So, where have we got in the development of an aftermarket system? Well, we end up back similar to what the SPi has, by making a wet manifold. But instead of using 1 large old skool injector, we use two modern ones, and have better ecu technology than ever before.

 

The SC system is good enough, and on an MPi engine, will pass emissions easily. However the D400 is not the greatest ecu, and the inlet system doesn't have an IACV, which it desperately needs for cold start and idle at differing coolant temperatures. 

 

Bagsport and CMDIY released drive by wire (DBW) kits, followed recently by SC (despite 5 years ago saying it couldn't be done effectively), eliminating the need for an IACV, or a throttle cable. These kits need an ecu that can support DBW control. On both my turbo minis, i have DBW and a Haltech ECU. They start like OEM, pass an MOT emissions test, and drive fantastically. Yes, you will get a bit of discrepancy between fuelling on inner vs outer cylinders, however I have mapped it to be richer than required when on boost to compensate. Felix makes c190bhp, and the van c160bhp. The spark plugs, and the knock sensor tells me a lot of data about what is going on in the cylinders, and the engines seem very happy. Boost makes a massive difference to fuelling, as instead of the engine taking a drink, the turbo is forcing it down its throat, so distribution is better.

 

Happy to answer any questions people may have! Thanks for listing to my TED talk