6 replies to this topic

[KiwiMiniGuy]

Has anyone successfully modified an SPI inlet manifold other than removing the crescents in the runners. I sense the is more airflow to be gained just looking at the intake area around the injector.

 

 

[Pops_Guild]

Honestly there is still much "myth and legend" around the SPi inlet manifold, by that I mean there's a bunch of theoretical thoughts but very little (and I mean none written down) actual facts from people who have tried stuff. Certainly not in isolation anyway, people have made a lot of modifications to SPI's but have not attempted to attribute what change does what, they usually do it all and 'it's better'...

 

One day I'll get the time to get some flow testing done. I have an array of modified SPi performance parts including modified inlet manifolds (6 versions). I hope that will help folks...just need more time :)

 

Mathematically (from measurements, volume etc.) removing the crescents would increase the potential air capacity, if the engine required it, but the capacity with the crescents is already mathematically non-restrictive. What I mean by that is that the engine will only ask for more air if it can, but it can do that already without hinderance. Even with head and capacity increases, the inlet manifold with crescents appears to be big enough from measurements for the engine to do that. The crescents also play an important role...

 

It is a wet manifold system, which means fuel 'pools' on the inside of it. The water jacket and the PTC heater are there to improve/lessen the effect, but it still happens. The early SPi manifolds didn't have the crescents and the resulting pooling and flow caused noticeable port robbing issues, and at worst, head gasket failure, so the crescents were added with no loss in performance to improve it. I've worked on a lot of SPi's and when you remove the plugs on those early SPI's it is really obvious that two of the cylinders are lean. It's obvious on standard SPI's and really obvious on performance SPi's (JCG conversions and similar), so I am inclined to believe that the port robbing gets worse as the engine pushes harder. 

 

So, although it flies in the face of the usual carb-mini thinking on inlet manifold tweaks, remove the crescents you'll reintroduce those pooling problems and that could be more detrimental. You need even fuelling across the cylinders. People do claim that those earlier inlet manifolds are better, I do not believe they are :).

 

You can do some mild cleaning up (smoothing) and polishing inside the inlet manifold and that could have a small benefit...I run that sort of manifold and it improved the torque across the entire power range by a smidge, but not hp.

 

The SPi overall is a carefully balanced setup. That MEMS ECU is actually pretty sophisticated, yes its annoying you cant easily over-mod/chip it for uber performance but it basically auto-tunes whatever mods you put on the Mini, all the way up to 85-90hp (from 52/62hp) and if your local motoring regulations require it, deals well with emissions all that way too, assuming the engine is in good health. In that sense its easier to enjoy engine mods with a SPi, you don't need to get it setup on a rolling road etc.

 

You can put a bigger throttle body on, you can put a bigger injector in, you can tweak fuel pressure, and there are other flow related mods too (the Japanese have done a bunch of playing in this area), but the ECU was designed with and hardcoded to standard versions of all of these.  It will cope with those sorts of changes but from my own experience so far, imperfectly. The ECU is not built to maximise power but is instead designed to be as fuel efficient as possible, you cannot trick it. It will give you more power but it wont know how to deal with the extra fuel and flow over some parts of the power range. What I mean by that that if you do those changes, expect flat spots in the power curve, the idle could be very rich (Lambda sensor damagingly rich!) and expect a dramatic drop in mpg.

 

I have no idea what other mods you've tried or have but my own honest opinion (until I test things for real) is that effort is better spent elsewhere...there's more bang for buck in changing other things.

 

Have fun! :)

[sonscar]

Top answer!The ECU uses tables to look up what to do based on what it originally was mapped on.There can be short and long term fuel trim tables based on llambda info.Steve..

[Pops_Guild]

Top answer!The ECU uses tables to look up what to do based on what it originally was mapped on.There can be short and long term fuel trim tables based on llambda info.Steve..

 

Ta! You have to try :)

 

I think there are 9 maps over 12 rpm ranges. The standard SPI and Cooper SPI seem to have different ECU maps as well. Replacing a standard ECU with the correct matched Cooper ECU definitely adds hp.

[Wazzah]

Pooling doesn't cause port robbing.

Port robbing causes pooling.

Siamese head design and a 1-3-4-2 firing order causes port robbing.

Cylinder 2 takes it's inlet charge immediately before cylinder 1's intake charge as they share the same port and follow each other in a 2-1-3-4-2-1 etc firing order.

Cylinder 3 takes it's inlet charge immediately before cylinder 4's intake charge as they share the same port and follow each other in a 3-4-2-1-3-4 etc firing order.

It is this closeness in the firing order and the shared port that robs 1 & 4 line of air and fuel.

The crescents are placed on the outside of the inlet manifold and you would think that they interrupt the flow to 1 and 4  and only add to the port robbing.

In fact the "diversion" of the air causes a low pressure area immediately behind the crescent and actually pulls the air stream back behind the crescent and towards the poorly performing cylinders 1 & 4.

It is certain that this bending of the air and fuel intake charge alleviated port robbing of 1 and 4 somewhat and it can certainly be argued that bending the air away from 2 & 3 can rob them of intake charge.

The mini engineers did go further with the MPI with injectors aimed directly into the ports as a further attempt to cure this robbing.

Even then they had port robbing problems and had to run the injectors at the high end of their duty cycle to cope.

I have tried 4 inlet manifold variations and 4 throttle body variations on my SPI engine when it was standard and now since the rebuild. 

SC Throttle Body with their supplied manifold. 

Standard Rover Throttle Body modified as a side draught on SC manifold.

45mm Throttle Body modified as a side draught on SC manifold.

Standard Rover Throttle Body on standard manifold.

45mm Throttle Body on standard manifold.

Standard Rover Throttle Body on standard manifold with 50% crescent's removed.

45mm Throttle Body on standard manifold with 50% crescent's removed.

Standard Rover Throttle Body on standard manifold with 100% crescent's removed.

45mm Throttle Body on standard manifold with 100% crescent's removed.

I've tried different injector sizes and adjustable fuel pressure regulators and set and checked AFR's with a separate wideband O2 sensor and gauge.

I have not performed any dyno/ rolling road testing but can report there is not much noticeable difference in any of these configurations.

I am currently running the 45mm Throttle Body on standard manifold with 100% crescent's removed and the standard injector.

This seems to be the "torquiest" configuration using the arse dyno.

My car goes very well and the plugs are the closest in colour that I have achieved.

It is my firm belief that there is not a hell of a lot to be gained in this area.

 

 

 

 

 

.

[sonscar]

Is not the MPI fully sequential and the injectors are large and run at high duty cycles as they only have a small window to inject the fuel into the cylinder before its partner starts its inlet cycle.This is based on research I did years ago on another siamese port motor EFI install.Remarkable achievement only using two injectors.I do not know but I assume it was done for emission reasons.

Porting and cylinder head mods must alter the MAP reading so for a given RPM you are injecting more or less fuel?The llambda is only capable of too  rich or too lean and the trims may be small?This is based on tuning my standalone ECU equipped cars.On the siamesed port fitting a new exhaust meant revisiting the fuel table as the trims were maxed out.Arent carbs wonderful,Steve..

[coopertaz]

did try myself with removing the crescents and found drop in power so had to get new manifold and just removed casting marks and smoothed corners again no real gain.definately more to be gained elsewhere.