Topic

Version 0420 is available.
This version adds tuning capabilities for the MEMS1.9.
They have been tested on an ECU in our lab but not on a running car.
Use this feature with extreme caution.



Details of the previous version are here .

Interesting, when I get the 1.8i running again I might give it a try.

I've tried this feature through MEMSDiag and am currently running the car with the Ignition Timing Advance set to the maximum it can be. No sign of pinking and the car definitely SEEMS to have more urgency about it. Fuel economy doesn't seem to be affected, returning about 40mpg in mixed driving, dropping to about 37 when a bit more spirited

God knows what I'm doing to the engine but it seems to be holding together ok.

When I get a chance I'll put the pscan on with this new update and see what that is showing.

hi, really very interesting this tool. I have a Mini MPI, will it be possible to have the tuning adjustment function also for MEMS 2J ??
Thanks

This would be great for me too but I fear that the ECU for the MPI is not tunable , unless the wizard that is Phillip tells me different . I had to get another ECU when our Mini started having emission issues a while back due a poor earthing in the boot ! Photo0054.jpg[/attachment]

John

I have not been able to find a tool (including the factory one) to tune the MEMS2J. The most likely reason for this is that it isn't tunable.

According to this
https://www.sevener.fr/fichiers_articles/Manuel_sur_ECU_Rover_MEMS.pdf
Some MPI engines have a knock sensor and so I guess are "self tuning" to some extent. Is this why they cannot be tuned by a diagnostic tool?

Spent Saturday morning at Andy Bowden's rolling road playing with these settings in pscan

We ran my sprint car, "Project Shed", a 1996 MGF with a 1.8 MPi engine running MEMS1.9. Not quite standard: it has R140 cams, a 52mm throttlebody (because I had one spare), a K&N panel filter, an MGF Centre 4-2-1 exhaust manifold and a Trevor Taylor single exit exhaust.

Advancing the ignition gave an additional 2 bhp, but there was little to be gained from advancing beyond 3 degrees.

The K-series is over-fuelled. By knocking back on the fuelling around 60 units (not sure what those units are) gained another 3 bhp Got the air:fuel ratio closer to the ideal 13:1 but could not get any further without parts of the map running too lean.

Went from this:


To this:


Because the settings only alter the "open loop" look up table and not the "closed loop" map, there is a little kangaroo'ing at very light throttle openings at very low engine rpm in low gear/ low road speed situations. But other than that, everything is gravy The acceleration is much smoother, pick up is very clean, and throttle transitions under all other circumstances are perfect.

Generally I am very pleased and all it has cost me is an hour's rolling road time. And I would recommend that you do use a rolling road if you really want to optimize the power/torque and establish the "safe limits" because both T4 and pscan will allow you to advance the ignition too far and lean off the fuelling way too much. That said, you can probably play with the settings (say to +2 degrees ignition advance, -30 fuel) within a reasonable safety factor with 60+% of the benefit...

Thank you Rob, this is excellent stuff and supports the 'subjective' feeling that I got. I will now turn my ignition advance back to +2degrees and play with the fuel trim - not something I had done previously.

I wonder what effect this will have on emissions at MOT time. I may have to take my laptop with me to the test in order to reset to default, or set it to default and then put it back afterwards.

Thanks once again for your efforts (and Phillip's).

Thanks Robin. Regarding MoT emissions, I don't know! I plan to do the same. That said, I suspect that we will be fine for a couple of reasons:
1. The test is done at steady-state rpm, and thus the engine is using the "closed-loop" map; we're altering the "open-loop" map here
2. The changes we're making are modest: the most likely impact of the changes we're making are likely on nitrogen oxides - and these hopefully will be contained by the catalytic converter.

PS, Philip mentioned that one of the tools had the settings the wrong way round? i.e. when you thought you were adding something, it was actually taking it away? Philip was particularly pleased when it appeared that pscan was doing everything correctly, as demonstrated by the rolling road and the broad band Lambda sensor in the tail pipe

yes I did play with another tool. It's not really that it's "the wrong way round", it's more that it's not obvious which way anything goes as it is all based on the raw numbers sent to/from the ECU. The risk of damaging your engine or not getting what you expect is therefore somewhat higher.

Obviously I don't want to debug how to use a competitor tool so enough said about that.

pscan on the other hand has very clear rich/lean and advance/retard buttons and the screen tells you what you have set compared with default.

Also pscan talks to every ECU on your MG Rover (well most of them anyway) and you get some proper customer support too. Just buy a pscan