I've been trying to solve this problem for a few weeks now, with no results whatsoever. Hopefully someone else has a better idea...

Basically i've completed fitting magna EFI to my old pajero. The little project has come along slowly, but well. Everything is running now, but I have one interesting problem... No timing advance from the ECU!

I've been over every sensor, the wiring loom and done a self diagnostic from the ECU, and I can't find a problem. I've been told by someone who specializes in ECU's that the magna ECU needs only two inputs (AFM and Coolant temp) to run with advance. I have actually replaced the coolant temp sensor, and checked the AFM as best I can, no problems there.

We've actually cleaned the ECU board, and replaced one of the capacitors that is known to cause problems on the board.

Interestingly, if I go through and unplug sensors, the AFM causes the revs to drop, but the coolant temp sensor seems to do nothing. It does however produce an error if it is unplugged. Continuity has been checked on ECU and sensor end of the wiring loom, and the voltages / resistances seem to match up as they should.

For obvious reasons, i'm reluctant to replace the ECU or AFM just yet. Although I have swapped the AFM with my spare. Although I cannot confirm that either work properly.

Any suggestions would be appreciated,

Cheers
Mick.

pretty sure the ignition timing advance is mechanical. if u take the distributor cap off, you should be able to turn the rotor with your hand and then it should spring back into its original place. This is spring loaded. and with the inertia of spinning faster as the revs rise it over-comes the preload of the spring and advances the timing.

Nope, double checked this. I don't know about the first gen, but second gen magnas the timing is controlled by the ECU. Ignition curve is programmed out of the factory and varies based on auto or manual. The ecu can also make adjustments based on inputs from sensors, such as the knock sensor. This can be proven by earthing the "Timing check connector" which will cause the ECU to default to base timing which is 5deg BTDC. Although in normal running conditions the ECU shouldn't run this retarded.

Timing advance uses the hall sensor (efi models). Check it... its a common magna problem for the sensor and/or loom & plug to fail.

I've checked the outputs from the dizzy TDC and Hall sensor, and I can't fault it. Pulses on one output, and 1 on the other (On cranking) No error from the ECU either. Perhaps this is a job for a mitsi dealer or good EFI mechanic. Although i'm sure they'll look at it in horror, not having seen this before.

Have you checked to make sure the timing check connector is not shorted to ground. This wire is used to set the base timing. If it is earthed it will prevent the timing from advancing. It is a white and yellow wire on pin 1 on the largest connector on the ECU.

Yep, i'm aware of that wire. No it's not shorted out. Used it to set the timing initially, and have also used it to check voltages. The symptom is exactly like what would happen if it were shorted, but it's not. There's no damage there which might cause that either. Although I haven't checked continuity on that wire to earth... ECU internal short maybe??? Will check that tomorrow and report back...

Thanks for the help! Much appreciated.

Mick.

Nope... just checked it with the multimeter... No continuity to earth. Guess that rules that one out too! :cry:

I'm in the process of trying to source a known good ECU to try, that would rule one thing out at least. Although the electronics engineer at work can't see any faults (under the scope) on the current board, Although spending some dime doing diagnostics would reveal the truth there. Problem is I don't want to spend money on a replacement ECU if I'm not sure... recond one is around $400 mark, a little cheaper because I know people, but still alot of money for trial and error. (Hint hint anyone local?? Eltham) isn't optimism great?

Unforuntately i'm not in the position of knowing how the donor vehicle was running, although it was "running" and had beel much unloved.

By the way, my old man has a TR with over 400,000km's on the clock which he's had since new and it looks / runs like brand new. I'm not kidding. Had a head overhaul at 300,000kms and no other major work done. Shows how robust they are if you look after them.

there is also a pin (pin 10) on the diagnostic connector that cuts timing advance and puts the ECU into tuning mode when it is earthed.

After a good look at the wiring diagrams, it seems there are two wires which connect to the diagnostics connector from the ECU. One of them is the yellow self diag wire, the other is a grey and red wire which i'm pretty sure is for an engine analyser to talk to the ECU. I don't know the pin locations because i don't have the Diag connector, just wires straight from the ECU. Perhaps one of these should be earthed. The wiring diagram does not however show this, although the connector is earthed, but I would have though this would be so the engine analyser is at the same potential to the vehicles earth.

Correct me of course if i'm wrong, because i'm only making an observation.

Earthing the grey/red wire puts it into diagnostic mode, so make sure that is not earthed either.

Going of the subject abit what's involved in converting a 2.6 Pajero (1992 NH) to EFI?.I got a complete TR motor with EFI , loom, computer and what not.I Know the inlet manifold needs to be cut and welded but what else?

I didn't actually change the engine over. Only the inlet manifold, dizzy, wiring loom and associated running gear. There's a bloke who has done a few conversions to magna engines, although not a pajero. He did however do an L300 4x4 which is pajero chassis anyway.

Marks mitsi projects, you'll find alot of info there if you look around.
http://users.chariot.net.au/~emengee/mitsubishiprojects/content.htm