Ford EDIS units

[tapkaJohnD]

I'm trying to fit Megajolt ignitiion to my Triumph.
But no ignition, none at all.

Using a Picoscope, I've established that the crank sensor is working and sending a signal to the EDIS.
But the EDIS is not putting out a PIP signal to the ECU or Megajolt unit.
In their absence, an EDIS should go into Limp Home mode (LOS) and fire the coil with a a constant 10 degree advance.
The EDIS has a good power supply and earth.

And by earthing the control leads to the coil, I've got those to fire.

So, good signal to EDIS, but no PIP and no LOS, coils work. Faulty EDIS??

JOhn

[eliot]

Is limp mode working or not? You said by earthing the control signals to the coils they fired, which is doesn’t sound like you have eliminated the edis unit itself.

[Penelope Stopit]

Cam Sensor? I don't know what you are working with

[eliot]

Penelope Stopit said:

Cam Sensor? I don't know what you are working with

Edis is is crank sensor only, it’s pretty simple. As long as it gets a valid crank signal it will fire the coils in limp mode - which is 10’ btdc

[Penelope Stopit]

eliot said:

Penelope Stopit said:

Cam Sensor? I don't know what you are working with

Edis is is crank sensor only, it’s pretty simple. As long as it gets a valid crank signal it will fire the coils in limp mode - which is 10’ btdc

Thanks for this, I'm now taking a refresher course as I had it stuck in my head that there was another signal and I thought it was camshaft or a pickup coil around an HT Lead

This is where I am studying http://www.megamanual.com/ms2/EDIS.htm

[MKnight702]

I had an issue with Megajolt when I first fitted it to my car. We tested all the inputs and outputs and they appeared OK. It wasn't until a friend dismantled the unit and had a poke around inside that he found one resistor that had been put in incorrectly, it was the wrong one so everything worked fine it was just that the signal that went to the EDIS was too weak to work. That was a bugger to find, I hope yours is easier.

[tapkaJohnD]

MKnight702 said:

I had an issue with Megajolt when I first fitted it to my car. We tested all the inputs and outputs and they appeared OK. It wasn't until a friend dismantled the unit and had a poke around inside that he found one resistor that had been put in incorrectly, it was the wrong one so everything worked fine it was just that the signal that went to the EDIS was too weak to work. That was a bugger to find, I hope yours is easier.

Thnaks!
I read that the earlier (?) Ford DIS system did use a camshaft postion sensor, but as said, EDIS uses only a crank sensor.

And as also said, EDIS has a limp-home mode, fixed 10 degrees advance, that it should use if no ECU or Megajolt connected. Which there isn't, so it should go into this mode and the state of the Megajolt unit isn't relevant. Yet.

The coil pack does work, IF you ground a control lead. But no PIP, no limp-home mode, no signal to coil-pack. The EDIS is said to be very robust, so a failed one seems unlikely, but I tjinbk I have one.

John

[eliot]

yep edis is solid. The sensor needs to be about the thickness of a business card away from the [36-1] trigger wheel. Have you tried swapping the polarity of the wires coming out of the ve (crank) sensor?

[GreenV8S]

eliot said:

Have you tried swapping the polarity of the wires coming out of the ve (crank) sensor?

That would be my suggestion too.

Also confirm the crank sensor wires are shielded and the shield is grounded at one end - preferably the EDIS end.

[cinquecento]

PIP goes to the Megajolt from the EDIS module, MJ returns a SAW. MJ also sees EDIS as a distributor..so signal in (PIP) and x millisconds later, SAW out.
Connect a laptop to MJ and see if you are seeing a RPM signal..if yes the MJ is seeing PIP..

[tapkaJohnD]

Swap the polarity? That's an idea. Bought the kit ready made ages ago, only just now got to fit it.

Connecting the MegaJolt unit would seem to complicate matters - the EDIS should run without it, and try to send the PIP. Why would putting the Mj in place be better?

Thnaks for thinking fr me!
JOhn

[Sardonicus]

GreenV8S said:

eliot said:

Have you tried swapping the polarity of the wires coming out of the ve (crank) sensor?

That would be my suggestion too.

Also confirm the crank sensor wires are shielded and the shield is grounded at one end - preferably the EDIS end.

This ^ one of the biggest causes of no trigger especially EDIS, which just to add in my career I have never witnessed a failed EDIS 4/6 module ever and its use on Ford are right through the middle of my time from early to later 90's they are rock solid if a little restrictive e.g no ignition cut for rev limitation etc only the hard coded one, I fear that RPM threshold may be a little high for your Triumph if your a thrasher but would be fine under normal use

Edited by Sardonicus on Monday 12th March 09:50

[GreenV8S]

One advantage of connecting it to the MJ would be that you can use tooth logging to see whether it is getting a sensible and consistent signal. The 'scope shows whether it's what you expect - the Mj shows whether it's what the Mj expects. Of the two, the Mj is more likely to be correct.

[Sardonicus]

Alternatively just look for a tacho pulse from the IDM pin on the module you can even scope from this too this also doubles up as the Ford diagnostic output when they was current issue

[tapkaJohnD]

Sardonicus, thank you for your long experience, that confirms my impression that EDIS units are unbreakable - so why is mine not working.

Lead swap on sensor ASAP? Would that reverse the direction of the stroke through the missing tooth on the oscilloscope? I borrowed a Picoscope and got this trace when I mounted the sensor and a spare trigger wheel on a lathe.



This shows that part of the trace sloping downwards from left to right.

But when I fitted the sensor back on the car, I got this trace on cranking:



And the slope is UPwards from left to right! Could be the clue I've missed!

John

[eliot]

You just reminded me about using a lathe and proof that limp mode works:


http://www.mez.co.uk/ms14-new.html

[GreenV8S]

tapkaJohnD said:

Lead swap on sensor ASAP? Would that reverse the direction of the stroke through the missing tooth on the oscilloscope?

On Megasquirt, the pulse detection logic is not symmetrical and the missing tooth detection fails completely if the sensor polarity is reversed. I wouldn't be surprised if the EDIS was similar.

It isn't causing the current problem, but that cyclic variation in the signal strength shows that the trigger wheel isn't centered properly. It also shows where the high spot is relative to the missing tooth.

[HughG]

When’s I first setup the MJ on my previous kit car I had 2 issues
- the bracket holding the crank position sensor wasn’t stiff enough and caused the occasional loss of signal
- the shielding on the cable wasn’t good enough, changed for some better cable sorted it.

[Mr2Mike]

GreenV8S said:

One advantage of connecting it to the MJ would be that you can use tooth logging to see whether it is getting a sensible and consistent signal. The 'scope shows whether it's what you expect - the Mj shows whether it's what the Mj expects. Of the two, the Mj is more likely to be correct.

You can't perform tooth logging with the EDIS based MJ, the PIP signal is the only thing being sent to the MJ and you get just two pulses per revolution on the four cylinder version. Irrespective, the MJ will do nothing at all without the PIP signal which the OP said is not present.

VR sensor polarity is very important. The EDIS detects a tooth on the falling edge from the sensor, so the missing tooth must be centred between two falling edges i.e. on a rising edge. This suggests the signal measured on the OP's car is correct, provided he has the scope connected appropriately!