running, but with a couple issues

[BMW320i]

After I got my toothed wheel and VR sensor mounted where I wanted, or at least close enough that it had a chance of working I wired everything up and sure enough the car wouldn't start. It turns out fuel was the problem and not spark though.

I have the Bosch L-jetronic from an '84 318i and apparently it won't do anything without a certain sort of RPM signal. This signal is the ignition pulse that goes to the coil and also the tachometer. So the only way I could get the engine running with the Mjlj and EDIS was to have the old coil also connected. I tried putting various resistors in place of the coil and it would get the tach working but not the fuel injection. Any ideas on how to get around this? I seem to remember reading that the coil can overheat when it's not connected to any spark plug, and aside from that I've been zapped by it a couple times already

While I was driving around a bit I noticed a few hiccups, like the engine would hesitate and the Mjlj program would momentarily display some absurd RPM reading (143,000RPM woohoo!). Is this more likely to be a VR sensor placement issue? wiring issue? power supply issue? The only source for switched 12V in the engine bay is one 16 gauge wire, is that enough for the EDIS and MJLJ together?

BTW I'm using the v3.2.1 firmware and configurator.

thanks for reading, TJ

[MartinM]

Hi TJ and welcome!

My Peugeot 205 was similar - I left the old coil in place as I also wanted to be able to switch back to conventional if I had a major problem during a competitive event (not that I ever needed to).

I've not heard about coils overheating when not connected to plugs, but I've never researched that.

You could try:

- connecting the L-jetronic sensor wire to the TACH OUT connector on the MJLJ - but if the sensor is detecting the flyback voltage on the coil (instantaneously, 10's/100's of volts), then that may not work, especially given your resistor tests, which would seem to confirm that

- connecting to one of the EDIS coilpack coils (pin 10 or 12 of EDIS) - this would be a good initial test to prove operation, but each EDIS coil is only operating half as fast as a single coil, so you should get half reading on the rev counter and the fuelling will be for half the rpm it is running at

- using the "Tachometers that require a high-voltage kickback pulse" circuit on
http://www.autosportlabs.net/MJLJ_V3_ve ... it_control
with (or probably without) the Zener diode


143,000 rpm is symptomatic of noise in the system somewhere. Standard techniques are, in order of most likelihood of fixing the problem:
- shielded wire for the PIP and SAW connections, earthed at EDIS end only
- shielded wire for the VR sensor to EDIS connection, earthed at EDIS end only
- running the PIP/SAW and VR sensor wires nowhere near the coil/plug leads
- having the MJLJ unit nowhere near the coil/plug leads
- triple checking all earth connections for soundness
- different and "cleanest possible" 12v connection to the MJLJ

Try some of these and report back...it will work fantastically when you've got it sorted, so don't give up!

Regards
Martin

[Gilesy998]

The latest firmware and controller should help too - V3.2.1 has the adjustable fucntion to "ignore" any noise-generated signals.
Is the VR bracket rock-solid in position? A bracket that wasn't quite rigid would cause the same problems as you describe.

[MartinM]

TJ says he is running 3.2.1, but I agree that VR sensor mounting is also important (both in position and rigidity) and can give these symptoms so it's worth checking that as well...

[Gilesy998]

TJ says he is running 3.2.1

*Must read properly*

[brentp]

Hi TJ,

How does the engine run without the MJLJ connected? with the MJLJ disconnected the EDIS module will run in 'limp home mode' with ignition set at 10 degrees BTDC.

Agree on the previous comments- you should source power and ground for the EDIS and MJLJ from a good source- preferrably connected to the same spot. The shield on the PIP/SAW cable as well as the VR sensor should be grounded to the same spot where MJLJ/EDIS are grounded.

16 gauge wire is a little suspect- if the MJLJ/EDIS are current-starved, the power supplies for the rest of the circuits will fluctuate as the coils driven. since the EDIS module is driving the coils, it's the device that needs the extra current, so make sure it it's connected to something beefy, like a 15A circuit in the fusebox. The MJLJ needs very little current, but it shouldn't be 'starved' by the EDIS module by residing at the end of that 16 GA wire.

Agree too with Gilesy- if your engine instability is in a particular RPM range, your bracket could be vibrating excessively at it's harmonic frequency. Running the engine without the MJLJ connected would be an indicator of this.

Let us know what you find out!

[BMW320i]

I had a couple of wires melted to my exhaust manifold... that may have been the cause of the occasional glitches.

The bracket for the sensor is pretty solid, I'm just a little concerned about the distance from the wheel. I'm going to go to the electronics store and get some diodes and try the suggested circuit for getting an ignition pulse from the coil packs. I'll get some shielded cable and a relay too in case I have any more problems.

Thanks for your input! BTW this car has never idled and run so smoothly and I've only just begun tuning.

[BMW320i]

OK I still can't get it to run right without the vestigal distributor and coil.

I tried connecting the two signals to the EDIS coil pack, through the diodes, to my tach and L-jet RPM input. Car doesn't start.

I tried connecting the tach output from Mjlj to the tach/L-jet and the car will start once or twice but die soon after.

I tried connecting ONE signal to the coil pack to the L-jet, and connecting the tach separately to the Mjlj. The car starts but the L-jet has a fit under load and cuts off the fuel, plus the tach drops out above 2000RPM (even while the reading on the laptop is fine).

The tach output on Mjlj is a 12V (peak-to-peak) square wave right? Is it conceivable that it can't handle the current draw and maybe this could be solved with some kind of amplifier circuit?

BTW I am now using a 20A relay for power.

[brentp]

It seems in order to get past the next hurdle you should understand the waveform emitted by the stock electronics, using an oscilloscope. However, if it's simply sensing the (-) side of the stock coil the diode circuit should, in theory, work.

If you can get your hands on the right equipment you can do some measurement and analysis and move forward. That's what I would certainly be doing!

[Hiperformance71]

Hello to all. I'm new here, sorry for my english, i'm from Italy.

I have about the same problem with a Bosch LE-Jetronic from a Fiat uno turbo 1400cc (or the Lancia Thema Turbo). I have mounted a MJLJ in a Fiat uno 1100cc fire engine converted to turbo (originally it has this injection sistem and a single coil and the distributor, the car is vary fast but we will find the way to go much better using the MJLJ).

After mounting the MJLJ in the car, it started ok, the idle is ok. but when we accellerate it, the engine goes like a rev-limiter or sonthing like that at about 3000 rpms, the rev-limiter in the MJLJ is set to the 8000 rpms. When the car is alimented with gas (LPG or GPL here) this problem disappear, the engine respond fine how espected. The LE-jetronic take the rpm signal from the coil. Because now we have 2 coils i had made the circuit described with 2 diodes and a zener, i have used a 18v zener. After this problem, i have tried too to use the circuit without the zener, the engine start but dies immediatelly. The same occurs when i had used the MJLJ tach out. Now i think to try some zener, from 10-12v to over 18v but i dont know what is best and i'm not an electronic expert.

Some ideas?? ideas are wellcome!!

Best regards
Antonio

[BMW320i]

I haven't got everything figured out here yet but there is one useful piece of information I remembered. On my car there is a relay that controls the power supply to the fuel pump(s) and this relay depends on the ignition pulse also. Maybe the tach signal satisfies the L-jetronic itself but not the relay? It's worth checking out.

[Hiperformance71]

Hello,

Tanks for the idea, effectivelly, this type of injection uses a "Tachimetric" relay. Well from the first time i have tried to start the car i have had the pump relay working, but i dont know if this work properly, i will made some tests tomorrow and document the results.


P.D. only few minutes ago i have discovered a good data about this type of injection. The signal for the rpm and the tachimetric relay is from 0,6 to 2 Volts. This is the answer i think, for the tach is good but not for the injection ECU or the relay, tomorrow i will made a mod to lower the output to 2v or lower.


antonio

[BMW320i]

I seemed to have success with this circuit, using the 1n4004 diodes connected the other way
http://www.hyakushiki.net/mcircuit.bmp

Unfortunately my spot welds gave up at 6K RPM and the toothed wheel disappeared across the highway so I'm kind of back to square one

cheers, TJ

[Hiperformance71]

I seemed to have success with this circuit, using the 1n4004 diodes connected the other way
http://www.hyakushiki.net/mcircuit.bmp

Unfortunately my spot welds gave up at 6K RPM and the toothed wheel disappeared across the highway so I'm kind of back to square one

cheers, TJ

Hello to All,

Finally, i have tested the circuit suggested by you, BMW320i, and it WORKS VERY GOOD!! Many thanks!!! The engine now can rev. freely at the redline (with the other circuit, the sistem is revving at max 4000rpm because the signal for the BOSCH LH2-Jetronic is wrong).

Cheers!!