3.5 hotwire efi won't start - solved, poor connection to coil -ve!

Re my 3.5 Hotwire efi auto from a '92 Disco1 in a Ninety.

Didn't use her for a few days and yesterday she wouldn't start. Spins over but no attempt at firing. Was starting and running perfectly before.
Fuel pump working, rail pressurised, relays clicking as they should, good spark from king lead and plug leads, checked plugs and they were dry and tan coloured.

but no fire up..............UNLESS

......some quick start is blown into the plenum, then she fires up and runs of her own accord. Switch off and won't restart.

Cleaned, checked and even swapped fuel pump and main relays but makes no difference, disconnected ecu and cleaned the terminals and multiplug. It seems as if the initial start up signal is not enabling the injectors to fire but once running they work fine - disconnected one at a time to hear engine change.

I'm stumped. Dodgy connection between coil to ecu and/or relay/s? AFAIK the first ignition position allows the fuel pump to prime the rail then once the engine is spinning on the second key position the coil informs the ecu which then powers and adjusts the injectors to suit conditions. So something is not working on the initial start up sequence.

Any thoughts received with thanks.

No thoughts anyone?

What should happen is that once the starter motor starts to turn the engine, a 12v pulse is fed back to the ECU from the negative side of the coil and the ECU then turns on the fuel pump and energises the fuel relay that provides a fixed 12 volt supply to all the injectors.

It's as if the fuel is supplied but the injectors won't fire, but once running (with help) the injectors fire!

My thought so far are:

~ As the engine starts and then runs of its own accord after help with a squirt of quick start or a squirt of fuel, is the issue is the signal from the coil to the ecu or the ecu to the injectors not turning them on... but they work once running (or it wouldn't run!!) ???

~ Just a thought but would a faulty water temperature sensor cause a non-start but allow normal running? If it failed and the ecu thinks the temp too high it will lean out the mixture but won't affect running as the ecu would compensate for the actual conditions? Ditto fuel temp sender?

~ something amiss with the wiring from the ignition to the ecu or to the main and fuel pump or to relays and injectors? Where on earth to start?

~ faulty ignition amplifier? They usual play up when hot not cold and cause running issues I didn't have?

Again, once fired up she runs perfectly but just won't fire up normally. Note this started last week, difficult to start one morning (all fine until then), got going, ran ok, stoped and switched off, wouldn't restart (same signs, spins but no sign of a fire up), waited awhile and she started again, got home all ok. Couple of days later refuses to start until battery going flat.

Please Is there anyone on here with V8 EFi knowledge and experience who ay have an idea what's going on? Many thanks in advance.

I keep thinking that if a 12v pulse is sent from the negative side of the coil to the ECU whilst cranking, the ECU energises the fuel relay to give a 12 volt supply to the injectors, so if the ECU isn't getting that signal from the coil the injectors won't work.

It doesn't explain why it fires up and runs when fuel/easy start is squirted into the plenum chamber...doesn't the ECU needs the -ve signal from the coil to stay working or is the signal just for starting? If the latter then perhaps a broken wire/dodgy connection between the coil negative and the ECU?

Calling @discool You're au fait with V8s, any ideas?

Had a pleasant afternoon with the multimeter going through the efi test checks. All values within range apart from the TPS which read higher than expected (12 vs 5 kOhm) but that wouldn't affect start up, and as said runs fine once going. with good throttle response.

As it starts and runs, with a little help from ES/fuel squirted into the plenum, it is only an issue whilst gong through the start up routine. After lots of cranking the plugs are dry suggesting a no fuel event. The ECU won't permit fuelling - aside from the initial priming - nor switch on the injectors, until it gets a signal from the coil -ve indicating that the engine is spinning. This goes to ECU Pin 39 - 'engine speed signal' - which switches on the fuel pump and injectors.

I am not 100% sure but I have been told that this impulse from the coil is needed only for start up, once running it doesn't matter.

So, I tested continuity between the coil -ve and pin 39 and got battery voltage as expected. Resistance between the coil -ve and pin 39 read between 7.3 and 8.3 kOhm. It should be 6.8 kOhm as it has the 6.8 kOhm resistor so not sure why it read high and fluctuating and if that is the problem.

Still stumped!

If you manually opererate the flapper does the fuel pump fire up?

In old racer we would wedge open flapper & fire it up.

Have you still got a fuel pump relay?

cappers said:

Calling @discool You're au fait with V8s, any ideas?

Click to expand...

Although I had a RR 3.5 flapper I didn’t know much about it as I didn’t have any issues and LR service it, but in my quest for knowledge I do have a Haynes Manual for the 3.5 RR I’ll have a look see first at the relevant wiring diagram.

With my 3.9 Disco there’s a 12v supply from the ignition switch to the +coil… also a 12v on to the EFI ECU also to one side of the fuel pump relay, the neg side of the relay is direct from the ECU so a neg is supplied when required by the ECU.
The neg side of the coil links directly to the ignition amplifier.
Thats from memory so maybe wrong.

So my task now is to find the correct diagram in the RR Haynes for you, then cut&past on here. Shouldn't take too long

Ok the 3.9 ECU also has a connection to to the coils neg terminal via an in-line resistor.
Attached is a copy of the 3.5 hot wire diagram. All i can find at the moment.

Hicap phill said:

If you manually opererate the flapper does the fuel pump fire up?

In old racer we would wedge open flapper & fire it up.

Have you still got a fuel pump relay?

Click to expand...

Thank you but it's 14CUX Hotwire Efi not the Flapper.

discool said:

Ok the 3.9 ECU also has a connection to to the coils neg terminal via an in-line resistor.
Attached is a copy of the 3.5 hot wire diagram. All i can find at the moment.

View attachment 320281

Click to expand...

Thanks @discool.

discool said:

Although I had a RR 3.5 flapper I didn’t know much about it as I didn’t have any issues and LR service it, but in my quest for knowledge I do have a Haynes Manual for the 3.5 RR I’ll have a look see first at the relevant wiring diagram.

With my 3.9 Disco there’s a 12v supply from the ignition switch to the +coil… also a 12v on to the EFI ECU also to one side of the fuel pump relay, the neg side of the relay is direct from the ECU so a neg is supplied when required by the ECU.
The neg side of the coil links directly to the ignition amplifier.
Thats from memory so maybe wrong.

So my task now is to find the correct diagram in the RR Haynes for you, then cut&past on here. Shouldn't take too long

Click to expand...

That is better as it is a Hotwire NOT a Flapper! From a 92 3.5 Disco 1; Flappers were on the 3.5 RRC.

All tests on the ecu etc were ok apart from the coil signal. It goes to ecu pin 39 which when tested showed the battery voltage minus about 1v and the resistance was 7.3 kOhm whereas it should be 6.8 kOhm with the in-line resistor in place. Would the increased resistance cause enough of a drop to fail to trigger the ecu? AFAIK once the engine is firing and running the coil signal is redundant, hence it running ok once started but not starting without help?

Tomorrow I will remove the 6.8kOhm in-line resistor and see if that makes any difference. I've read that it is really needed but I'm unsure and why is it there in the first place?! Perhaps the terminals are dirty/corroded.

If it's not that I really am at a loss as everything checks out ok and once started runs perfectly, just won't start via the normal method!

Cheers for the help.

ETA just found this:

"This circuit (WB wire with 6.8k Ohm resistor from coil -ve to ecu pin 39) is the only connection between the ignition system and the fuel injection system. A damaged or loosely connected resistor would likely result in an open circuit, leaving the ECU wondering about engine speed."

"

Opps and so it is. Been a long day.
My disco has two identical resistors which were taped to the coil but only one is used.
There was a post on it years ago so i went out and had a look and found both, apparently there was an issue (may have been engine not running) and the exact resistor was hard to get, but I found one better get one as a spare only to find the was unused one. So i re tapped with white and never togo there again.

Anyway its a quick check with your meter

Thanks @discool

discool said:

Anyway its a quick check with your meter

Click to expand...

The resistance between pin 39 and the coil negative was about 8.5kOhm whereas it should be 6.8kOhm. The voltage was about 1V less than battery level. Don't know if that is enough to cause non-start, but the ecu may not fire if the reading is correct?

Will take a look later today; if I remove it and all works does it need replacing or is it not vital for running, more a safety device to prevent accidental overload of the ecu?

Thank you for the help.

It isn't the in-line resistor

Unwrapped it, the terminals a bit dirty but not corroded and the sockets ok; gave a squirt of electrical cleaner and a rub with emery till shiny. Tested resistance and it still is about 7.5kOhm rather than 6.8kOhm unconnected to the loom. Tried starting with it inline once cleaned - nothing, and then removed it and bridged the wire - nothing....both times engine turns over but doesn't fire!

There could be a dodgy connection further along the loom and to exacerbate things the battery is losing power with all the cranking so it's now on charge. I will try again tomorrow with a fully charged battery.

I could cut into the loom near the ecu plug and splice into the WB wire to pin39 and run a new wire to the coil to bypass the original wiring but don't really want to do that until I know that is the issue. What else it could be is beyond me.

I have found a new independent LR mechanic just a couple of miles away (on a farm) who has extensive LR experience including V8s and Efis and he offered some advice and troubleshooting help over the phone. He said if I can't fix it to get it over to him and he'll take a look.

I'll update when I sort the problem. When forum members contribute to troubleshooting I always like to feedback the cause and solution to help others with a similar problem. Nothing worse than seeing a thread with your issue and not knowing how it was solved because to OP didn't reply!

Final Feedback!

Starting and running again nicely

Everything indicated the coil to ecu link - no fuel/spark on cranking but both there once running - but voltage and resistance measurements seemed ok.

Checking the coolant temp sensor today I moved the coil wire and it fell off the coil connector The piggy-back spade connector on the coil negative for the coil signal and the tacho had broken but the ends still touching under the plastic cover.
When I did the measurements yesterday I read between ecu pin 39 and the coil negative pole itself rather than the wire/connector hence reading ok. One of the first things I did was to check the coil connections but missed the break!!!

Let her warm, took her for a spin and so far AOK. Phew!

Thanks to all for advice and suggestions.

* Edited to add, obviously I put a new piggyback connector on and then she fired up and ran! Funnily I did notice that the rev counter was going haywire when trying to start before but didn't twig the 'connection'!

Well done finding that.

Resistance checks are all fine and dandy, but sometimes a LOAD test finds things like this... or burns out the last remaining strand :-D Isolate each end of the wire and load it up with a giant headlight bulb or something