Sorry, yet another TD5 Starting Thread

[Si Click]

My son has a 1999 Defender 90 TD5 with only 72k on the clock which has recently become a very, very reluctant starter. When it does start it runs beautifully, pulls well and will happily sit for long periods at motorways speeds without any signs of power loss. Apart from the usual slight puff of black smoke under acceleration it burns very cleanly. If air were gradually leaking into the fuel I could understand it not starting after standing overnight, but it will not start if you shut it down and try to start immediately.
We serviced it a few weeks ago including replacing coolant, engine oil, both oil filters, the fuel filter and the air filter - no sign of oil in the water or fuel in the oil. It has a new (month old) 100 AH battery and a new alternator which gives a solid 14V+ charge; we changed the aux belt at the same time. There was oil in the ECU ports, so we replaced the injector harness and cleaned the ECU with contact cleaner and then carb cleaner (which left less residue). I have replaced the air bleed valve and brass connector on the fuel filter - it does not have a non-return valve. My son has replaced the Crank Sensor. We have measured the fuel pressure at the temp sensor at 4 bar and the pump seems to work fine, and yes we have bled the system many times using the 5 throttle pump technique. Suspecting the ECU I have connected it to a Hawkeye Total and there was no sign of a Top Switch Fault. It gives non-deleteable Tacho Drive Open Faults (3066 and 3098). It showed an historic High Fuel Temp Fault which cleared and did not recur. It also gives Fault 3120 - High Speed Crank Synch Loss Logged which can be cleared, but recurs once the engine starts and is revved. I am assuming the Tacho faults relate to it not having a Tacho/ABS. Grateful for thoughts on the High Speed Crank issue, but not sure that this should affect start. The injector balance at idle seems excellent at 0 to 1 rpm, fuel temp sits at around 29 degrees C at idle. The only live reading that seemed odd was an Air Flow of 4.4 Kg/hour with the engine not running. Testing with the engine running showed normal response to throttle. I guess I'm back to suspecting air getting into the fuel most likely through the injector seals, so I will swap out the O rings and copper washers next. Anything we have missed? Grateful for the thoughts of your accumulated wisdom.

 

[johnlad]

Your maf sensor is not working and needs replacing with a genuine LR one, you can unplug the maf and use the stored default settings. But maf shouldn’t stop it starting,
It’s been known for a faulty starter motor to disrupt crank signal which would impact starting, so I’d start with starter motor test/ replacement.
Ignore tacho fault as not relevant to your car

 

[Si Click]

Thanks for that, very useful. I'll order the replacement MAF sensor now.
Given that it appears to be turning over strongly, is there any way to detect a duff starter motor other than replacing it?

 

[wireman]

Check the relays next to the ECU, make sure their connections are clean and tight and make sure the earth there is also good. Mine would sometimes not start and it turned out to be the relay holders had bad connections.

 

[tom1979]

With the engine not running 4.4 is normal. 4.4 or 0 with the engine running would be odd, but a 10P engine isn’t so reliant on the MAF and they run fine with it unplugged. I wouldn’t waste £70 on one just yet, and yours is possibly fine if you are getting 50-60 kg/hr with engine running

Sometimes when a starter motor starts to go it can interfere with the crank signal.

Oil in the ECU plug will continue for a little while as there is still oil in the main loom, just give them a clean every couple of weeks.

 

[Si Click]

I was getting a good 50-60 Kg/hr when running, so the MAF may be fine. What I might do is buy a Bearmach copy at a third the price as a spare, fit it and see if there is any difference. However, this is not going to affect starting.

Next steps are to;
a. Electrical Integrity. Check all the relays under the drivers seat, particularly the earths. I'll check the ECU again for residual oil coming off the harness and give it's wiring a check for any obvious faults. This sounds simple but as my son's 90 has some very comfortable aftermarket leather seats, I need to fully remove the seat first. We have fitted a battery cutoff and remote terminals under the passenger seat to avoid having to take this seat off too often, but I might check the earth integrity under there as well.
b. Starter Motor. Check the voltage drop from the battery on starting. If the starter is worn and taking too much current that could be affecting the ability of the ECU to fire the injectors regardless of any Crank Sensor interference. Assuming the interference results from arcing from worn contacts, hopefully I should be able to sort this with a repair kit rather than a replacement.
c. Change the injector washers and O rings. Could be air in the fuel and this is worth doing in any event.

Thanks everyone

 

[tom1979]

I wouldn’t bother getting a spare maf, and I definitely wouldn’t bother with a non-genuine one

 

[Si Click]

From a starting voltage of around 12.85 v and waiting until the fuel pump and glowplugs had both stopped, battery voltage dropped to 11.5 v with Ignition on and when cranking dropped below 10 v. To me that sounds low and suggests that something somewhere is draining power even before I engage the starter. Before I start pulling non-essential fuses to try to locate the culprit, any suggestions where this might commonly occur, or are these figures normal?

 

[tom1979]

From a starting voltage of around 12.85 v and waiting until the fuel pump and glowplugs had both stopped, battery voltage dropped to 11.5 v with Ignition on and when cranking dropped below 10 v. To me that sounds low and suggests that something somewhere is draining power even before I engage the starter. Before I start pulling non-essential fuses to try to locate the culprit, any suggestions where this might commonly occur, or are these figures normal?

Seems fine. Anything under 10v and the injectors won’t fire

Edit: just read the bit about going below 10v (couldn’t see this on phone). That is too low and you’ll never start it. Is the battery good?

 

[johnlad]

Voltage is too low it’s around 10.5v minimum before ecu will fire injectors,
When you start the car your not waiting for the fuel pump to go silent before starting are you?
Also on live data what’s your cranking speed whilst trying to start it

 

[Si Click]

Yes I am waiting for the fuel pump to go silent. I assumed this was priming the system.
The battery is about a month old, a Yuasa Silver 5000 HSB019 rated at 100 AH 900 A.
With another charged battery attached (not well parked for a jump start), the voltage with Ignition on was 12.1 v and under crank 10.7 v. Still no start.
Crank speed on live data shows 0 going to around 230 rpm for a second every 3 or 4 seconds as she tries to start.

 

[johnlad]

Yes I am waiting for the fuel pump to go silent. I assumed this was priming the system.
The battery is about a month old, a Yuasa Silver 5000 HSB019 rated at 100 AH 900 A.
With another charged battery attached (not well parked for a jump start), the voltage with Ignition on was 12.1 v and under crank 10.7 v. Still no start.
Crank speed on live data shows 0 going to around 230 rpm for a second every 3 or 4 seconds as she tries to start.

Starting should be done immediately after glow plug light goes out, without touching the accelerator.

 

[Si Click]

Everything under the driver's seat appears fine. All the fusible links were properly secured and in good condition, all fuses sound. The ECU had a tiny bit of oil, which was removed with a single squirt of brake cleaner. To get the aftermarket seat out only requires the four bolts holding down the rails to be removed, both my wrists and forearms to be made of rubber and about 2 hours of swearing. If it were mine I would be looking for some XS seats.
There were two wires (1 x brown, 1 x cream) coming from the RHS of the red ECU plug - looking towards the ECU, which had eroded insulationdown to bare metal. I wrapped insulating tape around each and it made no difference.
There is definitely air getting into the fuel as I heard that bubbly bleeding noise when the ignition was first switched on. If the seals arrive I will do the injectors tomorrow.

 

[sierrafery]

Try to bump start it and if it starts easy then replace the starter cos the problem is withing it and it disturbs the crank signal

 

[Si Click]

Thanks, I'll have to wait for the weekend to get help with that. In the meantime I'm going to replace the injector seals anyway.

Question is: if a worn starter is to blame, can I get away with using a repair kit which just swaps out the solenoid plunger and contacts, or do I need to replace the whole unit? Does the interference originate from the solenoid contacts or the motor itself?

 

[sierrafery]

IMO it's from the winding or colector not the solenoid

 

[Wimblowdriver]

the fuel filter - it does not have a non-return valve.

That is probably causing your problem.

 

[sierrafery]

Early D2s dont have a NRV from factory and no problems with starting

 

[Brown]

That 'high speed crank error' suggests to me a crank position sensor (sometimes called a ckp) fault. I had this with a Britpart crank sensor a while ago, and it stopped logging this fault once I had changed it for a more expensive one. I think it was a Lucas but it was a few years ago and I can't remember. Try a known good or famous-name-brand rank sensor and see what happens. Even the more expensive ones are cheaper than a MAF or starter motor. The fact that you got an RPM reading (in around the right neighbourhood) while the motor cranks on the starter suggests to me that the starter motor isn't interfering too badly.

 

[Si Click]

Thanks for the continued support everyone.
The crank position sensor is a new Lucas one (about £25), so should not be the culprit.
Today, between rain showers I swapped out the injector washers and O rings as I am not convinced there is not air in the fuel. None of the washers or O rings were blown so they were not the culprit, but a useful job to have done anyway. No 1 injector came out easily with a 5mm allen key held in some locking pliers. No 2 was not moving at all and I did not want to use too much force and damage it, so I bit the bullet and bought the slide hammer tool. Life suddenly became much easier.
One of the most difficult aspects was cleaning out the injector seat in the head. After some experimentation with towelling and cotton buds, I found the best way to do it was to take a sheet of blue roll, roll it into a tight tube, then leaving 2 cm free on each end, spiral wrap cellotape around it so that it holds its shape and is fairly robust. A squirt of carb cleaner down the hole and on the end and it is ideal for getting into the hole and clearing out any oil and grit. Use the other end for the next injector.
Tomorrow my son and I plan to fit a new Denso starter motor. After reading about the infamous top bolt I would hate to sweat blood taking it out, repair the solenoid, get it back in and find that the problem was down to the windings rather than the contacts and have to do it all again.
If that does not cure it then air in the fuel is about the only remaining diagnosis and it does not explain why the engine does not start when the system is fully bled. With all the system bleeding I have been doing there are signs of small amounts of fuel weeping down the tank from the pump. My assumption has been that this is a tiny amount of fuel expelled during system bleeding. Is this a valid assumption or could the top of the pump be leaking and allowing air into the system?