W124 230TE starting problems

[skwerl]

124.083 engine. Has recently failed to start in the morning. Crank the engine and it fires briefly (~ 1 second) but cuts out. Does this 2 or 3 times then doesn't even fire. So I thought fuel problem. The kicker is, if you leave it 20 mins it fires first time and runs perfectly.
I've yet to start poking around because there's no prevailing fault to find. If it just wouldn't start at all I could work my way around the usual diagnostic options.
Are there any components that are most likely to be at fault?
Had a new fuel filter July 2016 (20k miles ago). I'm ruling out the ignition side of things. I'm guessing intermittent issue with fuel pump or relay - I imagine both are original, 32 years old. The car has very detailed history and there's no mention of either being replaced

 

[skwerl]

Update - I pulled the fuel pump relay. Not sure what I was expecting to see. It's clean, no corrosion etc. Put it back and now the pump isn't coming on at all. So, most likely the relay but I can't work out why leaving it for 20 mins always results in starting/running as normal.
Going to replace relay as step 1

 

[219]

If you open up the relay , chances are you’ll see bad solder joints on the tracks of the circuit board , either where the solder has broken away completely around the leg of a component, making no contact at all , or has become ‘dry’ where it has crystallised to give no electrical contact .

The cure in either case is to go over the board with a soldering iron and ‘re flow’ the solder , just adding a little fresh solder as you go to form a heat bridge between the tip of the bolt and the track .

This is quite a common fault as cars with a lot of electronics start to age .

 

[skwerl]

If you open up the relay , chances are you’ll see bad solder joints on the tracks of the circuit board , either where the solder has broken away completely around the leg of a component, making no contact at all , or has become ‘dry’ where it has crystallised to give no electrical contact .

The cure in either case is to go over the board with a soldering iron and ‘re flow’ the solder , just adding a little fresh solder as you go to form a heat bridge between the tip of the bolt and the track .

This is quite a common fault as cars with a lot of electronics start to age .

Thanks. While cracking it open and re-soldering is worth a shot I'm not sure I'd trust it to remain dust and water tight so will replace. Guess I should have tried your suggestion before I ordered a new one but I'm reasonably confident it's the relay - no harm (other than the £106 cost) in replacing such a critical component.
I'll pull the old one apart once things are fixed.
The fact that it starts/runs fine after waiting a while is plain weird though. It may be heat expansion/contraction under current flow moving/loosening pins/solder etc.

 

[219]

I’ve opened up a few in the past and fixed them successfully.

The covers do snap back on and seal well , but if you want to be sure just run a thin bead of silicon sealant round the base before snapping the lid back on .

 

[skwerl]

I just pulled the relay again and it looks new. Went back through the history and sure enough it had a new relay in 2012 - about 30k miles ago.
So I guess fuel pump is next target for testing.

 

[219]

Check the electrical connections to the fuel pump as they can corrode and become unreliable.

Also , on that subject , check all the fuses in the fuse box . The torpedo type fuses used in W124 and earlier cars are prone to corroding at the ends resulting in a high resistance joint , which then gets hot and burns away a little circle at each end of the fuse .

Whenever I buy a new to me car with these fuses I routinely replace all of them , cleaning up the fusebox connectors as I go with a wire brush .

This video illustrates the point

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

The other possible culprit might be the ignition switch .

 

[grober]

Vintage mechanics are a bit like geriatricians- their "patients" may suffer from multiple problems- Taken individually they can be tolerated but together may add to an effect that's greater than individually. Sounds as if you don't have initial fuel pressure [ dropped overnight/standing]- by the time its built up battery voltage has dropped causing a FUEL PUMP RELAY/ PUMP / ELECTRONIC problem- these old electrical components may be "marginal "wrt battery voltage. Leave it a while for battery voltage to recover from cranking and not so long that fuel pressure has dropped again and off you go?

 

[optimusprime]

I would say the battery is flat give it a full charge up and try again... check the terminals of the battery for corrosion . And the relay will only start the fuel pump for a few seconds only .

 

[skwerl]

Definitely not the battery. It's turning over fine and that takes way more current than a fuel pump.
Pontoneer - terminals were quite corroded when I went to hot-wire the pump. The pump was getting power before as I could hear it kicking on. I cleaned them up the terminals. The failure of the relay, after re-fitting it, was my own fault. The harness was moving when I plugged it in and wasn't actually seating properly. Having refitted relay and cleaned pump terminals everything is normal but that's also normal behaviour.
Will have to wait overnight to see if it behaves the same way again. Be much easier if something just broke. At least that way it can be identified.

 

[grober]

When you first switch on the ignition---before attempting to start -- do you hear the fuel pump running to pressurise the system?- should do this for a few seconds before becoming silent [stage1 FPR]-----[stage 2 FPR ] running constantly-- will only be initiated when it gets an engine running signal either from the tacho or the EZL ignition unit via its CRANK position sensor on cranking over.

 

[skwerl]

When you first switch on the ignition---before attempting to start -- do you hear the fuel pump running to pressurise the system?- should do this for a few seconds before becoming silent [stage1 FPR]-----[stage 2 FPR ] running constantly-- will only be initiated when it gets an engine running signal either from the tacho or the EZL ignition unit via its CRANK position sensor on cranking over.

Yeah, I do hear the pump after the initial failed starting and the 20 min wait. But, no, first attempt in the morning there's no pump activation hence me focussing on relay or pump as the culprit. Pretty sure it's still something in the fuel supply. Could well have been pump terminals and cleaning them may help. Relay could have been slightly loose, though it seemed well seated when I removed it.
We'll see what happens tomorrow.
Thanks for everyone's input so far

 

[skwerl]

When you first switch on the ignition---before attempting to start -- do you hear the fuel pump running to pressurise the system?- should do this for a few seconds before becoming silent [stage1 FPR]-----[stage 2 FPR ] running constantly-- will only be initiated when it gets an engine running signal either from the tacho or the EZL ignition unit via its CRANK position sensor on cranking over.

How long would you expect the engine to run for if it doesn't get a stage 2 signal?

 

[219]

I’d say just a second or two

 

[skwerl]

I’d say just a second or two

Yeah. That's what I guessed. But the pump doesn't appear to kick in at stage 1 so don't think it's crank sensor etc. Unfortunately I need to wait for it to happen again, to be absolutely sure the pump isn't doing its initial priming.

One thing that I haven't mentioned is it has an Autowatch alarm/immobiliser fitted. It immobilises the starter solenoid and fuel pump via individual, 20A relays. Not sure where it's wired in but I assume/hope the fitter just spliced the 12V feed to the pump into the immobiliser circuit.

At this point everyone will say, "well it's obviously the immobiliser" but I'm reasonably sure it's not. Admittedly, I have no basis for that reasoning. If I have problems tomorrow I'll bypass it, to rule it out

 

[grober]

On my old W124 the two items that were switched/ isolated by the alarm relays were the fuel pump relay and the OVP relay--- NOT the main supply feeds but the relay coils activation circuits. [ reduced current demand]

 

[skwerl]

On my old W124 the two items that were switched/ isolated by the alarm relays were the fuel pump relay and the OVP--- NOT the main supply feeds but the relay coils activation circuits. [ reduced current demand]

I can try and trace the cable to see where they connect but I don't like removing any trim, if I can help it, as usually something breaks. Easy option is to cut the wires out near the immobiliser and splice them, bypassing the immobiliser altogether

 

[skwerl]

Everything ok today. I'm declaring corroded pump terminals as the culprit

 

[optimusprime]

Next time it stops , and wont start , look at the revcounter see if its moving at all.