14.8V is high. Should be 14.4V just after start. It could be the volt sense connection via a fuse or fusable link to one of the battery + terminals. Sounds more like this than a faulty alternator but needs investigating before it cooks the batteries.
Shayne’s point of priming just before starting is a useful one if the primer pump is intact. It sounds like there may be fuel leaking back to the tank overnight by air getting in somewhere. It’s not always a leak that lets fuel out.
Such a leak can have quite an effect as Shayne found out after changing his fuel filter on the way to Romania, and leaving it over night forgetting to prime it up. It allowed the fuel to leak back to the tank so much that it would not start and we had to push fuel through and up to the primer by pressurising the tank with air from my tyre compressor.
With code 23 showing its defo worth investigation. Might not be the sensor but could be a poor connection or corrosion, nick in a wire (or a puncture from a needle probe) letting moisture in. Test the resistance of the sensor against a known good one before condemning it.
Yes, I was being a fool
Checked my scribble and indeed it did say 14.8 - but the MultiMeter clearly says 14.2 (I checked it again).
I'd like to say I have a good reason for getting that wrong but all I've got is.... daftness.
Haven't done many miles over the break - just kinda switched off for a few days - but back to it now.
Problem persists but still running just lovely once started.
Not found a leak yet and hoses all seem well fitted. Priming the fuel filter still makes no difference (already pretty hard) ... so I'm going to fill up with new fuel at the end of the week (just one more thing I can illiminate) - then move on to all the other bits suggested.
Thanks a lot again for your help - hope you are all keeping well - I will update this when I have anything of value to add
Yep, as per Karl, and as long as that primer stays hard, cross leak and air off.
Thanks for update, all helps in the end for our greater knowledge, rather than as some do, just disappear and no one is none the wiser if/how fixed. Good luck with it.
Update time !!! I haven't forgotten to update the post - I just haven't been able to do anything about the problem. Time, money, blah blah blah.
Recap. Since December I have needed 3 attempts to start the LC followed by 5-10 minutes of being limited to 10mph - before a restart makes things good to go (with a pretty little orange comfort indicator on the dash of course). After those 10 minutes, all is good. Runs as well as an 03 with 130k on the clock can. Starts like it should on the button for the rest of the day - even if left for 8-10 hours. Seems it needs a solid 12 hours sitting to be a problem.
As per some of the posts here (thanks for the part number @karl2000 ), the Air Temp Sensor is indeed listed in the workshop manual as one of the suspects for the reported fault code. However, they are £140 quid. Hence the wait.
A couple of days ago I decided I had no choice, bit the bullet and ordered one. I then felt the panic and decided I had nothing to lose by seeing if I could work out if it's actually broken. Here was my thinking...
- I've got a new one on order - what can go wrong?!
- I don't know how it works - but it can only be physics.
- Remove engine cover. Unclip wire connector.
- Use voltmeter to check angry pixies are making it to the clip - assume that the measured "just under 5 volts" is ok
- Remove Air Temp Sensor. Look at it. Smell it. Poke it. Zoom in on it with phone zoom magic. Ponder.
- Note how much oil/diesel residue it has on it.
- Consider that it must be some kind of resistor with air temperature changing the flow of magic pixies through it.
- Remember the magic pixies in electricals like Isopropyl alcohol.
- Pour the pixes an Isopropyl drink. Dunk the sensor in said drink. Wiggle and woggle it.
- Reinstall sensor.
- Wait patiently for the morning to come to see what happens on a cold start.
- Just about starts on the first try.
- 5 minutes of being limited to 10mph.
- This morning it also started first time - but with a similar (though shorter) 10mph restriction.
So that's where I'm at. Picking up the new sensor tomorrow.
I did think about measuring the resistance across the sensor connections while I make the sensor hot and cold (hair dryer and ice bucket maybe). Physics suggests the reading should change if it's working - but without having a clue what reading I would be looking for across what temperature ranges (that I couldn't accurately measure) I'm wasting my time.
Decided I'm just going to risk the money. It only takes 5 minutes to fit. A garage would probably charge me that in diagnosis fees alone just to discover it's not the sensor - that I still would have had to pay for anyway.
Bought and fitted - but as expected, not the issue.
Also checked the magic pixie feed to the connector terminals again to be sure. Just under 5 volts.
There's this niggling fact that although it might take 2 or 3 long key turns to start, if I then turn it straight off, it starts again prefectly... all day. It might restrict the output to 10mph for a while (only the first few minutes - likely trigerred by the starting fault?), but if I off-and-on-it, even immediately after the fist start, it's perfect.
That fact makes it feel like it isn't a direct temperature issue of any kind - otherwise it would struggle to start all the time until warm - not just the first start.
Still feels like a fueling issue... but the primer stays primed (hard to push) over night.
Do you still get the check engine light? Might be worth trying to plug it into techstream (or equiv) to check the codes again now you've replaced that sensor. Try and get a full Pxxx or Cxxx type code rather than just the number if you can. Also if you can get a video of the live data when its acting up (10mph) that might give us some more clues as to what the problem is....
Doesn't sound like it can be much as the car works OK once you get it going!
If you still get code 23 (or equivalent) it could be the wiring between the (new) sensor and the ECU. The steps in the repair manual for code 23 should help narrow it down to the actual problem. Its easier if you have access to a tester or techstream.