Car struggles to start when under 40 degrees, at 20 degrees this morning it took about 10 seconds of the starter struggling to fire up, this is after the car sat parked outside for 2 days.

Changed the battery last month, no difference. Current and prior battery are the largest "Gold" (premium) batteries available in Auto Zone.

What should I do/check next?

 

Hi vdpnyc,

The first possibility that comes to mind is that your starting circuit has high resistance due to corrosion on one or more of the battery power connections between the battery and engine bay, and/or the ground/earth connections such as the engine ground strap and other ground points. You don't mention the year, model and engine of your car (always a good idea to add this info to your signature line), but assuming your car is an X308 you can find the appropriate Electrical Guide for your model year at this link:

JagRepair.com - Jaguar Repair Information Resource


The Electrical Guide contains not only schematics of all the circuits in the car, but also lists of components, power connections and grounds and their locations in the vehicle. Here's the power connection diagram for a '98 X308:





And here are the Ground/Earth point locations:




Sorry that the snipped jpegs are lo-rez. I'm attaching high-rez .pdf files of the above diagrams.

I would recommend cleaning all the power connections in the trunk and engine bay (the ones at the interior fuse boxes are worth cleaning too, but are less likely to be as corroded), and also clean the terminals and screws for the engine ground strap and check its resistance with a meter.

Below is a link to a phototutorial on cleaning battery power connections and grounds (the car shown is our '93 XJ40 but the same principles apply):

Welcome to Jag-lovers - Members Photo Viewing Page
Welcome to Jag-lovers - Members Photo Viewing Page


Cheers,

Don

 

Thank you - not sure why my signature isn't showing, it says '02 VDP in it...

If the corrosion-caused resistance were the issue, why would it be so temperature dependent? Is there any way to test for that before cleaning a million contact points? The battery-terminals and trunk-grounds near it are looking like-new...

 

I can't give you a quantitative answer, but I can give a qualitative one. Battery voltage and available cranking amps are inversely proportional to temperature and engine "stiction" and viscous friction are increased when cold. Also, to some extent, mechanical electrical connections can have higher resistance due to being less "tight" when cold, as well as the engine taking more crank turns to start. In other words, all factors are working against you and the more current you draw, the more voltage drop and consequently lower crank speed.

So, you did not really say (and probably don't know) if the cranking speed is the same on your car when cold, opposed to when it is warm, and if it is the firing that is different! A scope or other measurement of the ignition pulses would answer that.
If the crank speed is fine, then I would read the air temperature and the water temperature parameters with an OBD scanner to see if they were "reasonable".

Actually, if it were mine, I admit I would probably just make sure I had a good battery and let it crank!

 

You shouldn't be having problems at 20*F (-20* -- hell yes: see Sparks notes above). What weight oil? 5w30 works best. Check the ground straps, as noted, above. Try a can of fuel system cleaner: SeaFoam, BG44, Gumout Ultra. How old are the plugs? Air filter clean? Clean the MAF sensor and connectors. Clean the throttle plate and bore (carefully). Use an OBD II reader and check for codes and water temperature (to see if sensor is letting the ECU know it's cold out and needs to enrich).

 

At 20 degrees it sounds like it is cranking at a slower pace than normal, but this is an impression, not certain science. What does slower cranking mean?

What ground-points are the most likely candidates to cause this? It's too cold out for me to be under hood checking all of the ones on that chart, brrr....

Ran an OBDII and the only code is, as expected, a pending code about the thermostat not fully closing - a code that gets thrown whenever it's cold since I've swapped thermostat housings. I have a new thermostat but have yet to install it since it is just a dash light and the car still drives perfectly like this. Is that likely to have any effect?Water temp after 5 minutes of idling after 20-degree cold start was 40c, so it's heating up just fine.

Never changed plugs, they might be the originals (130k miles, lol) but the car runs super smooth once started, so an unlikely cause I think.

Oil is new and should be 5w30 (unless oil shop cheated me), air filter was clean last checked some months ago. MAF is not causing any other issues, so not opening that up for inspection. Throttle body is only about 8-10k miles old, so should be clean in there.

 

Well, lets see...
By asking "what does slower cranking mean", I assume you mean, "What causes it?". If that is so, the answer is lower voltage at the starter. Some guy named Kirchhoff said that the battery voltage minus the voltage drops in the wiring (the hot side and the chassis) equals the voltage at the starter. Well, he said something that leads to that anyway! If you are asking what the term means, you got it right-it means the engine rotates slower whilst the starter is engaged.

The most likely problem power points in the starter circuit include:
1) two battery connectors,
2) the point that the battery minus leads connects to the chassis,
3) the underhood "false bulkhead" connectors (the ones the big black cables connect to on either side near the firewall) (very likely candidate!)
4) where the same cables connect to the fuse panels under the hood,
5) the engine to chassis ground strap, which is on the right side near the starter (I think)
You should be able to check all of those (except the engine ground) in the few minutes before you freeze before you freeze to death. Remove one of the battery leads before you start and tighten each connection bolt to about 12 ft-lbs or so. Especially check the false bulkheads visually for corrosion.

The only way OBD knows the thermostat is not closing is that the engine is not coming up to temperature. That is NOT "just fine". The fueling system is calibrated for the temperature ranges the engine is supposed top run at and your fuel efficiency is taking a hit, your drivability cannot be right, you are potentially sludging your engine. It is not too far fetched to think that the engine management might be "winding up" the fuel adaptions with the bad values and contributing to your starting issues. Where did you read the 40 degree water? That is way too low for the engine temperature. Are you reading this with the OBD scanner? The engine temp should be in the 85 C range in spite of the ambient temperature.

A gasoline engine has to perform in a variety of ambient conditions and output requirements. Modern ECUs help to optimize the operation over those excursions, but it is not valid to say that if it runs well in one set of conditions, it will perform the same in others with failing components. That said, I agree it is not LIKELY to be the plugs, except that they can come into play if the system voltage is low while cranking (see paragraph 1).

BTW, the reason the MAF parameters are important is that the ECU uses the incoming air temperature (it is read by the MAF) to determine fueling. So, a problem with improper temperature measurement of the incoming air will most definitely cause poor performance when temperatures change! Also, you discount the water temp issue,but the ECU uses certain backup control strategies that MIGHT make it disregard temperature as a control parameter when a temperature fault is detected.

** In fact the ECU uses default temperatures (122 deg F) when the sensors are giving codes- including startup! I bet that is your root cause.

Frankly, you seem to have a certain chip on your shoulder about all of this. You do not know what the correct engine temp is, but you are convinced you have no problem there. You say your MAF is not causing "other problems", but you don't want to troubleshoot its temperature output parameter with a scanner. You came here asking what to do next, but you don't want to do anything in the cold. Maybe we did not give you enough background so you could understand why ALL of the suggestions you have gotten so far are valid, but they are. (Well, maybe I would argue Jim's Sea Foam, but that is just my opinion!)

So, with no malice at all, my advice is to take it to a Jaguar dealer or good independent mechanic and ask them to fix the thermostat, and then find the cause of your extended crank starting. Or, since other than this, you seem happy with the car's performance, then do what I said previously that I would likly do and live with it! **BTW, I said that before I knew you actually had a code you were ignoring. Any code might be causing the engine to go into a backup control strategy!

.

 

After five minutes of idling and water reaches only 104*F? You need a new tst, probably a new cross pipe with a new sensor. Clean the MAF, change the plugs, tighten the battery cable. Run some fuel system cleaner through. The good news, it still starts.

 

+1 on taking the advice offered. Most looks DIY so is cheap.

 

Actually, due to the report of pretty good drivability and performance other than cold starting, I think the engine temperature is possibly fine and the temperature sensor got buggered in the thermostat replacement mentioned!

 

+ 1 Or not connected

 

Thank you! All good advice (except the bit about having a chip on my shoulder, lol, you are reading too deep into what is written)

JHartz, 5-minutes after a cold start in 20f weather, a 40c coolant temp (read by OBDII) is low? Seems normal to me based on my previous cars - it's bloody cold out, lol.

Sparknzap, very good point about temp-sensor maybe being inoperative, never actually noticed if temp-sensor is even working or just always showing 40c - gonna check that next time I drive.

FYI, I'm anticipating a potential 2,000 miles roadtrip for work and am afraid to mess with the engine prior - mechanics have twice in the past broke those flimsy rigid coolant lines running above the engine block. You can drive fine with a stuck-open tstat, but you won't make 2k miles leaking coolant....

 

Good enough. Try charting the temperature rise as the engine warms up. Which code are you getting?

 

After five minutes, today, with outside 21F, mine was reading 165F and climbing to 185F at eight minutes; held 185 during a twelve minute dash down the freeway to the Oceanfront; rose to 196 idling in traffic (external 28F; IAT 65).

 

Temp sensor seems to be working, today is 40f and within a minute of idling after a cold start the Coolant was reading 50c and rose to 60c in another 2 minutes of idling.

It's been rainy, so have yet to pop the hood...

 

The best way to diagnose the cause of slow cranking is to measure the voltage at the starter, while cranking. The starter uses a lot of current, so there is always some drop, but the amount of drop will tell you there is an issue.

I would look for 9V or better while cranking. 10V would be good. If it is less, you can start trying to find where the drop is.

I made some 10ft long meter leads with big clips for diagnosing this kind of thing (on my boat, but the principle is the same)

In order, I measured (all while cranking)

Across battery while cranking (11V - fine)
At starter while cranking (9V - just OK)
From + battery terminal to + on starter (1.5V)
From - battery terminal to engine ground (0.5V)

So in my case I had quite a loss in the + cable. The problem was bad connections, and bad crimps.

 

That's fascinating! 1.5v loss from bad connections, maybe that's what I got...

You could probably make the same measurements via measuring resistance in the wires, no?

I don't have a voltmeter that is for a car's amperage, but my multimeter has a resistance setting.

 

The problem measuring resistance is that it is very low and your voltage drop, which is the important variable, could only be calculated from the resistance if you know the current (amps) while starting, which, as you pointed out is difficult to measure. For instance, with a 100 amp starting load, the .5 volt drop happens on a conductor with .5 ohms. Touch you DVM leads together and I think you will find you have something like that as the lead resistance. So, practically, a "normal" DVM is only good for resistances above about 3 to 5 ohms.

I think you may be a little confused. The test Mark performed is to read the potential drop (which is generally called "voltage") between one end of the wire to the other end. It is done with the "voltage" setting on the DVM (or multimeter). So the amperage flowing through the tested conductor does not flow through the meter and is not a factor.

Kirchoff's law says the voltage changes around the circuit equal zero. So, in this case, the battery produces say 11 volts under load, wire 1 drops 1.5 volts, wire 2 drops .5 volts, which leaves the starter to drops 9 volts. Each of these values can be measured, under load of course.