Hello everyone,

Last Saturday I took a car for a 1hour drive at high revs (3-3.5k on average) as I haven't driven the car for 3 weeks, so thought I'd give it a good shake. It was behaving no problems, however, next day returning from a 40 minute trip, I noticed the battery light lit up. The jag did manage to take my almost all the way home (just 10 meters of pushing), but the battery was dead.

I charged the battery at home, and on first start up I measured 14.08V between +ve and -ve terminals - whiich was weird to me, as this suggested all is good. However, as soon as I put on load (headlight, foglights, glass heating etc.) it dropped to ~11.5V, and I never saw that 14V again. I cleaned ground leads (actually, the battery to under fuse box bolt snapped off as I tried to remove it, as I didn't detect good continuity, had to ground via another bolt nearby), checked for any loose cables, but nothing out of the ordinary appeared. I measured from the voltage between the cable from behind the alternator that leads to +ve battery terminal, and the ground (both engine block, and the under fuse box ground) and saw essentially the same as the voltage across the battery terminals. Got AA guy to take a look, he basically did the same tests, and said it's the alternator (he tested the battery and said it was good).

I found some posts by Thermo where he suggests looking at the 4 pins from the alternator, but mine has 3 pins + 1 main cable.



Tried measuring these guys with the engine running (was fiddly trying to get a good connection so should try doing it again), but the voltages I got were weird, one of them was ~ 9.75V, the other about 10.7V, and couldn't get a good reading on the last one (was very low voltage). I checked the continuity of all the fuses and they seemed fine. I got a replacement alternator, but don't want to replace it without knowing the cause. I also tried doing some voltage drop tests, but those were idling (need an extra pair of hands to help with keeping the revs at 1.5k rpm), and didn't see more than 0.02V drop between negative and the alternator case.

Maybe someone could suggest what voltage I should expect to see for the 3 pins?

Also, does anyone know if someone on this forums rewired the harness using their home made cables? It seems that this Jag is a bit of a pain to get access to bits, so I imaging re-wiring everything could be a bit of a job...

 

Boletus, let me look and see what I can find for you. From the sounds of things, you got a wire or the alternator a little too warm and now it is having a problem making a good connection.

 

Thanks Thermo, I had a go at following the Workshop manual, particularly the PINPOINT TEST E : THE GENERATOR CHARGE WARNING INDICATOR LAMP IS ON WITH THE ENGINE RUNNING bit (pdf page 2095)

however, part E9: CHECK FOR IGNITION SUPPLY VOLTAGE AT FUSE 82 OF THE PASSENGER COMPARTMENT FUSE BOX
is actually not right for 2008 2.2D, as the only alternator related fuse is actually F36 under the hood (this is confusing, why is this Jag different?). The voltage measured there was basically the same as the voltage between battery terminals (at the time it already dropped to ~12.11).
So it seems to suggest that I need to "REPAIR the circuit between the passenger compartment fuse box and the generator. TEST the system for normal".

By the way, when I measured E7: CHECK IGNITION SUPPLY VOLTAGE TO THE GENERATOR with the key in RUN position (that's the last one before the engine actually starts turning over, right?) I got 3.5V, which is different to when it is running.

 

Boletus, the wires are as follows:

Green with a red stripe: battery voltage
White: From the ECU to control the alternator output
Brown: the common wire between the ECU and the alternator.

With this being said, the check that I would do is slide a pin along the brown wire and the white wires and make sure the connector is made up to power the alternator. Now, start the car. With everything off, you should have a fairly low voltage between the white and brown wires (probably something in the 3-5 VDC range. If you now measure between the brown wire and the chassis, you should get a very low voltage (up to say 0.5 V). Now, with the engine still running, turn on the dash fan and rear defroster. Measure between the brown and white wires. You hsould now see something more than likely in the 8-10 VDC range. Now measure between the green with a red stripe wire and chassis. You should be reading battery voltage. The final check will be to read the voltage across the battery.

This is where we are going to take these measurements and see where the problem lies.
-If the green wire with a red stripe has a lower voltage than the battery voltage, the green with a red stripe wire has a high resistance connection
-If you had 10+ VDC between the brown and white wires with all electrical off, then your alternator is toast as the field circuit is bad
-If you had about 5 VDC between the brown and white wires with all electrical off, but you are not getting the 10-12 VDC wtih a decent electrical load, then:
- If the voltages on the brown wire to chassis were about the same between the no electrical and decent electrical loading, then you have an issue with either the ECU or the green wire with a red stripe has a high resistance in it
- If the voltage on the brown wire to chassis were dramatically different, then your brown wire has a high resistance in it or your ECU is bad.

If your tests already are pointing at the green with a red stripe wire, what I would tell you to do is see how much of that wire you can expose to hand over hand it. You are looking for the insulation to have a brown spot in it (where it is getting hot) or you are looking for where the wire makes a funny sharp bend. This would indicate that the copper wire inside the insulation is bad. This is where a small thermal camera would be handy as the high resistance spot is going to be warm and will show up on a thermal camera.

Another way to check things is to slide the pin up along the side of the green wire with a red stripe and then connect your multimeter across the battery. You are now going to get a wire from a known good power source and run it to near the pin in the green wire with a red stripe (make sure the end does not touch anything metal). Now, start the engine, put a load on the electrical system (dash fan, headlights, rear defroster). Now, while watching the multimeter, touch the extra wire to the pin in the green wire with a red sripe. If you see the voltage jump up to around 13.7 - 14.5 VDC, then you know the wire between that point and the fuse box is bad and you need to figure out a way to correct it. If it doesn't come up, then we need to take a step back and figure out what we did wrong.

Hope this helps you figure out what is going on. Not sure if something like a circuit tracer would help you out. This allows you to inject a signal on a wire and trace it with the wand. Where the noise to the wand stops, that is where your break in the wiring is.

 

Hi Thermo,

Thank you for the thorough reply. It will be interesting to see where this leads (pun inteded). The only thing - I have an "off orange" wire, rather than a green with red stripe. While I manage to find time to test this, do you know the part number for this 3 pin wire? If it's toast, it's probably easiest to replace the whole thing.

 

Jaguar electrical wiring says generator/alternator is at connection db1 with a black 4 way plug.




https://www.google.com/url?sa=t&sour...X4lfAw0yBkYXpy

 

@Thermo so here are my findings:

1) Everything off: Voltage between white and brown: just over 2V
2) Brown to chasis ground: Around 6V! (White to chasis ground about 8.5V).
3) Fan+Rear defrost on. Still about 2V between white and brown. Brown to chasis still 6V.
4) "green plus red stripe" (on closer look it looks more like faded red with a red stripe!) Matches battery voltage to within 0.04V.


I couldn't really see much of the brown wire. Should I try stripping the wire harness (anyone know anyone who has done this in the past?) Or is there a way to test this further (perhaps look into ECU?). To be honest, I didn't quite follow what you meant by :

Did you mean like a "bypass". So would be best to bypass as close to the fuse as possible, and then as close to the alternator connectors. Hopefully the colour coding is consistent throughout the harness haha!

For completenes: to perform you suggested tests, I had the following set up (shown unplugged).

 

Boletus, we need to try and do these checks with the wiring together as much as possible. What I have found that works is you run a pin along the side of the wire and slide it in to the back side of the connector next to the wire. YOu can slide it in and you will feel it touch the metal parts inside. You then slide a pin in for each wire you want to check. From there, you simply should need to touch the pin to get your reading.

With the connector disconnected, this will throw off the readings as any high resistance wire is not going to be seen as the resistance of a multimeter is really high and this will mask the wire issue. The other option would be to make a set of jumper wires that you can plug into the factory connectors and then have some sort of terminal to touch where it is easy to get to. Then you just have to find some spare pins that fit on to the pins in the connector

 

@Thermo thanks for the reply - at first I wasn't sure what you mean, but I think I got it now, is that it?:.



I checked for continuity and all was good with these pins. Measurements, however, were pretty much the same, with the exception that now the voltage difference between striped wire to ground is 0.1V lower reading compared to voltage across terminals (11.65 vs 11.75).

Ugh, I had hoped it was just a case of replacing an alternator. I am NOT looking forward to stripping the wiring (not even sure local auto-electricians would bother with such a job...). But maybe I can just do the "bypass" trick instead.

 

Boletus, because the voltages are not changing when you are applying a load to the alternator, this would make me think that the ECU may be your source of problem. As you put more load on the alternator, the alternator is going to need more field to generate more power. This would be seen as a larger difference between the white and brown wires or possibly a difference in the voltage between the brown wire and chassis ground. But, because you are saying all these checks are coming up with the same voltage, this is telling me that the ECU is not registering the changing voltage in the car or it is not able to cause a change. If you are seeing 11.70ish volts on the car's electrical system, I would expect the voltage between the white and brown wires to be almost that same voltage. Thinking about it, I can probably narrow it down even more. Sounds like the oscillator circuit in the ECU is failed. If you want me to go into detail about why I say that, I will. But, probably a little bit above your head unless you know what I am talking about with pulse width modulation.

 

@Thermo thanks for the reply. That is interesting, maybe I need to borrow an oscilloscope from work to see what(if anything) the ECU is outputting. The problem is, I wasn't able to find a wiring diagram for this guy that would match the wire coulour I see - would make my life easier! I was under the impression that ECUs rarely fail on these cars.

Regarding PWM, happy to hear it, I'm a physicist by training, just haven't had to use this stuff for a long while, so might be a good refresher!

 

Boletus, pulse width modulation is a way of taking a single voltage (lets use 12 VDC in this case). and creating any voltage between 0 and that voltage. You start with a circuit that would be pulsing at a constant rate (lets use 100 times a second, so 10 milliseconds between pulses). If at each pulse, I have a circuit that would turn on for 10 milliseconds, I would fill the whole time between pulses, so, it would be like applying 12 VDC because it would be on 100% of the time. We change out the 10 millisecond circuit with a 5 millisecond circuit. So, now, you are going to have a circuit that is going to output a 5 millisecond pulse and then be off for 5 milliseconds before the next pulse comes and turns it back on for 5 milliseconds, etc. In this case, you have 5 of 10 milliseconds of on time, so, it would be on 50% of the time. Therefore, your multimeter takes an average of this on and off time (voltage wise) and gets and average of 6 VDC (50% of 12 VDC). As you widen or narrow the on time of the pulse, you can set whatever voltage you want.

In the case of our alternators, this same concept is used. The ECU pulses 12 VDC to the alternator field windings at a rapid rate. This causes the field to build and shrink, causing the output of the alternator to rise and fall. But, this gets averaged out by the battery which acts as a filter to give a flatline DC voltage. This is why it is bad to run your car without a battery in it. The voltage would be bouncing all over the place and the computers would HATE!!!!!! you as they can't make a reference voltage and this causes all sorts of errors to appear. As you apply more electrical load to the electrical system, the windings of the alternator drop a certain amount of voltage. Well, this has to be made up for some how. All that happens is the on time is adjusted to be a little longer, raising the output voltage back up to the desired 13.7 VDC when the motor is running. This is why I was expecting a change in voltage between the brown and white wires as you applied an electrical load.

Granted, our alternators would appear to be taking this concept 1 step further. They are using 2 of these PWM circuits. The brown wire would switch between 6 and 12 VDC while the white wire would switch between 6 and 0 VDC. The benefit of this sort of circuit is you can build in a fail safe condition that will provide some voltage, but maybe not 100% of what is needed.

Did this all make sense?

 

Thermo, yes it makes sense. Do you know the specs of the ECU output (frequency/dut cycle). I'm tempted to check the output with an oscilloscope, but also to maybe even play with function generator to see if I can be spoof the alternator haha.

By the way, the available electrical wire diagram online seems to suggest that diesel doesn't even have "smart charge", only the petrol versions. What's up with that?

 

Boletus, I am not seeing anything about how the ECU controls things I am suspecting that what you are going to run into is that both the brown and white lines are going to be at 6 V when the engine starts. Then what you are going to see is the brown wire switch between 6V and 12 V while the white wire switches between 6V and 0V. Then the ECU is controlling how often the brown and white wires are going to their limits and for how long to regulate the voltage that way.

As for the "smart charge", yes, diesels use a different thought process for some reason. For diesels, they are using a mix of battery voltage and ambient air temp to figure out where to put the voltage. I can see where this is going to be a similar to the petrol setup, but, yet, is different. In short, as the air gets colder, the car is going to raise the voltage slightly (anticipating more electrical load, which the petrol version does not do). It also looks at the state of the battery charge and will raise the voltage as necessary to maintain the battery at 100%. This is similar to the smart charge feature of the petrol engines, but instead of only being available for the first few minutes of driving the car, for petrols, it can do it at any time. So, if you ask me, kinda sounds like the diesel guys got a better setup if you ask me. The ohter interesting thing is that the generator goes into a wake-up and sleep mode based on what is happening with the ignition key. As the key is put into the car, the ECU preps the alternator for starting the car. Kinda like when you turn off the car, the ECU will sense if you are removing the key. At this point it puts the generator to "sleep" by dropping any voltage going to it to a very low voltage.

 

Dont Forget to Check the Integrity of the Battery Cables? seems that most of our Swagg! problems are the Simplest VOLTAGE!!.

 

OK, so I got a circuit tracer (fluke networks pro3000) and tried tracing for anything broken - couldn't find anything. Decided to just change the alternator (wanted to fit a clutch pulley, so needed to test fit anyway). With the alternator replaced it charges again (14.10V with everything on, no problems so far).

After driving for an hour, the positive cable is warm, but not "hot" - I can comfortably hold my hand on it for an extended period of time. In the future I plan to add some thick (2 Gauge or above) copper wires just in case. Hopefully it was just a faulty alternator, and there is no "underlying cause" after all. Thank you Thermo for all your suggestions.