Battery drain, bad jump start, blown ECU, and a cluster gone mad
#1
![Question](https://www.jaguarforums.com/forum/images/icons/icon5.gif)
I will start this thread with my questions, but also share my experience to date regarding some electrical debugging that will help to pull together info from disparate threads. I will reference links and docs to help others.
First, ASI in Tennessee confirmed the result of my testing (methodology will follow, with tips) that I fried my ECM (aka ECU). They said it is irreparable, so I ordered a replacement on eBay (from a good source, tomsforeignautoparts). However, Tom’s only knows the part number (LNF1410AC), which matches mine, but they don’t know the full VCAT code (the last 3 digits, which on my model are located on a sticker in the spare tire wheel well in the trunk). So I know the part will work, I just need to know:
OK, next post will be the story of what went down and what I did to diagnose.
First, ASI in Tennessee confirmed the result of my testing (methodology will follow, with tips) that I fried my ECM (aka ECU). They said it is irreparable, so I ordered a replacement on eBay (from a good source, tomsforeignautoparts). However, Tom’s only knows the part number (LNF1410AC), which matches mine, but they don’t know the full VCAT code (the last 3 digits, which on my model are located on a sticker in the spare tire wheel well in the trunk). So I know the part will work, I just need to know:
- Do I really need to get it programmed? The 418-03 Technical Bulletin (“418-03AM VCATS Compatibility.pdf”) suggests that it’s safe to try it and see if it works. My XKR is a 2000, the same year as the replacement ECU which is also from an XKR with a close VIN and both are convertibles. I don’t think I need to do anything with the immobilizer, but I could be wrong. Sources on the forum seem to indicate the immob is only required around 2003 or later.
- If I do need to get it programmed, I assume it’s OK to install it and drive it to the dealer/mechanic? I want to avoid anyone touching the car other than to do the programming. (I’ve never used a mechanic, being one, so I just hate to pay someone to work on my car; I simply don’t have the equipment to reprogram it).
OK, next post will be the story of what went down and what I did to diagnose.
#2
![Default](/forum/images/icons/icon1.gif)
With a crossed wired jump start it will blow some fuses including the instrument cluster so you'll have to pull all the fuses and check . I blew the alternator voltage regulator at the same time on mine of similar electrical layout .
After you have done the fuse survey the ECU may be fine and don't put the car on a possible bad battery charger but charge the battery up off car then test things out with pure sweet sweet loving DC power that only a battery can put out . A instument cluster going wacko wacko wacko is a indication of a unpure DC power source , done it myself .
Someone put a bad external power generator on 3 different aircraft once before he clicked , not me . What a mess .
After you have done the fuse survey the ECU may be fine and don't put the car on a possible bad battery charger but charge the battery up off car then test things out with pure sweet sweet loving DC power that only a battery can put out . A instument cluster going wacko wacko wacko is a indication of a unpure DC power source , done it myself .
Someone put a bad external power generator on 3 different aircraft once before he clicked , not me . What a mess .
Last edited by Lady Penelope; 04-27-2018 at 09:16 PM.
#3
![Lightbulb](https://www.jaguarforums.com/forum/images/icons/icon3.gif)
OK, here is the story of what actually went down and what I did to diagnose.
The brand new car battery kept draining after 2-3 days, which is annoying as it isn’t my daily driver, so I had it on a trickle charger. But one day, as I was about to go for a drive, it was dead even on the charger. Test showed 1 amp draw (when lights were activated) where the charger maxes at 750 mA. I never knew about the “sleep mode” so I never set the parking brake and I never lock the car, so it probably never went down to the lower draw per spec. I didn’t even know that there were these modes. Look for “Quiescent Current Testing.pdf” to learn more.
OK, so what bonehead thing did I do to fry my ECU? Well, I was in a hurry so I hooked up a boost charger, and when that didn’t jump it (like I said, I was in a hurry), I hooked up a second lithium battery booster. (Yes, I can picture you slapping yourselves on the forehead… sorry, it’s my first Jag). The lithium booster alarm went off, and I could tell something was wrong. I disconnected it, put on a 10-amp charger, and opted for a different ride.
There is a proper procedure for jumping the car. Look in the Handbook. Ignore it at your peril. Cost me big time.
When I came back later, the car started, but the cluster was going mad. It displayed numerous faults: ASC unavailable, incorrect part fitted, gearbox fault, low brake fluid, failsafe engine mode, engine light, ABS light, red and amber warning indicators, plus there is no temp gauge reading and it shows 65MPH standing still and no tach. Plus weirdly, every time I started the car the odometer incremented between 100 and 5000 miles (the odo had increased by 104,000 miles when I finally removed the cluster).
OK, so on to debugging. First I removed the instrument cluster and performed Service 413-S484 (look for “413-S484am2 INSTPK CAN failure.pdf”) to fix the flowed solder (short) on the cluster, which was likely the original cause of the battery draining. (I verified that it had really shorted: the repair is easy, and you can see a forum link here.)
I then re-installed the cluster and cleared the various OBD error codes (mainly related to CAN bus errors) and did a hard reset several times, including a 20 minute test drive to make sure systems were refreshed, but no change, still crazy cluster.
BTW, I also checked all connectors, grounds, and fuses. Note that for some reason they don’t suggest pulling fuses one at a time anymore; I forget why, but I have a meter with a continuity beep so I can test across the fuses quickly with probes, which is a lot easier. It won’t verify that the fuse is actually connecting within its spade terminal, but I have no corrosion so it’s a pretty sound test (yes, pun intended).
OK, so then I replaced the cluster with one off eBay, and it only displayed the errors related to its inability to connect to CAN bus systems downstream from the ECU.
Pause for explanation: These cars have 3 networks, an optical network which runs the entertainment and navigation (if so equipped), a high-speed CAN bus which connects realtime sensors and computers (including the ECU, ABS, speed sensor, and the shift illuminator). The medium-speed SCP network runs the door, top, and other interior electrical systems.
So when you get CAN bus errors (which any OBD2 device will read), you are debugging either the network (which is wired on two twisted yellow and green wires, one for CAN high and one for CAN low), the electrical components on the CAN, or the instrument cluster itself, which happens to connect to both the CAN and the SCP networks.
You don’t need too much high-tech gear to test the CAN. There are some awesome diagnostic manuals (look for “DTC Summary.pdf” and I found a 148 page “2000 Model Year XK8 Range Electrical Guide” on this forum) but there are some simple things the shade tree mechanic can do as well.
The DTC guide is nice. In my case I saw for example:
The DTC summary for P1638 is:
Fault
CAN INST token missing
Default Action
None
Engine speed and coolant temperature data missing at instrument pack
Possible Causes
INST failure
CAN open circuit fault – INST to ECM
CAN short circuit fault
ECM failure
The CAN wires can be tested for continuity, but one trick is to use a digital multimeter to test the voltage and resistance of the wires. These sometimes short to ground or to each other, or break leaving an open circuit, so testing can be a hassle. But here’s a trick: you can check for correct voltage and ohms at every component, but there are 120 ohm resistors on each end. So if you get 60 ohms connecting CAN high to CAN low, through the OBD port (slot 6 is CAN high, 14 is CAN low), then the wiring is probably fine. My test reported the CAN wire continuity end-to-end. On to the components then.
I saw that from the cluster the first stop is the shifter (J-Gate). If the gear shift lights up, then the CAN wires are good to there. Next they go to the ECU. From the ECU they branch and one goes to the ABS and the other to the transmission control and speed sensors. My shifter lights worked, but data from both the ABS and the speed sensor were not coming through, and since it is a branch according to the electrical diagram, I guessed that either both had fried or it was the ECU itself, the latter being by far the most likely. Visual inspection of the ECU board looked good, and I saw no leaking capacitors. But ASI confirmed it was fried.
Bottom line: don’t ignore battery discharging, check to see whether your model requires the instrument cluster modification, and don’t, just don’t, jump start without following proper procedure!
The brand new car battery kept draining after 2-3 days, which is annoying as it isn’t my daily driver, so I had it on a trickle charger. But one day, as I was about to go for a drive, it was dead even on the charger. Test showed 1 amp draw (when lights were activated) where the charger maxes at 750 mA. I never knew about the “sleep mode” so I never set the parking brake and I never lock the car, so it probably never went down to the lower draw per spec. I didn’t even know that there were these modes. Look for “Quiescent Current Testing.pdf” to learn more.
OK, so what bonehead thing did I do to fry my ECU? Well, I was in a hurry so I hooked up a boost charger, and when that didn’t jump it (like I said, I was in a hurry), I hooked up a second lithium battery booster. (Yes, I can picture you slapping yourselves on the forehead… sorry, it’s my first Jag). The lithium booster alarm went off, and I could tell something was wrong. I disconnected it, put on a 10-amp charger, and opted for a different ride.
There is a proper procedure for jumping the car. Look in the Handbook. Ignore it at your peril. Cost me big time.
When I came back later, the car started, but the cluster was going mad. It displayed numerous faults: ASC unavailable, incorrect part fitted, gearbox fault, low brake fluid, failsafe engine mode, engine light, ABS light, red and amber warning indicators, plus there is no temp gauge reading and it shows 65MPH standing still and no tach. Plus weirdly, every time I started the car the odometer incremented between 100 and 5000 miles (the odo had increased by 104,000 miles when I finally removed the cluster).
OK, so on to debugging. First I removed the instrument cluster and performed Service 413-S484 (look for “413-S484am2 INSTPK CAN failure.pdf”) to fix the flowed solder (short) on the cluster, which was likely the original cause of the battery draining. (I verified that it had really shorted: the repair is easy, and you can see a forum link here.)
I then re-installed the cluster and cleared the various OBD error codes (mainly related to CAN bus errors) and did a hard reset several times, including a 20 minute test drive to make sure systems were refreshed, but no change, still crazy cluster.
BTW, I also checked all connectors, grounds, and fuses. Note that for some reason they don’t suggest pulling fuses one at a time anymore; I forget why, but I have a meter with a continuity beep so I can test across the fuses quickly with probes, which is a lot easier. It won’t verify that the fuse is actually connecting within its spade terminal, but I have no corrosion so it’s a pretty sound test (yes, pun intended).
OK, so then I replaced the cluster with one off eBay, and it only displayed the errors related to its inability to connect to CAN bus systems downstream from the ECU.
Pause for explanation: These cars have 3 networks, an optical network which runs the entertainment and navigation (if so equipped), a high-speed CAN bus which connects realtime sensors and computers (including the ECU, ABS, speed sensor, and the shift illuminator). The medium-speed SCP network runs the door, top, and other interior electrical systems.
So when you get CAN bus errors (which any OBD2 device will read), you are debugging either the network (which is wired on two twisted yellow and green wires, one for CAN high and one for CAN low), the electrical components on the CAN, or the instrument cluster itself, which happens to connect to both the CAN and the SCP networks.
You don’t need too much high-tech gear to test the CAN. There are some awesome diagnostic manuals (look for “DTC Summary.pdf” and I found a 148 page “2000 Model Year XK8 Range Electrical Guide” on this forum) but there are some simple things the shade tree mechanic can do as well.
The DTC guide is nice. In my case I saw for example:
The DTC summary for P1638 is:
Fault
CAN INST token missing
Default Action
None
Engine speed and coolant temperature data missing at instrument pack
Possible Causes
INST failure
CAN open circuit fault – INST to ECM
CAN short circuit fault
ECM failure
The CAN wires can be tested for continuity, but one trick is to use a digital multimeter to test the voltage and resistance of the wires. These sometimes short to ground or to each other, or break leaving an open circuit, so testing can be a hassle. But here’s a trick: you can check for correct voltage and ohms at every component, but there are 120 ohm resistors on each end. So if you get 60 ohms connecting CAN high to CAN low, through the OBD port (slot 6 is CAN high, 14 is CAN low), then the wiring is probably fine. My test reported the CAN wire continuity end-to-end. On to the components then.
I saw that from the cluster the first stop is the shifter (J-Gate). If the gear shift lights up, then the CAN wires are good to there. Next they go to the ECU. From the ECU they branch and one goes to the ABS and the other to the transmission control and speed sensors. My shifter lights worked, but data from both the ABS and the speed sensor were not coming through, and since it is a branch according to the electrical diagram, I guessed that either both had fried or it was the ECU itself, the latter being by far the most likely. Visual inspection of the ECU board looked good, and I saw no leaking capacitors. But ASI confirmed it was fried.
Bottom line: don’t ignore battery discharging, check to see whether your model requires the instrument cluster modification, and don’t, just don’t, jump start without following proper procedure!
#4
#5
![Default](/forum/images/icons/icon1.gif)
Rich, Thanks for taking the time and posting easy access sites to get the full monty.
Don't know if I have the knowledge of the where the heck the start and stop of all those wire locations, but now I at least have a understanding of where I would need to go.
Which section of NC is your location? Jon89 and I have got together to preform a bunch of things and I'll give you a invite on the next get together. Hopefully not for fixing anything in the near future. LOL
Wayne
Don't know if I have the knowledge of the where the heck the start and stop of all those wire locations, but now I at least have a understanding of where I would need to go.
Which section of NC is your location? Jon89 and I have got together to preform a bunch of things and I'll give you a invite on the next get together. Hopefully not for fixing anything in the near future. LOL
Wayne
#6
![Default](/forum/images/icons/icon1.gif)
Worth noting it's not just Jaguars that are susceptible to damage from jump starts - most modern cars state a correct procedure in the handbook. I once fried the TV /monitor screen on my BMW 7 series through not following the recommended method. Car was under warranty but BMW refused to cover - cost me dearly. I'm always very cautious these days about jump starting anything. Lesson learnt(expensively).
Thread
Thread Starter
Forum
Replies
Last Post
Terry007
XJ XJ8 / XJR ( X308 )
6
06-24-2015 06:29 PM
eurojourney
XK8 / XKR ( X100 )
6
08-24-2007 10:59 AM
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)