LOL neither do i but the fact is after doing what you suggested the answer although i hate to say is the battery is tip top and the fault codes are still present as before. Will proceed to check the tcm to ecm loom for issues plus the 2 lever bolts on the trans while dropping the oil pan and swapping out the seals and mechatronics sleeve. Hopefully on the way i will find a wiring fault preferably before dropping the sump. Well i think the battery thing was a long shot but definitely worth the extra effort and good to know it is holding good charge. I will just await a non rainy day and proceed

 

With P1643 still, I'd follow the workshop manual. Hopefully it's a chafed/chewed wire.

 

Yes thats the plan but appreciate your help with the battery thing and will update here as i go. I think coms between tcm and ecm is the key to this somewhere and the faults are a result but im sure to find eventually

 

What would be the best way to test the tcm to pcm coms for breaks or shorts because nothing visible after checking everything. I have a meter for this but honestly not sure what way to go about it.

 

It's hard to do with out a breakout box. You would need to remove the big ECM plug (It's on the passenger side of the engine compartment on the firewall down low). Not easy access! Then follow the wiring diagrams to see what wire is hooked to what. But if you still have codes I think that would be the best way to attack it.

JagV8 mentioned P1643. Do you still have this AFTER clearing all the codes and starting the car? And again to me this points to the TCM harness that plugs into the transmission and is known to be caused by this leaking sleeve. So until you can verify and fix that I would not be chasing anything else. Is there anyway you can get the car scanned by an SDD? These types of errors tend to sent more codes as the other network communications will also fail. Sometimes these combination of codes can give us a clue on what to do next.

We don't see bad TCM's so I don't think that your problem at least not yet?
.
.
.

 

Cheers for the reply but already in midst of doing the valve body seals and mechatronics sleeve and if i need to then check the loom after this then just wondering the best approach other than splitting myself in two!

 

OK I think that is the way to go. Get the sleeve installed and stop any transmission fluid from leaking at that point.
Might not fix it but it does need to be repaired anyway.
.
.
.

 

Well atf changed plus filter and seals including the sleeve and still the same issues persist so i guess back to chasing my **** around this rock!! bloody unbelievable that this is such a pain to diagnose. Started up to complete the atf change then once turned off would not restart or shift so stuck in park also. Must be a short or broken wire within the ecm/tcm loom or bad earth somewhere so will endeavour to push on with it

 

P1643 is verry rare. Some have had the mechatronic sleeve loose. At least one had to change the TCM.

 

With due respect i dont think its very rare upon reading many posts on this fault code but its just unfortunate that most have a range of causes that never come to light because reading about them is one thing but if people dont report back on there solutions or fixes? . There are still a few things to go through and check before i find out the cause but rest assured i will report my findings in the hope it helps someone else maybe.

 

I got 108 matches across all models out of the hundreds of thousands (millions?) of posts and some had fixes.

 

Today after much brain stretching on how the multiplexing system works before going all out to test, i thought it might be wise to get every earthing point cleaned and secured! After starting at the front left corner behind the headlight the earth was very grubby and didnt look to be doing much but definitely needed removing to sand and clean off the crust that has built up. I bolted it back on and just out of curiosity tried to start the car and with positive results! It started and the red symbol on the gearshift was lit up red in park as it should be, not for long but long enough for the tcm to come alive thus losing two of the fault messages on the dash. Switched off to try and restart and no go.. As darkness was down i locked up and left until tomorrow but before this happened i noticed the indicator flash and interior lights were on? went outside to check and car was unlocked and making a clicking sound from relays maybe one under the bonnet (key out) of ignition so in favour of saving the battery i removed it to save power and locked the car manually. Although good and bad results i think overall i could be getting to the root cause of the problem as its sounding like a short to ground all be it intermittent which kind of leads me to the wiring loom in that vicinity.

 

+1 on that. Have you seen the pinpoint tests given in the service manual?

http://www.jagrepair.com/images/Auto...M-Workshop.pdf

I did a Control-F search of the document, and got 6 hits for P1643. Scroll down to page 2845, and it tells you to do pinpoint test E, which starts on page 2853. This test has 13 specific steps using a multimeter, checking resistance between two points, voltage, etc. Most of the tests are done from the diagnostic connector under the dash, so that part should be very easy.

Normally I don't recommend skipping around the manual, but some of the steps do require access to the TCM. Lazy as I am, I'd do the other easy steps first and come back to the TCM later.

 

Now thats gonna save me some searching and alot of time karl thanks!! As usual the minute you start to make pace the rain hits lol so good time to get my head around it. All this learning is great but my eyes are so pinned out i look like ive been on lsd for a year!! Was about to remove the front bumper as im sure i read there was a trouble area there where the loom can suffer from abrasions causing shorts etc. Yesterday was a brief moment of joy when i see the tcm come to life if only for a nanosecond or two but small steps....

 

Have now done the checks and im pretty sure the fault lies between the tcm plug and ecm on the +- of the can signal wires! so the tests you pointed to karl were definitely helpful. Have yet to complete the ecm part of the test until get to the dammed thing lol for sure the security bolt will not be reused and a nice 10mm will do the job. The ecm test refers to the ip plug fc015? Injector pressure sensor maybe i will have to check but otherwise slow progress

 

Groovy! But I must play devil's advocate and ask specifically what lead you there? The pinpoint test E in the service manual has specific steps, labeled E1 - E13. Did one of those fail?


This one?:

E10: Check Continuity of the CAN+ Circuit

I was scratching my head why the test of the CAN+ circuit would send you to the fuel pressure sensor, but it's called the IP sensor in the wiring diagrams. Then I realized in this case, IP means "Instrument Panel", even though it's called the instrument cluster in most Jaguar documentation.

For the CAN circuit, see figure 20.1 in the wiring diagrams. This shows the wire pair connecting various modules, and it has the corresponding connector/pin designations on the ECM as called out in pinpoint E10. (Note PI001-124 in the service manual is shortened to PI-124 in the wiring diagrams. Also note the second digit is the letter I, not the number 1.) But over at the instrument panel/cluster in that same diagram, the designations don't match up. What gives?

It appears the designation on the instrument panel/cluster changed somewhere around 2005. If you look at figure 20.1 in the 2004 wiring diagram, the designations match the service manual. 2005 wiring diagram? Not so much. You will have to look at both years for the wiring diagrams and extrapolate for your car. This is somewhat typical in the world of documentation, as some minor changes often never made it to the finished manual, especially near the end of production.

Also note it would appear you can do the same basic wiring checks via the diagnostic connector under the dash, where you'd connect a scanner. Rather than pull the instrument panel/cluster, I'd check continuity to the diagnostic connector first.

 

I am abit confused as just following the e series tests is fine to a point but it leads me to think that somewhere between tcm and ecu/diagnostic socket this fault exists? Basically on the Etest E2 was a no-E3 was a yes resistance 1 ohm-E4 was a no-E5 was a no E6 was a no-E7 was a yes-E8 was a yes-E9 was a no. I stopped here because now i feel the issue is located and to further test forwards to the ecm would further confuse me although i did check E12 pin 123 to 124 on the ecm plug as i had it unplugged at the time and read 1 ohm? Im not sure all these tests will lead me to the fault or send me further away from it at this point. Sorry if i sound confusing just trying to make sense of what im testing. Overall looking at what i have from those tests there is a short to battery on can + between diagnostics connector and the tcm.

 

Can you doublecheck this result, please? This points to an unwanted connection between CAN+ and the battery.

This particular step (E3) seems to have another copy/paste error. The header says you'll be checking for a short between two points, but then it says to measure voltage. But the very next line asks if the resistance is less than 5 ohms. That makes me almost positive the previous line was supposed to say measure resistance, not voltage. And if so, resistance less than 5 ohms is out of limits, and you measured 1 ohm. Sounds like a problem, so please check this one again.

When working through those pinpoint tests, there's a trick to know what results are expected. Each question has a good or bad answer. The "good" answer simply sends you to the next step, which means the results are within limits. The "bad" answer tells you to repair or replace something, which means the results are out of limits.

I know all this testing may seem overwhelming, but I'd say you're hot on the trail. Keep going!

 

Cheers karl, I did the test twice anyway karl as i wanted to be sure myself and the second try i wrote this all down just so i didnt forget lol a lot to take in when your not used to it but this is partly why i didnt complete the test to 13 as i already know the ecm is fine as is the instrument cluster so it has kind of got me convinced! only problem is where abouts in the loom it is? Do i start testing the can+ fromt the tcm plug in increments and just work towards the next connector as i believe it leads back to the jgate hence it not illuminating the red position icon?

 

I’m sorry I don’t want to be a negative Nelly.
And you guys are doing great with this diagnosis.
but I just wanted to point out that I have been here in the same situation.
Exhausting all of these avenues. for diagnosing
no CAN comms between ECM and TCM.
and I noticed somewhere further up this thread
and hopefully this is not the case.
But I just want to point out that a replacement TCM was my fix.