XK8 / XKR ( X100 ) 1996 - 2006
Sponsored by:
Sponsored by:

2001 XK8 Won't Start - Code 1638

Thread Tools
 
Search this Thread
 
  #1  
Old 02-24-2022, 06:55 PM
J-Bone's Avatar
Junior Member
Thread Starter
Join Date: Feb 2022
Posts: 6
Received 0 Likes on 0 Posts
Default 2001 XK8 Won't Start - Code 1638

Hello all, new member here with a slightly dead car and in need of some electrical advice. I have a 2001 XK8 convertible that was running perfect until last weekend when I went to start her up. The engine gave a sort of gasp/cough like it wanted to start, but it didn't. And it would crank freely afterwards, but wouldn't fire up. Pulled out my OBDII code reader and the DLC is giving me P1638 - CAN IC token missing. Tried clearing the code and restarting, but the same thing happened again. Now the dash is displaying a litany of alerts in addition to the MIL: Failsafe Engine Mode, Low Brake Fluid, ASC Not Available, Gearbox Fault.

Checked the battery and it is good at 12.5V. Checked various fuses and relays related to the instrument cluster, no obvious issues there. My initial suspicion was that the instrument cluster blew a circuit somewhere. However, I can see the tach moving when cranking, and the fuel gauge still works. The gear selector is lit up and moves freely, and I can hear the parking solenoid engage and disengage.

So I started testing the network per the JTIS 418-00 manual, starting with the prefix "S" steps for CAN open circuit faults. First I checked the CAN network for continuity by measuring the resistance between pins 6 and 14 on the DLC. Got a steady 61.5 - 62.0 ohms. That seems to be within spec.

Next I checked the circuit between the DLC and the IC for continuity. Disconnected the negative battery terminal, then the IC electrical connector FC25, and re-measured resistance at pins 6 and 14 on the DLC. The multimeter gave me an OL reading. Thought that was a bit odd, but I checked it multiple times and kept getting the same reading.

Now I'm kind of puzzled what to do next. JTIS advises that if the resistance is less than 10.000 ohms, there's a short circuit betwixt the DLC and the IC...and if not, then to proceed on testing the gear selector module and other downstream modules. At the risk of sounding stupid, if I'm getting an open circuit reading between those two points, does that mean there IS a problem between the DLC and IC? Or do I need to move on and start checking other modules on the car? Any help would be most appreciated.
 
  #2  
Old 02-25-2022, 08:02 AM
fmertz's Avatar
Veteran Member
Join Date: Nov 2014
Location: Eastern USA
Posts: 2,604
Received 1,487 Likes on 1,043 Posts
Default

In case you haven't yet, check the TSBs on jagrepair.com, there is a well known issue for instrument clusters failing CAN. It requires basic soldering skill to fix apparently. The TSBs have a well identified range of VINs, but take that with a grain of salt.

Best of luck, keep us posted.
 
The following users liked this post:
motorcarman (02-25-2022)
  #3  
Old 02-25-2022, 09:06 AM
Join Date: Dec 2009
Location: Wise County,TX
Posts: 11,910
Received 7,896 Likes on 4,770 Posts
Default

S484 is for 2000MY cars and the 2001 SHOULD have an M stamped on the back of the INSTPK cover.

POSSIBLE CAUSES
CAN open circuit fault – INST to ECM
CAN short circuit fault
INST failure
ECM failure
 
The following users liked this post:
cjd777 (02-25-2022)
  #4  
Old 02-25-2022, 10:01 PM
crbass's Avatar
Veteran Member
Join Date: May 2019
Location: Durham, NC
Posts: 1,090
Received 924 Likes on 455 Posts
Default

Originally Posted by J-Bone
...Next I checked the circuit between the DLC and the IC for continuity. Disconnected the negative battery terminal, then the IC electrical connector FC25, and re-measured resistance at pins 6 and 14 on the DLC. The multimeter gave me an OL reading. Thought that was a bit odd, but I checked it multiple times and kept getting the same reading....
Did you check continuity with FC25 off? If so, this disconnects the instrument cluster from the CAN network on both sides (both coming from the DLC on pins 10 and 24, and going to the illumination module on pins 11 and 23). An OL reading would be expected if this is the case.
 
  #5  
Old 02-26-2022, 08:35 AM
J-Bone's Avatar
Junior Member
Thread Starter
Join Date: Feb 2022
Posts: 6
Received 0 Likes on 0 Posts
Default

Originally Posted by crbass
Did you check continuity with FC25 off? If so, this disconnects the instrument cluster from the CAN network on both sides (both coming from the DLC on pins 10 and 24, and going to the illumination module on pins 11 and 23). An OL reading would be expected if this is the case.
Hello crbass - yes, I checked continuity on pins 6 and 14 of the DLC with FC25 disconnected from the instrument cluster, and that’s where I got the OL reading. I thought that it would be expected too given that the circuit between the two was disconnected, but the JTIS confused me somewhat by asking if the resistance was less than 10 ohms. OL is infinite resistance, so it seems like the answer is “no” to that JTIS step and it’s time to move on to the gear selector.

What’s puzzling me is that the resistance was within normal range in my initial measurement of those pins at the DLC (with FC25 connected). From doing some research on others with this fault, there’s usually a higher resistance like 120 ohms if there is a short in the CAN system.

fmertz and motorcarman - thanks for your posts, I did come across that TSB and my car’s VIN is way out of the specified range so I discounted it. It is however a very early 2001 model so it’s possible It could have the same issue. Still not ruling out IC failure at this point but I don’t have a smoking gun yet. I’ll check the VCATS and see if the cluster has the “M” mark on it. Will keep you all posted.
 
  #6  
Old 02-26-2022, 01:19 PM
Join Date: Dec 2009
Location: Wise County,TX
Posts: 11,910
Received 7,896 Likes on 4,770 Posts
Default

There is a TERMINATING resistor at EACH end of the CAN bus.
ONE in the INSTPK and ONE in the ABS module.
The CAN bus resistance with a functioning bus is about 60 Ohms. (two 120 Ohm in PARALLEL=60)
 
  #7  
Old 02-26-2022, 02:22 PM
J-Bone's Avatar
Junior Member
Thread Starter
Join Date: Feb 2022
Posts: 6
Received 0 Likes on 0 Posts
Default

Originally Posted by motorcarman
There is a TERMINATING resistor at EACH end of the CAN bus.
ONE in the INSTPK and ONE in the ABS module.
The CAN bus resistance with a functioning bus is about 60 Ohms. (two 120 Ohm in PARALLEL=60)
Understood. Problem is, I am getting ~62 ohms of resistance at the DLC which indicates a functioning CAN bus, but the car thinks otherwise and it’s throwing me a 1638 code indicating the CAN isn’t working. A mystery…
 
  #8  
Old 02-26-2022, 03:24 PM
Join Date: Dec 2009
Location: Wise County,TX
Posts: 11,910
Received 7,896 Likes on 4,770 Posts
Default

The INSTPK for the 4.0 liter X100 does not need configuration so you could substitute a 'known-good' unit (with correct VCATS) to see if the fault follows the INSTPK?

The ECM can be substituted also (with the correct VCATS) BUT you need to perform 'IMMOBILIZATION-SETUP' everytime any different ECM is connected to the CAN bus.(even the original after a different one was installed)

There can be only one ECM configured to the car at any one time.

The gear selector illumination module is the only thing between the INSTPK and the ECM.
Too bad it is a LISTEN-ONLY node on the CAN bus.
I wish at times I could interrogate the selector lever module for DTCs?
 

Last edited by motorcarman; 02-26-2022 at 03:28 PM.
  #9  
Old 02-26-2022, 09:15 PM
crbass's Avatar
Veteran Member
Join Date: May 2019
Location: Durham, NC
Posts: 1,090
Received 924 Likes on 455 Posts
Default

Originally Posted by J-Bone
Hello crbass - yes, I checked continuity on pins 6 and 14 of the DLC with FC25 disconnected from the instrument cluster, and that’s where I got the OL reading. I thought that it would be expected too given that the circuit between the two was disconnected, but the JTIS confused me somewhat by asking if the resistance was less than 10 ohms. OL is infinite resistance, so it seems like the answer is “no” to that JTIS step and it’s time to move on to the gear selector.

What’s puzzling me is that the resistance was within normal range in my initial measurement of those pins at the DLC (with FC25 connected). From doing some research on others with this fault, there’s usually a higher resistance like 120 ohms if there is a short in the CAN system.
.
Great, so you appear to have ruled out CAN short and CAN open from Bob's (motorcarman) list.

So, next two things are INST and ECM. Easier to buy an INST with the correct VCATS on ebay and check to see if this fixes your problem (you probably want to have one around your mileage to prevent issues with mileage checks from the DMV) than to do the immobilization setup from the ECM unless you have the software, then that's easy too...
 
  #10  
Old 03-08-2022, 08:53 AM
J-Bone's Avatar
Junior Member
Thread Starter
Join Date: Feb 2022
Posts: 6
Received 0 Likes on 0 Posts
Default

Update - I bought a used instrument cluster off eBay and installed it, but got the same exact issue. Engine cranks, coughs once, then throws a P1638 code. Measured 61.5 ohms on the CAN network and voltages are spot on with the substitute cluster as well, So it seems like I can rule out the instrument cluster and the CAN network as potential causes of the problem.

I’m leaning towards the ECM being the issue but as I don’t have the equipment or expertise to repair/reprogram it, next step is to have the car towed to a shop (if I can find one). Stay tuned…
 
  #11  
Old 03-12-2022, 12:29 PM
J-Bone's Avatar
Junior Member
Thread Starter
Join Date: Feb 2022
Posts: 6
Received 0 Likes on 0 Posts
Default

Update 2x - problem solved. The issue was the key. Before all of my woes started, I had dropped the key by accident - not a long drop and it landed on soft plastic, but apparently that was enough to damage it somehow. So when I tried to start the car using that key, the car thought it was being stolen and went into immobilization mode. The shop reset the immobilization flag in the modules and then the car started and ran fine with the second key I have.

Not sure why the car was throwing the P1638 code - seems to have been a red herring. That would explain why all my CAN testing didn’t reveal any issues. The network and modules were fine all along. On the plus side, I now know how to change the burnt-out bulbs in the instrument cluster.

Now the problem is that I only have one working key for the car and if something happens to it, I have a dead car on my hands. Going to stop by my local Jag dealer next week and see if they can repair or replace the damaged key. Wish me luck…🍀
 
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
SLazz
XK8 / XKR ( X100 )
5
05-08-2018 01:05 PM
johnkent
X-Type ( X400 )
3
09-26-2017 12:19 AM
gthawk1
X-Type ( X400 )
2
07-18-2015 05:44 AM
Rlover
XJ XJ8 / XJR ( X308 )
11
08-31-2011 05:18 PM
Rlover
XJ XJ8 / XJR ( X308 )
9
06-23-2011 09:30 AM

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 


Quick Reply: 2001 XK8 Won't Start - Code 1638



All times are GMT -5. The time now is 08:31 AM.