Following on from my previous post on the airbag warning message, I did some testing of the Air bag Module (ABM) today. I identified the correct output signal that the Instrument panel expects following a “power on” of the ignition system. First, I used the original, yellow sheathed Airbag wiring loom and some resistors to fake the existence of Airbags. It turns out that the steering wheel Airbag reads approximately 1.2 ohms when in good condition.
Please be very careful when doing anything with “live” Airbags, they are dangerous hence I used resistors instead. Please, Please view any one of a number of videos available on the subject at the well known online video site prior to connecting anything electrical to an Airbag. If you do not know what you are doing, get someone who does. I suggest you do not touch them at all. I have measured mine so you don’t have too. Just keep them locked away somewhere safe or dispose of them safely. Do not come back to me of one goes off accidentally, you have been warned!
I did not have any 1.2R resistors. I ended up using four 6R resistors in parallel and just pushed them into the appropriate plug in the loom. It took a little debugging to stop the Airbag module from buzzing an error code. Once I had the rig working, I could see what the output signal should be in normal operation. The issue I had was that the Safing module does not have a separate wire for its ground connection. It relies on the connection to the chassis via the mounting bolts. As I had the module sitting on a wooden panel, I needed to provide it with a separate ground to get it working.
As expected, it turns out that although the circuit diagram says Ground and +B as the active/inactive conditions the reality is a little more complex than that.
With power permanently applied to the module the output signal is +B
Turn on the ignition power and the signal drops to ground for approximately 10 seconds
This is the Self test Pass signal “sequence” I was unaware of prior to this test
After that, the signal returns to +B and remains there unless an error is identified
I did try disconnecting the resistor packs (pretending to be the Airbags) one at a time and the signal goes from high to low as follows:
3 times high to low
A short pause at high
Followed by either two or three high to low again depending on which “airbag” (actually a resistor pack) was disconnected
It then repeats the sequence again until the problem is resolved
This, I presume, equates to error codes 32 or 33 which is similar to the way the error code was displayed by the ABS module on my old XJS. It flashed the ABS light on the dashboard after you shorted a couple of pins together on the ABS module
The question is now, do I generate the same signals using just a few components or do I hard wire the required signals and use the Airbag module?
Debugging the Transmission MIL light
Having identified the requirements to resolve the AIRBAG warning on the digital display. It was time to look into the Transmission MIL light issue. First there was the connection of the transmission loom to the transmission control module. I removed the panel below the gear leaver assembly which enabled me to push the transmission wiring connectors through a hole in the tunnel and connect up to the transmission module. The transmission error light was still taunting me! I did a slight tidy up of the wiring and re-connection of the Transmission module ground but still had the error. A little more debugging and discovered that I had not connected the permanent power feed to the transmission control module. I connected this up to the appropriate permanent feed, the MIL light went away.
This post follows on from the last post where I explained the issue I was having when attempting to use the digital display from the XJR6. The display was showing the mileage then FLUID and AIRBAG messages. Having shorted the appropriate input to ground for the engine coolant sensor the FLUID message went away leaving me with the AIRBAG message to resolve. I tried shorting to ground or +B on the appropriate input but nothing seemed to clear the message. A photo of the Airbag circuit diagram is below, click for a larger version.
Since that time I have done a little more debugging. I proved that the feed to the main processor in the instrument circuit was receiving different signals depending on the status of the signal that should come from the Airbag Module. The Instrument display circuit, therefore appears to be receiving the appropriate signals to turn off the error message but it is not. The signals on the “Y” output of the HC151 chips looked like below with the signal grounded or at +B so you can see there is a difference and it was being propagated through the circuit board.
A little more research showed that there is potentially an error message generated due to “No diagnostic module selftest Pass signal”. Does that mean that there is a short signal sequence that is emitted from the Airbag module after power on that the Instrument circuit is expecting? There is no clock signal synchronization between the two modules so it has to be pretty basic pattern. I suspect something like a high signal initially and then low for maybe a second or two in each state. I really don’t want to have to add the Airbag module into the car. Using a test rig is needed to see if I can work out the signal states for the Airbag module in normal operation. Hopefully a very simple workaround can be found other than actually adding yet another module into the car ;-(
The Airbag module has a single output signal (Airbag Failure Warning, connector AB1, pin 4). The manual states this is either ground or +B. It does not have any other connections to the CAN bus or directly to any other module. The only connections are to the two impact sensors, the Safing sensor and the airbags themselves. Doing a little research, it appears that Airbags have a resistance of around 2-3 Ohms each. This should be replicated rather than actually connecting the Airbags to the module along with the potential safety implications! I still have the full Airbag system wiring loom along with one of the impact sensors and the Safing module. I will use this as the test setup and fake the actual airbag connections to identify what the “Selftest Pass” signal looks like.
This will dictate one of three possible outcomes:
Replaced the module by a very basic circuit to replicate the required initial signal(s) required to stop the error message
I will fit the Airbag module in the car just to eradicate the error message and allow the use of the XJR6 digital milometer
I will not use the digital display from the XJR6 to provide a milometer and use something else like a GPS based milometer instead.
Having connected up the XJR dashboard display, I managed to get the Mileometer working showing the previous XJR milage. However it also displayed AIRBAG and FLUID repeatedly so not so good. The FLUID was straight forward to resolve by connecting the appropriate wire to ground. I misread the circuit diagram and thought the AIRBAG signal needed to be at +B. Having temporarily connected the wire to a +12V battery feed it still said AIRBAG. I tried grounding the wire, still AIRBAG! I tried adjusting the voltage to the wire using a variable resistance box but still same AIRBAG error. I took the display board out of the assembly and studied it closely.
I traced the input wire (pin 45) to an input on a surface mounted integrated chip with HC151 printed upon it. This is an 8:1 multiplexer chip. My guess is that the inputs are sequentially scanned and the outputs read to determine the status of the input signals. What that means is that if the chip is faulty (possibly damaged by me directly connecting 12V to the input?) then I cant just tie the outputs to one state as I will get other errors flagged as well. I am not sure this is the problem because the detail in the circuit diagram does say it is either ground or +B. The input circuit seems to have the signal going to the middle of a pair of resistors so the input voltage would be split in half anyway. The spec of the chip states 6V maximum input which is a little close, especially as my old XJS with an AJ16 engine used to run at 13.8V most of the time!
I have ordered a replacement IC (or 5) just in case. They are not expensive but I will have to wait a few days for them to arrive. The suspect chip (11) is highlighted by the red arrow in the picture below.
The image below shows the location of the input signal, the +B and the ground points and the signal feed to pin 12 of IC 11
The Engine is back in again, hopefully finally this time 🙂
Firstly, sorry the site was down yesterday. Something wrong at the ISP.
I had a week off and managed to spend a day in the garage. I set things up ready to install the engine and just as I pushed the engine & hoist a few feet out of the garage to enable me to push it into the engine bay it started raining. Well obviously it was time for a tea break so I left it like that for about an hour and then proceeded to get the engine back in the car. It is now sitting on the engine mounts and the gear box is supported underneath with wooden blocks. I do have a slight problem now as I cant get under the car and into the pit without moving it and obviously can’t move it with the gearbox on blocks. I will have to support the rear of the engine to allow me to move it and carry on fitting the gearbox mounts in the new location.
The engine loom was pushed through the hole in the bulkhead as you can see below and then I did a little debugging of the electronics.
I connected the power, checked for no smoke 🙂 and looked to see if the ODB2 port was working.
OBD2 wireless module powered up, app on iPhone could not see/connect to the ECU
Unplugged the ECU plugs and checked for 12V (+B) – not present
Checked the ECU power relay was switching on – it was not?
Checked the power in the new fuse box in the engine bay and discovered the relay in the fuse box enabling the “switched” power was not being energised
Followed the cables back because this was working a while ago and discovered the 48 pin connector I built into the bulkhead was not connected together fully.
Fuse box Relay was still not being engaged
Traced this back to the cable I was using for testing was not finally connected to the relay I wired up. The S2 ignition switch switches its outputs to +B when in each position. All the ignition switch positions on an XJR switch the associated output to ground. This means I need relays to convert the +B outputs to a ground signal to feed the appropriate signals to the various control units.
Wired up the correct cable to the relay output and now the Engine Bay fuse box is alive!
reconnected up the ECU plugs and now the ODB2 module/iPhone can talk to the ECU – hooray 🙂
Gave error code P1620 which looks like immobiliser error
Ok now I know the iPhone app will talk XJR I will purchase the full app and update you on further progress.
I have been working with HortonWorks recently and they wanted to see the installation process of an Hadoop Cluster using Isilon as the storage layer and any differences from a standard DAS based install. I ran into numerous issues just getting the Linux hosts into a sensible state to install Hadoop. It thought I would summarise some of the simple issues that you should try to resolve before starting to install Hadoop.
Selected a Basic Database Server as the install flavour
Choose a reasonable sized OS partition as you might want to make a local Hadoop repository and that is a 10GB tar.gz file download. You have to extract that so over 20GB is needed to complete that process. I ended up resizing the VM a couple of times so I would suggest at least 60GB for the Ambari Server VM including the local repository area.
You might want to set up a simple script for copying files to all nodes or running a command on all nodes to save you logging into each node one at a time.Something a simple as ( for nodes in 1 2 3 4 5 6; do; scp $1 yourvmname$nodes:$1 ; done) will save a lot of time.
Set up Networking and Name resolution for all the nodes and Isilon (use SmartConnect)
Turn off or edit the IPTABLES setting so the nodes get access to the various ports used by Hadoop
I needed to update the Openssl package as the Hadoop install process fails quite a few steps along the way and you may run into other issue if you restart the process again. (# yum update openssl)
Disable the transparent huge pages (edit the /boot/grub/grub.config file and reboot)
Set up password less root access for the Ambari server to the other compute nodes in the cluster
The only real changes during the Ambari based install process occur during the initial set up as per below:
Add all the compute and Master nodes into the install process and use the ssh key.
Go to the next page so they all are registered and install the Ambari client.
Then press the back button, add the Isilon FQDN to the list with a manual (not ssh login) and then continue.
Later, during the services/node selection process, just have the NameNode and Datanode services on the Isilon only.
Just follow the install process (change the repository to a local one if you set that up) I did, as my link to the remote repositories was limited to around 500k so it took ages to install multiple nodes without the local options.
I now have two Hadoop clusters up and running and using Isilon as the HDFS store so more to play with 🙂
Life has been busy recently, I have been on a couple of work trips, great trip to Spitzingsee Lake in Schliersee, Germany not far from Munich for an EMEA team meeting. Great venue even if it only stopped raining on the last day.
As per usual, DIY, Gardening and even the occasional trip to a car show took place and kept me out of the garage. I did snatch a few hours here and there. I had refitted the refurbished steering rack but a much longer task was running the fuel pipes over the rear suspension cage.
Rear Suspension Cage Drop
To run the fuel feed and return lines back into the spare wheel area in the boot I had to run a the fuel pipes over the rear suspension cage. This meant I had to drop the whole rear suspension cage. This is not too much of an issue as there are only eight bolts holding the four mounting bushes to the chassis and then the bolts holding on the rear trailing arms. The full suspension cage is really, really heavy so if you need to do this please take extra care. I managed to lower it without incident and ran the two pipes over the top of the cage area exiting via an existing hole in the boot. I also took the chance to change one of the small trailing arm bushes whilst it was off as well. No drama so not really worth writing up specifically. During the refitting of the cage I discovered that one of the four Metalastic mounting bushes had come apart and needed changing. As I had four new bushes on a shelf I decided to change them all. I had refitted the suspension about 3 years ago and the car has not really moved since then so it was just time based deterioration of the bush as they were all OK when I last looked at them.
Changing the bushes is not that hard, just a little awkward to get to some of the nuts inside the cage. It is however a bit of a pig getting the holes lined up when reassembling things. My recommendation is to loosely fit the bushes to the cage, offer the cage up to the chassis and only tighten the bolts once you have managed to get the bolts that go through the chassis pushed all the way through and nuts screwed on. Trying to align the bolt holes through the chassis with the bushes firmly attached to the cage is not going to work! I am speaking from experience here so don’t waste the time trying to align it with jacks, using leverage and so on.
The failed bush is shown below with a little sideways pressure is applied. The faults are highlighted with the arrows. The light brown side has come away from the inverted “V” shape in the bracket. The right hand rubber has also come away from the outer bracket.
Its all reassembled now and back on its wheels.
The twin tanks in the S1,S2 and S3 models are notorious for rusting away and leaking. I did have a spare tank in the garage. It came from an old Series 3 car that I had scrapped probably 15 years ago. The fuel tank had been stashed away waiting to go into the car to replace the rusty/bodged repaired ones that are currently in the car. The S3 tanks have the low fuel drain that I will need for the single external pump option I am going to go with. Unfortunately even though I stored this tank with the view to use it, on close inspection I discovered it does have some pin holes up at the top around the fuelling hole. The pipe assembly appears to be braised in place so no chance of just welding up the holes and I don’t have the facility to braise. I will probably just put some kind of epoxy or liquid metal over the area as a temporary fix whilst I find the funds to replace both tanks with new ones. (approx £250 each)
Fitting this tank temporarily, running the outlet pipes into the boot area and searching for the switch-over valves are the next things on the TODO list.
I decided to rebuild the steering rack whilst I had easy access to it with the engine out. The bushes were “cream crackered” as can be seen by the photos. I had some replacement Poly based bushes that came with a car I purchased years ago. The bushes are in fine condition so I used those. The old bushes were pushed out using a couple of sockets and a threaded bar to get it moving and eventually gently tapped out. The new ones were pushed back in and I am just awaiting delivery of the rack rebuild kit.
The first challenge was to identify the rack type. There are a few variations on steering racks over the years and it was not straight forward to identify the one straight away. I did a little searching on the internet and through the parts/service manuals and got the impression I had the Adwest manufactured part. A close look at the steering rack once it was cleaned up and on the bench revealed a few words and numbers cast into the rack. You can see these in the photos below:
I found the following identifiable words/numbers POW-A-RAK and three numbers HBE12251, HBE 12241 and HBE 12211. A little more searching on these and I found this page which lists the casting numbers as ADWEST CAST NO HBD12241- HBE12251- HB12211.
XJ6/12 80 cm Short Pinion Series 2 & 3 Adwest HBD 12211, 12241, HBE 12251 & HBE 16400
Compare the pictures with those in the parts manual and the repair kit part number is AAU1503. I have ordered a repair kit and new bellows to finish off the job. As it happens I also have a spare steering rack from an older S2 XJ6 which is the Alder type so I should be able to rebuilt at least one of then regardless 🙂
As previously mentioned, I planned to fit the fuel pipes today having finally decided how/what I was going to do!
I did spend at least an hour or three looking through cupboards and boxes for the Series 3 valves. I gave up and started working on the pipes. I did find some of the S3 fuel system parts as per below, but not the valves. I know I have them somewhere as I can picture moving them a few times but they are hiding at the moment.
I did then work out a route with very little modification to the body save a few screw holes and a minor trim of a flange along the floor. The two pipes are fixed approximately every 12 inches using either modified (cut smaller) versions of the XJR pipe clips or new Stainless rubber covered P clips. There is still a little tidying up to do to make them look a little straighter but they are in place.
I have made up a couple of pipes to go up and over the rear suspension cage and into the spare wheel space in the boot. I will have to drop the cage down a little to install them so I did not manage to get to that today. The pipes in the photos and the two going over the suspension cage will be joined using short rubber pipe sections. The copper ends of the pipes should exit into the boot area. There is quite a large round grommet already there which I will try to use but if unsuccessful there is a large rectangular removable panel that I could put the pipes through if needed. I think these panels are there to assist in changing the awful handbrake pads which are almost impossible to change without dropping the suspension cage.
After some research, discussions, Email and yesterday over the phone with my Nephew who is a car mechanic and has done numerous engine swaps, I have come to the decision to Keep It Simple Stupid or KISS! In essence, I will replace the two internal pumps, one in each tank, with a pick up pipe and then feed the two outlets via a three way valve to a single, high performance pump. The tanks already have their own vents. This is pretty much how a Series 3 XJ6/12 fuel system is designed. I am sure I have the appropriate S2/S3 valves “somewhere” in the garage. It should end up looking something like below:
Run two new fuel lines from the engine bay back into the boot area
The original fuel lines ran down the drivers side. The XJR lines ran down the passenger side. This means I have to find a new route that has some protection from the heat of the exhaust, the potential for damage from the drive line and and of course the ground such as speed bumps etc.
The biggest challenge will be getting the lines over the rear suspension cage and into the boot area
These two lines will be the fuel out and return lines and will use new copper fuel pipes plus the rubber end pieces from the original fuel lines so they connect into the engine fuel rail as per before.
Remove the two internal pumps from the two XJ6 tanks and make up a pick up pipe for each tank that will collect the fuel from the lowest area of each tank.
Each of those pipes will go via the original three way Switch Over Valve to a new high performance pump.
The outlet of the new pump will go to a filter
The fuel filter will be the the standard XJR filter (which I have a couple of new ones already)
Then the fuel out line will connect to the filter on onto the engine bay
The return lines will go back to the appropriate tank via the original Switch Over Valves.
Thanks to James G. and Richard B. for their knowledgable input.
I have been trying to decide on the requirements for the fuel system needed for the Super Charged engine. It had two fuel pumps in the tank, the second one came on around 4000RPM. On top of that, there is a vapour control system with a carbon canister and various valves. There is also a small pump and a release valve in the engine bay. I have read that there is a test carried out by the ECU after about 5 minutes of running where the vales are closed/opened to monitor the pressure changes in the fuel tank. This could be a problem unless I manage to include all the components somehow. I don’t know what the end result of this test failing will have of the engine running. Will it just generate an error code and carry on or cause other issues?
The fuel lines required for the XJR setup consisted of 3 separate 10mm fuel pipes and a smaller vacuum pipe that ran from a valve next to the Carbon canister all the way to the engine bay. I did remove the old pipes from the XJR however being steel and obviously not designed to fit in the S2 XJ6 they would need major re-bending and after a couple of attempts to adjust them I decided to purchase new copper based pipes all round. These will also match the new brake lines I have run too. I have also ordered some rubber covered stainless P-Clips to support the pipes along the underside of the car.
OK, so we will have new fuel lines and I have found a suitable location to fit the rectangular Carbon canister which is under the car roughly below the rear passenger seat area. That then leaves me with the issue of the two existing XJ6 fuel tanks, changeover valves and the need for a high pressure fuel feed.
My initial thoughts were to do away with the internal pumps in the two fuel tanks, feed the fuel lines via the change over valves to the new high pressure pump and then onto the engine.
I did try putting the XJR tank into the boot of the XJ6 but the boot hinges are in the way so it would need a lot of work to get the fuel filler pipe to one of the original filler caps.
I read on the Jaguar forum, a post by Larry Lowden, who suggested a smaller “surge” tank in the boot fed by the original pumps and still using the twin fuel tanks. The surge tank would be connected to a high pressure pump to feed the higher requirements of the engine. I doubt that with today’s level of traffic, I could sustain high revs long enough to empty the surge tank faster than the standard pumps can refill it if it was say over a gallon or so in size.
This means I need to have find a specially made up tank or maybe I could do something with the XJR one.
I thought about getting the XJR tank cut down to size? I took the round access panel off of the tank and looked inside. The two pumps sit inside a circular anti swirl assembly and there are two baffles either side of them about a foot apart. I could therefore get the tank reduced in size either side of the baffles or even just remove the side that has the filler pipe/cap assembly. That would reduce the overall capacity of the tank to around two thirds of its original size whilst resolving the sealing off of the filler pipe and retaining all of the original functionality of the pressure testing, dual fuel pumps and so on.
There is a local company I have used in the past to make up one off metal work items I could ask if they would be prepared to work on the tank. There is a real threat of explosion as the tank has had fuel in it, in fact it still has a small amount of fuel in it right now so a major safety issue. My welding is not good enough to make a fuel proof seal anyway so I have to get some one to do it for me.
The alternative to modifying the XJR tank is to look into the racing type fuel tanks that are available.
Looks like some research is in order, where would we be without the internet 🙂