Richard Hamilton

I think you need to have the extended features option on PCM2.1 to be able to adjust the individual memory settings yourself.  Oddly, it came with the Sport Chrono package.  If you have it, the option is in Main>Set.  Otherwise, you have to change the individual options for each key with a PIWIS tester.

It sounds like you have the I666 Telephone Module option.  There are 2 ways to enable HFP:

1) With PIWIS, code the PCM to Telephone Module "Not Installed".  This has the effect of making the system behave like the I619 Mobile Phone Preparation option, which enables HFP Protocol.

2) Update the system to V2.23 and V2.24 (a 2-stage update with CDs).  However, if you do this, you will also need to have the maps updated.

The transponder and radio remote are two separate elements.  Perhaps they are confusing it with a 997, where the alarm module reads the radio remote during programming. The 996 alarm module doesn't do that.  You have to code the module manually. 

Unless things have changed with PIWIS3, but I don't see how it would be possible. Let us know if they were able to do it.

A used circuit board will be no use to you.  New 9x6 keys (with circuit board) come with a bar code label with a 24 digit code printed on it.  That code is entered into the alarm module with a PST2 or PIWIS tester.  If you haven't got the code, the circuit board is useless.

The navigation system update is done by DVDs in the head unit.  It depends on the initial software version you have, but there are up to 3 updates required before the latest maps will run.  The software updates can take up to an hour to run.  (Note that after the updates, old map data will not run).  After the system is updated, an Activation Code is entered using PIWIS2 or PIWIS3.   (The Activation Code is based on the VIN, and comes from the Porsche online PIWIS system).  Then the map data is loaded from DVDs.  On PCM3.0 it takes around 40 minutes per map DVD.  In Europe there are 4 DVDs.  PCM3.1 updates are quicker, as they have a faster DVD drive. 

You have to code it in the instrument cluster, not the PCM.  Here are the coding steps when retrofitting PCM2.0:

8 Programming PCM2 with PST2
8.1 Activate PCM2.
8.2 Read out PCM2 fault memory and erase.
8.3 Match the MOST required installed components list to the actual installed components list.
8.4 Code the PCM2 according to vehicle type, versions, country-specific tuner settings and MOST switch.
8.5 Check MOST components.
8.6 Activate navigation in the "instrument cluster" control module system.
8.7 After programming is finished, read out the fault memory of all systems, remedy any existing faults, and erase fault memory.

Was it a 52 plate?  If so, it might well be MY2003.

You should be able to connect to the DME with a code reader, even if the immobilizer is active. 

Mike is correct that DME7.8 was introduced for MY2003.  If the engine was post-September 2002, it would be MY2003, but it might be worth checking the engine number, just to be sure.

If it was a 2003 engine the third digit would be "3".

Did you use the immobiliser pill from the new DME/Alarm/Key set?  The alarm module needs to read a pill which was coded to a matching DME and Alarm Module.

Also, if you can't connect to the DME with a code reader, check the W-Line diagnostic wire between the alarm module and DME.  It uses the diagnostic line to transmit the immobiliser signal.

Just to clarify, for anyone reading this in the future, on a 996 you can't code the lights for DRLs.  All the lighting is hard wired, and there are no control units involved.  The 9x7 series introduced the Front Control Unit for the lights (and other items) and the DRLs can be coded in that.

It could be a problem with the brake light switch or clutch switch.  There are two sets of contacts in the switches, so although the brake lights and clutch interlock may work, it doesn't mean the switches are OK.  You can check both contacts for each switch in DME > Input Signals, using a Porsche tester.  Also worth checking the cruise switch Input Signals in the Instrument Cluster on the tester at the same time.

PET shows the instrument cluster bulbs only for cars up to 2001.  You are also right that the 2002 doesn't need an additional wire from the DME to the cluster - only wires from the stalk to the cluster.

It sounds like it might be a coding issue to me, and I think you should track down someone with a PST2 or PIWIS.  Get them to check the Order Code is saved in the DME Vehicle Data, and then re-recode both the DME and cluster.

On the 986 the CDR23 used the speaker outputs for the basic system.  If an amp was fitted (Harman or Bose) it used the fibre optic output which connects the head unit, CD changer (if fitted) and amp in a loop.

It was the CDR22 which used the yellow plug for the line outputs for the amp.

They aren't line outputs on a CDR23.  The 4 wires connected to the yellow plug are:

Yellow - CAN Comfort High

Black - CAN Comfort Low

White/Red - Wake up/Diagnosis Line

Green - Key Code In

The KTS520 does not click, click, click when you connect to a 16bit car (986/996, etc).  The Bosch serial cable is not wired the same as a standard serial cable.  Ubox2 is not what you need.

It is probably best to continue our conversation by PM and email. Check your PM inbox.

When you run PIWIS diagnostics, does it try to connect to the KTS520, or do straight to simulation mode?

Does the computer recognise the KTS520 when you plug it into the USB port?  (Does it make the sound?)  If so, right-click My Computer, Properties, Hardware, Device Manager, and see if you can see a Human Interface Device.  It won't be labelled KTS520. 

If the KTS is connecting to the computer it might be a firmware issue.  To update the firmware, you should try doing it from within PIWIS first.  Go to Tools > Firmware Update.  The firmware flashing process should take 2-3 minutes.  If the Firmware Update screen appears for a couple of seconds and then disappears, the flash was unsuccessful.  In that case, try updating it with the Firmware tool.  Run RBPLFWconfig in your Firmware folder.  Choose KTS520, USB connection, ForceUpdate, and Firmware version IP09.  A COM1 connection would be better, but you need the special Bosch serial cable to use it.  If it flashes correctly, you will see a progress counter.

Good call DBJoe.  I hadn't noticed that.  Check fuses C1 and B8 to start with.

Two possibilities I can think of:

Firstly, there is a break in the diagnostic ("W" Line) lead between the DME and Alarm control unit.  This wire carries the diagnostic and immobiliser signals.  If the wire is broken, you can't communicate to with the DME via the diagnostic socket, nor can the Alarm unit connect to the DME to pass the immobiliser signal.  There is a plug and socket joint between the ACU and DME (Plug X2/3 - 30 pins) in the rear left passenger footwell.  I have known a case where the pins were corroded, and gave exactly the symptoms you describe.

Second possibility - the DME has failed.  I have know that happen too.

Has it been running with the flashed DME, or has this happened immediately after the flash?

AS JFP says, there are several options for DRLs which can be set with a PIWIS tester in the Front Control Module.  Here are a couple of the coding screens for the 997.2 in PIWIS.  The 997.1 will be similar, but it should give you an idea of the possibilities:

As far as removing the seat is concerned, you won't have any airbag problems with a UK 996.  The seat buckles are not connected to the airbag control unit like they are on a US 996.

Edit: Also, before disconnecting the battery, you can switch the ignition on, and then switch it off again, but you must leave the key in the ignition.  The alarm siren is signalled by the loss of the Terminal 86S supply, and this supply remains on until you remove the key.

It should be coded to the car, unless the security coding has been removed by a specialist.

I would email Durametric, to see what they can suggest, but I think you would be best finding someone with a PST2 or PIWIS tester.  It should not matter if the wiring & stalk are in place or not, you should be able to code it.  I have done it both ways, and it has always accepted the coding.  However, I prefer to code it after everything has been wired up, so that I can check all the connections and switches with the Input Signals menu on PIWIS.

You could try Maple Audio in Bath for checking and repairing the head unit. 

Coding is where settings are made in the control units to tell the car which options are installed. Very occasionally, perhaps after a battery disconnect/change, a unit can lose its settings.  The screenshot below is from the Gateway coding page for a car with PCM3, but the principle is the same for PCM2.1.

Another thing they could try is running the Handover Routine with the PIWIS tester.  If any units have slipped into transit mode (again, after a flat battery, etc.) it will wake them up.

However, if you have a wiring issue, it would be better to sort that out first.  Putting the optic loop in place of the phone module would have eliminated that unit from the MOST loop, so I suspect your issue is elsewhere.  Either more than one external unit is at fault, or maybe the suspect earth is on the main PCM unit?  It is unusual to get corroded connectors on these units, but being a cabriolet, I wonder if it has had a leak at some time and the phone module has got wet.

Did PCT check the PCM coding?  It would be worth getting them to check that the "Required" and "Actual" coding values are the same.  And, as has been said before, check that all the installed units are coded in the Gateway module.