After more driving today I'm starting to suspect that there could actually be an intermittent fault with the torque reduction control line from the gearbox ECU to the engine ECU after all. This is a dedicated line between the two ECU's that from memory sends either a high/low or simple PWM signal.
The purpose of it is that just before a gearchange the gearbox ECU signals to the engine ECU to cut torque dramatically, which it does by retarding the ignition timing by about 20 degrees. So even though you keep your foot down during the gearchange it's as if you'd lifted your foot.
The gearbox ECU then ramps the pressure to the new clutch up and the pressure to the old clutch down and after the gear change completes (about half a second) it signals to restore engine torque to normal. This is so that during that period of clutch slip and overlap the engine torque is reduced to reduce strain and wear on the clutches and give a smoother change.
This feature gives this auto box a distinctive "manual-like" gear change characteristic under strong acceleration where it sounds like you've lifted your foot off the throttle for the gear change, both for up shifts and also kickdowns, which have a sort of double declutching sound to them when it's working properly.
When you're accelerating under throttle and you get a gear change you normally hear that half second lull in the "engine power note" where the engine becomes a lot quieter because it's not putting out much power. What I've noticed is that when I get the really blatent flares during an upshift the engine note does not quieten down like it normally does, it remains noisy like it's still putting out full torque demanded by the throttle. Busted!
Likewise during a kickdown it sometimes lacks the "double declutch" sound where you should get a momentary burst of engine note when you first mash the throttle, followed by a lull in engine note for half a second as it changes gears then the engine note again.
Full torque during the gearchange is very likely to case the engine rpm to flare up and certainly won't be conducive to a smooth gearchange, and if it's happening all the time will eventually wear the clutches prematurely as they're not designed for full power during the slipping/crossover period.
This would all sound quite far fetched if it wasn't for the fact that Citrojim already fixed an almost identical problem on Trainmains ill fated Red Xantia V6! I don't recall the exact symptoms but I think the gearbox going into limp mode due to clutch slip may have been one of them - clutch slip provoked by constantly changing gears without the normal engine torque reduction operating.
From memory Jim found that there isn't a single wire run between the ECU's, but that they are two separate wires with a poor quality crimp connection between them part way along inside the wiring loom trunking, and that crimped joint had become corroded and intermittent, thus the intermittent nature of Trainmain's V6 problem.
Unfortunately that's all I can remember of that issue, as it was about 5 years ago now - I can't remember if Jim ever detailed exactly the path the wire runs and where in the loom the spice is he repaired, or for that matter whether the details were discussed in Citrojim's blog or Trainmans blog. Can anyone help me find the posts in question ? They date back to approximately 5 years ago. Hopefully Citrojim and/or Trainman are reading this and might rembemer!
I think it's well worth me trying to bypass this wire - rather than trying to find the splice and repair it, I'd run a completely new wire through the trunking between the ECU's. I already know I can thread another wire into the cable bundle going into the engine ECU connector and solder it to the pin on the inside of the plug as the connector can be disassembled and I've done that before. I'm not sure whether the same can be done on the gearbox ECU connector (the ECU is under the battery tray) and whether that connector can be disassembled but I would think so.
The route that I suspect the original wire follows, or at least the route
I would follow for my replacement wire is below:
Looks like a pretty easy run of trunking to thread a new wire into - once the battery and battery tray are removed (which is necessary to get to the gearbox ECU connector hidden underneath) and the air filter and top and side plastic covers are removed from the engine the entire trunking run should be fully visible from end to end.
Before I can do so I need to check the wiring diagrams to find exactly which pin number at the engine ECU end connects to which pin number at the gearbox ECU end. Unfortunately Sedre only covers the Series 2 Xantia and I am
not sure that the S1 V6 wiring between the ECU's is identical - it's quite unlikely in fact when they use different immobilisers! (Thus the pinouts for the engine ECU are almost certainly different)
So is anyone able to help me with the wiring diagrams for both engine ECU and gearbox ECU for a 1997 S1 Xantia V6 so I can positively identify the torque control line ?
For the sake of replacing a wire, it seems like a good idea to at least rule it out as a posible contributor to the gearbox problem.
Just not at work where I am at the moment...
Pitch black, thin and runny like water compared to new oil, and when I caught a 1 litre sample in a measuring jug as it was draining and tipped it out I found significant quantity of particulate matter:
