This will be a long post, so my thanks in advance to those who read to the end.
The story so far: In December last year the car experienced a temporary loss of drive with "Transmission Limited Function" message. Ford reported there were no trouble codes stored, so were unable to provide a diagnosis. The car also occasionally exhibits weird transmission shifting behaviour usually in 1st/2nd gear, such as harsh engagement (shunting) and momentary disengagement (hesitation or balking) of drive.
In April I bought an OBDLink MX and have been using FORScan on my smartphone to stream and log live data from the car's computer as I drive around. I wanted to be prepared in case the car ever breaks down again so I can see exactly what the error state is. I was also hoping that examining the data might give some insight into the health of the transmission sensors which have been known to fail causing the aberrant shifting behaviour as a symptom. I'm not sure if I've proven anything yet but I wanted to share the following for discussion.
I've been mainly concerned with logging the following PIDs (Parameter IDs) exposed via the
TCM:
- RPM is engine speed, naturally in revs per minute.
- GISS is the Gearbox Input Shaft Speed sensor value and is also in revs/min. This is presumably the shaft driven by the engine through the clutch(es).
- VSS is the gearbox output speed sensor value and is in km/h.
- GEAR_SEL is the currently selected gearbox gear which I've normalised to the range -2 thru 6 (0 for neutral, -1 for reverse, -2 for the explicit error value anything unexpected).
- TGT_GEAR is "Target Gear" which I expected would be the value of the gear that the electronic brain is anticipating will be requested next.
○ We are told one of the features of a dual-clutch transmission is it always has two gears selected, but only one clutch engaged; by pre-selecting the next gear on the other clutch, changes can be lightning fast as it just swaps clutches. In practice though, this PID seems to generally sit at "GEAR_SEL+1" (and even goes to Reverse when the current gear reaches 6th!), preceding the current gear just a little bit, so I'm not too sure what he represents.
○ It could be that Powershift, being less of a bleeding-edge performance product, just omits prediction algorithms for downshifting. Certainly in most driving it's only the upshifts you'll notice, as downshifts are usually masked by braking.
I've been smooshing this all together into composite graphs like the one below. This shows a typical trace of accelerating from (near-)stationary to 80kph, maintaining that speed for a bit, and then braking back down to about 20 kph.
image
At the leftmost end, GISS starts off low (not quite zero as we were crawling in traffic) and rises to meet RPM as the clutch is fed in. You can see the characteristic sawtooth pattern of GISS and RPM as the car shifts through the gears.
Between shifts, GISS matches fairly closely to RPM but I've noticed at 100kph and below there is a discrepancy of about 10-20 revs/min when the transmission is loaded. When I'm accelerating or maintaining speed, engine RPM will be 10-20 higher than GISS, and when coasting and engine braking GISS will go 10-20 above RPM. You can just about see this on my chart, at the areas labelled Accelerating and Coasting. When I get up to 110kph, GISS and RPM become closely matched.
It seems to me that there isn't enough clamping force in the clutch(es) at speeds below 110kph (or more accurately, the corresponding RPM which is almost 2000) and it's slipping the clutch. Does this seem like it could be a concern? I've read that an increase in viscosity of the transmission fluid (from impurities or wrong type) can cause the clutch to slip, but my car is only halfway to its next change (done at 60K, odo 93K). I'm not happy with the thought that it's constantly slipping though so should I be considering an early service?
Apart from the GISS and RPM, there's no aberrant behaviour in the example above. Generally, the Powershift works perfectly for me 95+% of the time. When it is acting up, what I get is stumbling and lurching behaviour, like any of the following:
not in gear when it's needed; drops out of gear momentarily; goes for the wrong gear; shunts roughly into gear.
So the Sunday before last, the car had a very rough day. In six trips, I had gearbox events across three of them.
On the third trip of the day, while moving off from the lights on an uphill incline among
very slow-moving traffic, the car suddenly disengages drive causing the engine to free rev. It then engaged again abruptly; hard enough to chirp the front tyres under power. I've had it happen before, but this was the most uncomfortable.
I looked through the logged data and found what I believe is the record from that event, you can see the RPM spike and a bump in the VSS.
image
On the fifth trip, it seemed to be mis-shifting at every second intersection and traffic light down the main street. I gathered up what I believe to be the three instances I could feel. They are marked by a spike in the RPM and flattening of VSS as the gearbox disengages, followed by RPM going squiggly as it re-engages drive in a usually un-graceful manner.
image
I can't see any evidence in this for why it happens. I thought perhaps I would see GISS going squiggly confirming the sensor going haywire. About all I can say is that it seems unrelated to actual gear changes.
Speaking of the sensor, at day's end it had thrown the diagnostic trouble code P0715 to the
TCM. Interestingly enough, I know from my scan log that it threw the code during the sixth trip when, as far as I was concerned, it was actually changing gears just fine.
This is the third time since April that I've seen (and cleared) the P0715. I'm wondering at what point I need to consider this a problem that'll need to be $$$rectified$$$?
Thoughts?