Issues only appearing at low ambient temperature?

[otolith]

I had some issues with the Elise last winter, not revving out, power loss, coughing and spluttering. The revs the problem was happening at made me suspect the variable valve timing mechanism. And then it just went away.

Now that the weather has turned cold again, it's suddenly back, but it's happening below the VVTLi changeover point, and it feels a little lumpy at all revs. This is all on a warm engine.

Would this be consistent with an issue with the MAF sensor? I'm not getting any error codes or engine management lights.

[George111]

Air temperature sensor causing it to run too rich ? That would cause it to feel lumpy. If it's only at low ambient temp (not low engine temp) then this is possibly the cause. Have also found a faulty cam sensor didn't work well at low temp but as engine warmed up it was fine.

Could be the maf sensor but not the first thing I'd check.

[otolith]

Ah,interesting, the IAT sensor on this car is part of the MAF sensor.

[otolith]

I did try removing and cleaning the MAF and IAT sensor assembly, to no effect.

When this problem first appeared, I thought that it was related to the variable valve timing (Toyota 2ZZGE in an Elise), because it was happening around the point where the cam change happens, but I'm now pretty sure it isn't. It will rev past the high lift engagement point on part throttle, only bogging when the throttle is opened fully, and I've also discovered that it will happen at motorway speeds in the higher gears at 4000-5000rpm and full throttle, well below the engagement point.

So, coughing and spluttering with full throttle at high revs or medium revs under high load, problem appears to show up in the winter and go away in warmer weather (or perhaps it's just intermittent and decided to go away for 9 months of its own accord).

[ch427]

Has the management light triggered at all throughout this problem even briefly?

[otolith]

Nope.

Well, actually, once, last year, while sitting with the engine idling in heavy traffic on the way to the garage to get it looked at. I think that was a red herring, though. It was a misfire, the code was cleared and it never came back. I've checked for trouble codes with DashCommand since the problem started again, and there are none.

I can also log live OBD data, is there anything that would be worth looking at?

[ch427]

You can look at the maf readings first, expect to see around 3 g/s at idle and roughly your engine horsepower times 0.8 at wide open throttle ie 200 x 0.8= 160 g/s)
can be difficult to monitor unless you have some sort of flight recorder type application that can be done on the move.
Also worth monitoring the iat too.

[otolith]

This is a ~ 3 second snapshot of the MAF data during the problem happening - you can see from the acceleration trace when it chokes;

[otolith]

That's interesting - looking through the data further, that big jump in MAF reading when the problem occurs seems to be consistent.

This one is building the revs up on part throttle and then going WOT, at which point it coughs and splutters and won't accelerate.

[stevesingo]

Is your air filter arrangement stock, i.e no cone filter or such like?

[otolith]

It's a TRD airbox, so an in-house Toyota tuning part which Lotus fits to the supercharged Elises. Fairly common modification to the NA car, doesn't normally cause problems. Takes an ordinary type of filter, not an oiled one.

[stevesingo]

Have you tried the original, i.e eliminate any possible unknowns?

[otolith]

No, don't have it any more. It was changed six or seven months before the trouble started, though.

[ch427]

looks like a big leap in the maf values, what are the figures on the left of the graph in?

[otolith]

g/s - so those 220g/s + figures are clearly spurious.

[wolf1]

You need to test the maf itself with an ociloscope as the ecu will substitute it's own best guess to try and get the engine to run as best as it can. We get it a lot where techs look at O2 sensors on live readings and see nothing wrong. I test them directly and the component is faulty.

[otolith]

You mean those spurious flow values might not be coming from the MAF?

[stevieturbo]

wolf1 said:

You need to test the maf itself with an ociloscope as the ecu will substitute it's own best guess to try and get the engine to run as best as it can. We get it a lot where techs look at O2 sensors on live readings and see nothing wrong. I test them directly and the component is faulty.

it may substitute some values, but if that is a raw MAF reading, then it will be a raw MAF reading.

On some calculated values you might log, these may appear more normal if the ecu does substitute dummy values..

You'd need to also log timing and injector pulse width, to see if they're doing anything odd during the problem

[otolith]

I've ordered a cheap aftermarket MAF, which will probably be crap, but for 30 quid or so should at least show me whether mine is goosed or not.

[George111]

otolith said:

I've ordered a cheap aftermarket MAF, which will probably be crap, but for 30 quid or so should at least show me whether mine is goosed or not.

Looks like it is the MAF sensor ! It appears to be believing there's a lot of air all of a sudden and the ECU is then increasing the fueling to cope but the air reading is probably invalid, so it's going too rich.