A remap will cure it for sure, and make the car vastly better to drive. I may still follow this route if a simple like-for-like exchange fails to offer any improvement...

 

Have you unplugged the lambda and the tried it ?

From your descriptions this is normal behaviour, some cars to feel this more than others, if the afr correction is working overtime then that just indicates that the lambda sensor is fine, it would be sluggish or stuck if the sensor was faulty.

Boost will be relative, what were the atmospheric conditions when checked ie atmospheric pressure etc ?

bob

bob

 

Bob,

I haven't tried running the car with the lambda sensor unplugged, but I will do if you believe it will help to diagnose the problem. I can't believe the magnitude of correction seen (and felt) on this car is normal, otherwise Subaru would have a car park full of returned vehicles; to the uninitiated, it feels like a significant misfire.

Assuming that the car appears to run better with the O2 sensor disconnected, what course of action would you recommend?

Point taken on boost pressure, but the car has always felt thin and weedy, so I wouldn't be surprised if it was consistently lower than the majority. Could this be hardware related (wastegate, etc.), or is it likely to be a software issue?

Thanks for your help.

Andy

 

I'd be taking it to Bob for a proper look. Did TSL check and see what the ECu actually was? If they plugged in the logging equipment to it, it will be the existing ECU, but may have been remapped, ie Tek2 etc?

MB

 

Unfortunately, we were both very short of time, so just managed to do a pre/post lambda change test run which showed no improvement.

Previous owners were a senior lady in Suffolk and a middle-age family in Brighton so I don't believe the car has been modded at all - just a standard AE801 as far as I know; what I don't know is which version of the 801 map it is.

Other than an ECU change (ideally for a different iteration of the standard 801 map), or a bespoke re-map, I don't see many options....

 

You have to watch these old ladies This would be a classic if it turned out to be running on a Motec M800 with antilag and launch control

MB

 

If the feeling you describe disapears with ithe lambda sensor disconnected, ie it either drives consistently "ok" or "not ok" then that will determine what to do next. These cars tend to the lean side when running off the "closed loop" portion of the map in "open loop" but the surging feeling should go.

So how much correction was there, plus minus 3%, 10%, 20% ? ECU can add or subtract up to 25% if it wants to however somethings wrong if its going that far.

As far as boost is concerned try pulling off the pipe from the lower port of the boost control solenoid, this is the one that goes back to the inlet pipe, do not remove the side exiting one under any circumstances. If this pipe is restricted it will limit boost, however it could just be that some previous dealer has resolved an overboost problem by fitting an oversize restrictor.

bob

 

Thanks for this Bob, helpful and rational, as ever !!

I could be a mile off the mark, but if memory serves me correctly we were seeing swings of plus/minus <16%, typically 8-12% - I THINK.....

The weather is awful at the mo', but I'm just about to head off for a test drive with an unplugged O2 sensor....

I will also try disconnecting the lower pipe from the boost control solenoid - please excuse my ignorance, but can you tell me what this will show? I would also be interested to know more about the restrictor you mention above; where is it located and are there any hard and fast rules on sizing versus boost pressure?

Many thanks

Andy

 

 

Ok Bob, I've just come back from a 15 mile test drive with elements of town driving and open "A" roads, all with the O2 sensor disconnected. The car felt loads better, definitely pulled much more smoothly and cleanly. I can't be 100% sure, but I also believe it felt a little more urgent than before, but that might be a function of the fact that it felt smoother, rather than any real performance improvement. The other thing that really surprised me was that at no time did the CEL light flash; there was absolutely no indication to suggest that the O2 sensor had been disconnected.

Truth be told, I had planned to remove the boost control solenoid pipe half way round the circuit, but standing under the bonnet in the pouring rain I completely forgot which of the three widgets on the O/S strut tower I was looking for so I played it safe and left all hoses connected - bl**dy idiot .....!

Given the above findings, what do you suggest I do from here? At this moment in time, leaving the O2 sensor off is an appealing option !!

 

An update on the saga!!

Finally managed to get the car to Scoob-meister Rawle in Swindon. Within 5 minutes, Bob identified that the car had been fitted with an oversize boost restrictor (about 1.5mm diameter instead of 1.2mm for this 99 uk car).

On arrival, boost pressure was peaking at 0.8'ish bar and tailing off to 0.6-0.65bar. With the restrictor replaced for the correct item, it immediately improved to 0.9bar - 0.7'ish bar. We tried a 0.9mm STi boost restrictor (as you do ;-) and the car immediately overboosted (- if that's a word??!), to a peak of 1.1 bar but not so much as to trigger the fuel cut-off (@1.2 bar IIRC). For obvious reasons, we then reverted to the 1.2mm restrictor.

Following Bob's advice, I have since tightened the wastegate actuator rod by one full turn, and plan to tighten it a further turn as and when time permits. I don't yet have a boost gauge, but it is clearly evident that these changes have made an enormous difference to the perfomance and feel of the car. A boost gauge will help me to optimise the actuator rod adjustment and should rectify the running problems permanently.

Incidentally, it's still early days, but the lumpiness previously experienced under light load appears to have been all-but erradicated. Could this be as a result of insufficient closing force being applied to the wastegate, causing it to flutter & vent pressure leading to unsteady turbo spin-up and consequent pulsation?

My thanks go to Bob for taking precious time out of his already manic schedule. What he doesn't know about Subarus isn't worth knowing.