Why couldn't this state of the art workshop fit the bits that you had to get MM to fit, and by the sounds of it, the bits that they did fit, were not fitted that well

 

Am i right in saying the "sensor issue" on all occasions was it throwing up that Rich code ?

If all 3 sensors flung up the same code i don't think it's a sensor issue,

Could be another sensor or the map overfueling and "killing" the sensors

Or is it intermittently working ?

I would invest in a Tactrix cable or Vag-Com

Download RomRaider logger

Do a few runs logging

Post em up and let us keyboard techs diagnose it from there.....

 

Got a feeling someone would be eating a big slice of pie by now after 8 pages + 6k views

Seems like OP is gonna shady off so MM + Bob dont actually find out what is wrong with it, then cover up real issue with some spiel on here about how he found it and its now working yet its MM+Bobs fault still

 

vag-com on a subaru? ****ing hell this place - no wonder people have **** thats broken after been on here

 

For logging not mapping ??

Is there something wrong with that ??

Do you even have a clue what I am talking about ??

 

No wonder the folk who give decent advise stay quiet and all thats left on here is B.S and poor advise

https://forums.nasioc.com/forums/sho...1914025&page=4

Do a little research you may learn a thing or two..........

 

yes - a cheapo vagcom lead with the opensource romraider software will give you access to realtime ECU data that can be very useful in diagnosing a fault.

 

Also handy because even if your not an expert can send data to folk like yourself and others who know what their looking for,

On that thread it says you can flash with Vag-Com cable too

Need to play around with settings tho

 

Nonsense they had car 3 times with sensor like that

 

Fit 1000cc injectors a 4 bar map sensor a fuel pump
And a 3 port and drive it 272 miles dude sit down and have a rest

 

yeh *some* of the cables could flash *some* cars but it's not worth the chance unless you've a spare ECU sitting about and time to dismantle your vag com lead. For the price of an openport if you want to flash your own car then that's the cable to buy.

The ECUTek ECUConnect dongle can do logging as well now along with virtual dyno and dash displays so worth considering if you've an ECUTek mapped car.

 

Bob and mm handed me the car back 3 times knowing there was a sensor issue what’s your point?
My mechanic can’t find a mechanical issue
Seems neither can murray Motorsport
And bob can’t find a mapping issue
Somethings the matter would you think not?

 

To reiterate my car was starting driving 100 percent perfect prior to the first map
It has never been the same since
I drove it up north with turbo and intercooler on off boost car dove like a dream
After bob mapped it that’s when my issues started
And it’s never been the same since
Quite straightforward

 

I learn a lot from your posts - I wish there was more input from the mappers. I think it's a real shame that esl don't post anymore.

 

thanks! I try and help where I can - the ESL facebook page is pretty active and I think that's where Andy spends time to respond to technical questions now rather than jumping in and out of posts here.

 

My first input on here. I own a Scooby Hawk, but my other (****ty french)car has been a nightmare and taught me a lot. So,

1. Does the JDM Hatch have a CANBUS network? If so, have you discounted that the Lamda sensor hasn't been hand shaked with the car? My French c**per is very weird when you change sensors if the car hasn't been coded with them using the correct Manufacturer's diagnostic setup.

Thus have you taken the car at any point to an authorised Subaru dealer to have it diagnosed? I can do all sorts of stuff with my own dongle, but my DPF codes keep coming back, despite the system now working, unless I get the new sensor talking properly to the ECU. CAN networks are bloody strange things.

Other than that, I can't think of what to suggest to help. Checking third party parts onto a customer's car must be a perennial nightmare for tuners. Mismatching, hidden problems, you name it.

You do your research, assume it will all work together, and try your best to get everything at the keenest price, who wouldn't? Any of us could therefore be in your position i guess. Not that I'd have gone public about it liek this.

Anyway, without any finger pointing, CANBUS. It's the only logical reason why a new sensor doesn't work, despite testing OK, and if the wiring is also OK. You'd think the O.>E sensor would go in no problem, but these EURO4 complient CAN systems are not like logical in any way!

Best of luck, but keep an open mind. Bob mapped the car as best he could at the time, and the fault, when it happened, he quite logically asssumed you would get a new sensor fitted, and mapped it with the data he got prior to the fault. I would do and expect the same.

Modern cars are a pain. I'm lucky, the Subaru Dealer near where I work has very modest labour rates. All the special tools and gear, so for something like this, i'd book it in without a moment's hesitation.

Open mind and best of luck.

 

There's no coding involved - canbus isn't used like that for the O2 sensor. However he's only mentioned testing the heater circuit not the amps coming to the ECU for signal, they are two different circuits and need seperatly tested to confirm the sensor is good.

 

Also, it's a jdm, so UK dealers do not have the correct diagnostics for it

 

Shouldn't it continuously go up and down between 0.1 and 0.9V, by observing with a simple multimeter?

 

Thats the rear o2 your thinking of, front one is based on ma reading for scale if I remember rightly so different readings expected.

 

I getcha.

 

aye - just checked so front o2 uses mA readings to determine AFR reading, that's where your cheapy vag com lead comes in as you can read what the ECU thinks it's getting from the sensor as well as the resistance reading. Sometimes when the sensor dies it will just feed a continuous lean or rich reading to the ECU causing it to continually add or remove fuel trying to reach stoich/14.7 which of course it can't do. Once the corrections max out the car will become quite rich or quite lean causing poor running even on a stock car. Once you add giant injectors into the mix you can end up with a difficult to drive bus until the problem is sorted. Or the sensor dies out and has Mr Rawle has already said the ECU goes into limp mode which will instantly make things richer than they should be, but again the issue should be rectified when the new sensor goes in and the ECU is reset. Unless there's a fault with the ECU, sensor or wiring between the two in which case again the ECU makes the wrong decision until the situation is rectified.
08 sti front o2 sensor scale.

 

Thanks for that graph.Shows clearly what the sensors do.
That being the case,if there was a problem with the AFR sensor,which I assume is absolutely critical & a basic parameter to the correct running of the engine, why wasn't any further work on the engine brought to a stop until the problem was corrected?

So,when was the faulty sensor/sensor reading to the ECU noticed?
Was any further work done with a known basic fault?

 

even with a faulty lambda the mapping session should have went ok. i seem to remember when one of my cars was mapped by AF (cant rem if it was the wagon or the bug) they screwed up the threads on my lambda taking it out of my old manifold to put into the ported and wrapped ones i wanted fitted but andy said that would be fine as he would be using his wideband anyway for readings and we went ahead and it was fine. i drove around a few days while i waited for a new lamda to turn up and fitted it and no probs whatsoever.

 

yes that was you driving with no lambda on a 16bit ECU. 32bit drive by wire cars are a different beast.

 

ahh ok. never realised they were so different (hence the problems im guessing)

 

...which is why I mentioned CAN issues. I know older cars just look at voltage and current. Post 2006 DBW cars are all EOBD2 plus loads of other tricks for complience and JDM cars will no doubt have other hurdles to pass for domestic approval too.

I fly for a living and in aviation we say there's no such a thing as a silly question....

 

I thought BR had told someone with a spec C hatch on here that the ecu was 16 bit. It would be good to have a definitive answer, as I may have got the wrong end of the stick

 

JDM hawk would be 16bit - drive by wire hatch 32bit if i recall correctly.

 

I have to say i have just sat and read 9 pages .... great read its like emmerdale but on a forum.

I am fairly new to this forum but over the years i have had many Japanese high powered (mapped cars).

I read this simply, even after all the different avenues the thread has takeN.

1. You and your "mechanic" fitted parts but took the injectors fueling parts with you as clearly you could not drive the car with those fitted without a map understand that.

2. Parts were fitted there by MM car was mapped you were happy car left the premises you paid.

3. Months went by car was fine then develops a fault you blame mapper ? cars go wrong all the time things happen.

4. You and your "mechanic" try to fix problem cant find so still blaming mapper.

5. Car goes back stays there 8 WEEKS way to long really why was it not fixed then ?

6. You go collect car you pay anyway and take the car why ?

7. I could go on ..........


In my past experiences despite people saying different on here its best to do all the work yourself then go to your preffered mapper for just mapping OR give it all to the garage/mapper combined to do all the work.

Blame is much easier to place then ........you have so many people involved it will never get worked out and know one will really know where to place blame and gain any compensation.

Conclusion......either take it to a new "mechanic" down your way have you thought of that to get a 2nd opinion rather than your usual guy. Take it back to Bob as he has offered or just bite the bullet get syvecs on it and move on as you cant place blame after you fit that ECU as you'll never know what the true problem was.