STI 7 : important information about engine failure
06 April 2003, 10:43 AM
#31
Scooby Regular
Join Date: May 2001
Posts: 720
Likes: 0
Received 0 Likes
on
0 Posts
Xavier,
Yeah - Thats why I asked earlier if it was the Euro or JDM versions that we exhibiting this strange phenomenon. I've had my car a while now, and I've never noticed it do anything like that. (Fingers still crossed tho.)
I also agree with your fuel pump point. If its a genuine fuel pump issue, it will be most apparent at WOT.
Ray,
I'm no expert with DD, never had a chance to use it. But I would think you would log rpms, throttle position, AF ratio (or lambda ratio) and boost as a minimum. You should be able to tell if there's a trend on the AF ratio even with the factory 'narrowband' sensor, if it tends back to 14.7.
Yeah - Thats why I asked earlier if it was the Euro or JDM versions that we exhibiting this strange phenomenon. I've had my car a while now, and I've never noticed it do anything like that. (Fingers still crossed tho.)
I also agree with your fuel pump point. If its a genuine fuel pump issue, it will be most apparent at WOT.
Ray,
I'm no expert with DD, never had a chance to use it. But I would think you would log rpms, throttle position, AF ratio (or lambda ratio) and boost as a minimum. You should be able to tell if there's a trend on the AF ratio even with the factory 'narrowband' sensor, if it tends back to 14.7.
06 April 2003, 03:32 PM
#34
Scooby Senior
Join Date: Oct 2000
Location: Zurich, Switzerland
Posts: 3,105
Likes: 0
Received 0 Likes
on
0 Posts
LHD/RHD shows no difference in the maps...not that I can actually *read* a Euro LHD rom though..but the RHD roms work fine in a LHD car 
What version of code on the cars you've logged?
Richard
What version of code on the cars you've logged?
Richard
06 April 2003, 10:23 PM
#37
Scooby Regular
Join Date: Sep 2000
Location: London
Posts: 7,039
Likes: 0
Received 0 Likes
on
0 Posts
I don't think the JDM ECU's are any different or "superior" to the Euro spec ones, they just have different maps in them, which are designed for higher octane fuel....
Matt
Matt
07 April 2003, 09:30 AM
#38
Scooby Regular
Join Date: Sep 2000
Location: London
Posts: 7,039
Likes: 0
Received 0 Likes
on
0 Posts
Here is a reply from Stephen Done:
"The lambda sensor in the 2001-> Impreza is accurate around stoich mixtures, but not that good when things start getting richer. It also cannot measure richer than around 0.77 lambda or 11.3:1. This varies slightly with atmospheric pressure, so I can't be more specific about its limits without getting more complicated.
For our mapping, we use Motec PLM systems. These also have an analogue output that allows them to be connected to DeltaDash for logging along with other engine parameters. See external sensor logging section on the EcuTek web site.
The ECU will switch from closed loop to open loop based on throttle
position. If you have derestricted the exhaust so that small trottle openings produce more boost, then you will see weak fixtures at high engine loads.
{shameless_plug}
A remap allows the throttle opening at which open loop (non-stoich i.e. richer) fuelling can begin. This map is based on RPM, so you can say that (for example) >30% throttle at 3000 RPM causes open loop, dropping to 10% at 5000 RPM if that's what you wanted.
{/shameless_plug}
Scooby50WRX does not state the spec. of his car. If it is totally standard and he sees this condition, then there _may_ be a general problem. If he has changed exhaust components for example, then he has caused this. Whilst the standard ECU is highly capable, you cannot expect it to cope perfectly if parts of the rest of the system are changed without recalibrating the ECU accordingly. This is just such a situation. In fact, our development car would run 16 PSI (1.1 bar ish) of boost at stoich mixtures, if we had not
altered the ECU calibration.
A faulty throttle pot could also be the cause. He should check (via a select monitor or deltadash) that his range of throttle values is 0-100% according to pedal position. Rule this out before anything else."
Matt
"The lambda sensor in the 2001-> Impreza is accurate around stoich mixtures, but not that good when things start getting richer. It also cannot measure richer than around 0.77 lambda or 11.3:1. This varies slightly with atmospheric pressure, so I can't be more specific about its limits without getting more complicated.
For our mapping, we use Motec PLM systems. These also have an analogue output that allows them to be connected to DeltaDash for logging along with other engine parameters. See external sensor logging section on the EcuTek web site.
The ECU will switch from closed loop to open loop based on throttle
position. If you have derestricted the exhaust so that small trottle openings produce more boost, then you will see weak fixtures at high engine loads.
{shameless_plug}
A remap allows the throttle opening at which open loop (non-stoich i.e. richer) fuelling can begin. This map is based on RPM, so you can say that (for example) >30% throttle at 3000 RPM causes open loop, dropping to 10% at 5000 RPM if that's what you wanted.
{/shameless_plug}
Scooby50WRX does not state the spec. of his car. If it is totally standard and he sees this condition, then there _may_ be a general problem. If he has changed exhaust components for example, then he has caused this. Whilst the standard ECU is highly capable, you cannot expect it to cope perfectly if parts of the rest of the system are changed without recalibrating the ECU accordingly. This is just such a situation. In fact, our development car would run 16 PSI (1.1 bar ish) of boost at stoich mixtures, if we had not
altered the ECU calibration.
A faulty throttle pot could also be the cause. He should check (via a select monitor or deltadash) that his range of throttle values is 0-100% according to pedal position. Rule this out before anything else."
Matt
07 April 2003, 11:32 AM
#39
Scooby Regular
Join Date: Feb 2002
Posts: 1,343
Likes: 0
Received 0 Likes
on
0 Posts
matt,
I am just going on what tuners have told me, I think the ecu's are indeed different, not only the maps but the actual ecu itself.....and as I said before the JDM is supposibly superior. Ask some of the mappers that have done Uk, Euro and JDM sti7 remapps.
Cheers,
Ray
I am just going on what tuners have told me, I think the ecu's are indeed different, not only the maps but the actual ecu itself.....and as I said before the JDM is supposibly superior. Ask some of the mappers that have done Uk, Euro and JDM sti7 remapps.
Cheers,
Ray
07 April 2003, 01:24 PM
#40
Scooby Regular
Join Date: Oct 2001
Posts: 141
Likes: 0
Received 0 Likes
on
0 Posts
Thanks Matt and Steven.
Just want to add informations regarding Scooby50's car and the other STi7 :
Scooby50 : 2.5" Prodrive sport cat exhaust and MBC (Dawes Device type) @ 1.25 bar
Other STi7 : 2.5" Madac sport cat exhaust, standard boost control
Lambda measurements have been made using the same MOTEC sensor on both cars and the same behaviour has been observed.
I can understand that using an MBC, a small amount of throttle can produce a lot of boost. But what surprises me is that we did saw the close-loop to open-loop transition where the fuel mixture became rich (this happens at about 0.5 bar of boost pressure), but the fuel mixture became lean again at about 1 bar, for a couple of seconds, before becoming rich again !
Don't know if a STi8 has the same behaviour. The only think I know is that the duty cycle boost solenoid MAP has been changed on the STi8 so that a small amount of throttle will produce more boost than it produces on an STi7 !
Just want to add informations regarding Scooby50's car and the other STi7 :
Scooby50 : 2.5" Prodrive sport cat exhaust and MBC (Dawes Device type) @ 1.25 bar
Other STi7 : 2.5" Madac sport cat exhaust, standard boost control
Lambda measurements have been made using the same MOTEC sensor on both cars and the same behaviour has been observed.
I can understand that using an MBC, a small amount of throttle can produce a lot of boost. But what surprises me is that we did saw the close-loop to open-loop transition where the fuel mixture became rich (this happens at about 0.5 bar of boost pressure), but the fuel mixture became lean again at about 1 bar, for a couple of seconds, before becoming rich again !
Don't know if a STi8 has the same behaviour. The only think I know is that the duty cycle boost solenoid MAP has been changed on the STi8 so that a small amount of throttle will produce more boost than it produces on an STi7 !
07 April 2003, 01:28 PM
#41
Scooby Regular
Join Date: Feb 2003
Location: STi8 G spec (400+hp)
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
As said above, it's a typical upgrade... so if I agree with Stephen comments on the standard ECU and exhaust effects, I definately can't understand why those std maps run 16 PSI (1.1 bar ish) of boost without getting out of the loop by reading MAF or MAP sensors ? They obviously see high values, but no effect on the soft.
Does this mean that the ECU is reading those sensors at partial throttle for stochio and advance only ? I don't have any idea about the boost a stk STi can built at partial throttle with stk exhaust system back pressure but this sounds weird. I expect at least 7 to 8 PSI with 80-90% : at 14.7:1, this already means high EGT.
With ref to steve remark about TPS, I read 0.58v to 3.94v on my select monitor... I hope this is a correct 0-100% range.
Does this mean that the ECU is reading those sensors at partial throttle for stochio and advance only ? I don't have any idea about the boost a stk STi can built at partial throttle with stk exhaust system back pressure but this sounds weird. I expect at least 7 to 8 PSI with 80-90% : at 14.7:1, this already means high EGT.
With ref to steve remark about TPS, I read 0.58v to 3.94v on my select monitor... I hope this is a correct 0-100% range.
08 April 2003, 02:12 PM
#43
Scooby Regular
Join Date: Sep 2000
Location: London
Posts: 7,039
Likes: 0
Received 0 Likes
on
0 Posts
Presumably, with modified breathing, part throttle @ 1 bar takes it into a part of the fuel map it wasn't supposed to go under "normal" (i.e. unmodified) use......?
Matt.
Matt.
08 April 2003, 02:18 PM
#44
Scooby Regular
Join Date: Nov 2000
Location: 32 cylinders and many cats
Posts: 18,658
Likes: 0
Received 1 Like
on
1 Post
This is the I-club new age WRX MBC full boost part throttle high EGT lean as f**k scenario again? Doesn't need to be an MBC that gives you the high boost on part throttle especially with the rather frisky nature of the small P15 or whatever it is exhaust housing on the VF35. The fact that they can boost creep and surge if you sneeze in the wrong direction suggests they can be very frisky on part throttle. This would be great if you are near surge but not on it and have a MAP based ECU, or the MAF/TPS setup is mapped correctly. If a little frisky for some and tricky to control in the wet mid corner on the throttle?
The older ECUs mainly have MAF as an input and a little from TPS and possibly from MAP.
With a TD04 on the old cars you could get it a bit lean with a Dawes with a very small bleed hole. Maybe the Denso ECUs have more of a contribution from TPS than MAF and MAP to the load points?
Maybe at lower midrange RPM you are spooling into areas of the map that are pretty lean because of the "progressive" standard spool up with the factory boost tables on the factory ECU?
No answers just a few more thoughts to throw in.
The older ECUs mainly have MAF as an input and a little from TPS and possibly from MAP.
With a TD04 on the old cars you could get it a bit lean with a Dawes with a very small bleed hole. Maybe the Denso ECUs have more of a contribution from TPS than MAF and MAP to the load points?
Maybe at lower midrange RPM you are spooling into areas of the map that are pretty lean because of the "progressive" standard spool up with the factory boost tables on the factory ECU?
No answers just a few more thoughts to throw in.
08 April 2003, 02:35 PM
#45
Scooby Regular
Join Date: Jul 2001
Posts: 326
Likes: 0
Received 0 Likes
on
0 Posts
mmm...so it is better to avoid part throttle, high boost scenarios? Better to use WOT when you want to "use" high boost...
Would there be a cheap and easy way (POT) to just fool the ecu into thinking you're using WOT when you're not?
Would there be a cheap and easy way (POT) to just fool the ecu into thinking you're using WOT when you're not?
08 April 2003, 04:36 PM
#46
Scooby Regular
Join Date: Oct 2001
Posts: 141
Likes: 0
Received 0 Likes
on
0 Posts
That's what I think : it's better to drive ON/OFF (low throttle/WOT) than smoothly to avoid lean mixture.
And I'm pretty sure that STI7 fuel pump is efficient enough to support up to 350 HP.
STi7's engines might have suffered from lean mixture at part throttle/high boost as measured rather than from lean mixture at high rev which has not been measured on the 3 tested cars.
And I'm pretty sure that STI7 fuel pump is efficient enough to support up to 350 HP.
STi7's engines might have suffered from lean mixture at part throttle/high boost as measured rather than from lean mixture at high rev which has not been measured on the 3 tested cars.
08 April 2003, 05:15 PM
#47
Scooby Regular
Join Date: Nov 2000
Location: 32 cylinders and many cats
Posts: 18,658
Likes: 0
Received 1 Like
on
1 Post
It is a pretty dodgy setup if you have to drive in a certain way to avoid it blowing up though.
A truly complete and enjoyable car in final form should be reliable and allow you to concentrate on delivering all that power safely to the road surface with the mechanicals looking after themselves?
A truly complete and enjoyable car in final form should be reliable and allow you to concentrate on delivering all that power safely to the road surface with the mechanicals looking after themselves?
08 April 2003, 05:23 PM
#48
Scooby Regular
Join Date: Feb 2002
Posts: 1,343
Likes: 0
Received 0 Likes
on
0 Posts
John,
I couldn't agree with you more
No way I want to 'baby' my car after it has warmed up and ready to rumble...
I have confidence that with a new fuel pump and remapp by Pat and full decat running 98 RON gas (mixed with 114 octane !!!! on drag days) I should have a safe happy car....if not I will drive it into the ocean with it tied to my leg
Cheers,
Ray
I couldn't agree with you more
No way I want to 'baby' my car after it has warmed up and ready to rumble...
I have confidence that with a new fuel pump and remapp by Pat and full decat running 98 RON gas (mixed with 114 octane !!!! on drag days) I should have a safe happy car....if not I will drive it into the ocean with it tied to my leg
Cheers,
Ray
08 April 2003, 06:08 PM
#49
Scooby Regular
Join Date: Oct 2001
Posts: 141
Likes: 0
Received 0 Likes
on
0 Posts
I agree with you John. The question is : why Subaru has made such a mistake in the fuel MAP ?
I'd have like someone in the UK to reproduce our wide band lambda measurements. The more we have logs, the more we can understand what the problem is. And I'm sure that the fuel mixture is perfect (even too rich) at high rev !
I'd have like someone in the UK to reproduce our wide band lambda measurements. The more we have logs, the more we can understand what the problem is. And I'm sure that the fuel mixture is perfect (even too rich) at high rev !
08 April 2003, 09:49 PM
#51
Scooby Regular
Join Date: Aug 2001
Posts: 95
Likes: 0
Received 0 Likes
on
0 Posts
If we follow you Matt, how can you explain the lot of pistons failures which were overheated or molten (that's sure and said by Subaru) on standards cars ?
Running lean at a specific moment could explain that, I don't see other reasons but I'm open to all explanations...
Running lean at a specific moment could explain that, I don't see other reasons but I'm open to all explanations...
08 April 2003, 10:29 PM
#52
Scooby Regular
Join Date: Feb 2001
Location: Greece, previously Syd Australia
Posts: 2,833
Likes: 0
Received 0 Likes
on
0 Posts
I agree with you John. The question is : why Subaru has made such a mistake in the fuel MAP ?
To give tuners and engine builders some business!
Cheers,
Wrexy.
08 April 2003, 10:43 PM
#53
BANNED
Join Date: Feb 2001
Posts: 624
Likes: 0
Received 0 Likes
on
0 Posts
RT,
I seem to be getting similar readings on the SECS as you.
JDM STi7.
At 1.0 bar approx - hard to hold exactly - reading 0.76
Tried to watch the readings at different speeds coming along the motorway today. Private test track for the last one !!!
At 80mph / 3700RPM in 6th / -0.1bar / 0.95-0.96 reading
At 60mph / 2800RPM in 6th / -0.3bar / 0.99 constant reading
At 100mph/ 4600RPM in 6th / +0.4bar / 0.80-0.85 reading
At higher boosts 0.5bar in different gears readings from 0.76-0.8
I hope this makes some sense to some.
John Banks - if you want to try and get some better readings, I'll be free most of Friday.
David
I seem to be getting similar readings on the SECS as you.
JDM STi7.
At 1.0 bar approx - hard to hold exactly - reading 0.76
Tried to watch the readings at different speeds coming along the motorway today. Private test track for the last one !!!
At 80mph / 3700RPM in 6th / -0.1bar / 0.95-0.96 reading
At 60mph / 2800RPM in 6th / -0.3bar / 0.99 constant reading
At 100mph/ 4600RPM in 6th / +0.4bar / 0.80-0.85 reading
At higher boosts 0.5bar in different gears readings from 0.76-0.8
I hope this makes some sense to some.
John Banks - if you want to try and get some better readings, I'll be free most of Friday.
David
08 April 2003, 11:13 PM
#54
Scooby Regular
Join Date: Feb 2003
Location: STi8 G spec (400+hp)
Posts: 179
Likes: 0
Received 0 Likes
on
0 Posts
As JDM and Euro spec STi drivetrains mainly have different exhausts, considering that we certainly have higher back press in our dual cat system, can we imagine the Euro spec fuel maps were basically JDM maps corrected to provide less fuel according to rpm and tps in any but not WOT conditions...?
In this case, back pressure relief would explain the lean situation at part throttle as software is not expecting such high boost values... but what about A/F Correction maps then ?
What a pity to have all these sensors (comparing to my classic 98') and such a stupid software not ready to cope with this more-than-typical mod : exhaust upgrade is generally the first thing we all go for...
I think I'm gonna bring my STi8 to the doctor
Any further trip planned in south France JB ?
I can arrange hotel reservation, shopping for ladies and track entertainment for gentlemen
In this case, back pressure relief would explain the lean situation at part throttle as software is not expecting such high boost values... but what about A/F Correction maps then ?
What a pity to have all these sensors (comparing to my classic 98') and such a stupid software not ready to cope with this more-than-typical mod : exhaust upgrade is generally the first thing we all go for...
I think I'm gonna bring my STi8 to the doctor
Any further trip planned in south France JB ?
I can arrange hotel reservation, shopping for ladies and track entertainment for gentlemen
09 April 2003, 10:22 AM
#56
Scooby Regular
Join Date: May 2001
Posts: 720
Likes: 0
Received 0 Likes
on
0 Posts
Tks David. Those results sound about the same as mine.
I think the most likely reason for cars to run 14.7 lambda at 1bar part throttle, would have to be the Dawes Device setup. Certainly on my car with the std boost solenoid in control, it takes quite a bit of throttle to get it to hold 1bar in 4th or 5th gear. Perhaps the TPS indeed does have a significant input to the ECU.
I think the most likely reason for cars to run 14.7 lambda at 1bar part throttle, would have to be the Dawes Device setup. Certainly on my car with the std boost solenoid in control, it takes quite a bit of throttle to get it to hold 1bar in 4th or 5th gear. Perhaps the TPS indeed does have a significant input to the ECU.
09 April 2003, 10:38 AM
#57
Scooby Regular
Join Date: Oct 2001
Posts: 141
Likes: 0
Received 0 Likes
on
0 Posts
Thanks David and RT.
Once again, the JDM STi7 shouldn't have the same behaviour as fuel and ignition MAP are different, that's why you didn't found anything wrong.
One of the EURO STi7 tested car did have an MBC, the other not !
Once again, the JDM STi7 shouldn't have the same behaviour as fuel and ignition MAP are different, that's why you didn't found anything wrong.
One of the EURO STi7 tested car did have an MBC, the other not !
10 April 2003, 02:47 PM
#60
Scooby Regular
Join Date: Jul 2001
Posts: 326
Likes: 0
Received 0 Likes
on
0 Posts
swaussie, I think so...
it's almost instinctive with me these days to slowly feed in the throttle, instead of just using WOT...just drives so much better that way...Is WOT causing it to run overly rich?
it's almost instinctive with me these days to slowly feed in the throttle, instead of just using WOT...just drives so much better that way...Is WOT causing it to run overly rich?