|
|
| Author |
Message |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
 Posted: 13:39 - 18 Aug 2008 Post subject: Rolling Road argument (Car thread but sort of bike related) |
 |
|
Hi Guys,
On another forum I occasionally frequent, there is a minor argument about fuel mapping on turbocharged cars.
Apparently the best way (according to the members of this forum) to build a fuel injection and ignition map for a turbocharged car is to have a passenger sit in the passenger seat with a laptop connected to the ECU and a pair of det cans (which I assume is some sort of knock sensor on a pair of earphones) and set up the car that way.
I find this strange, because as far as bikes are concerned, the only way to set them up after exhaust or intake mods is on a dyno. I've always been schooled with the idea that a Dynamometer is the only way to measure and set up fuelling. The Dyno can produce varying loads on the engine, and it can test the engine running through the gears (we all know there are different fuel maps for different gears on bikes, but what about cars?).
So how the hell can these guys say that 'generic maps' or a guy with a laptop being driven around can get the fuelling as perfect as a trained Dyno operator and a proper calibrated rolling road? The guy with the laptop can't even tell if the engine is making more power? Driving around cannot let the engine rev from idle to redline in every gear and cannot measure the load on the engine.
I'm confused, and I'm sure that I'm right in this case. If I'm wrong then why do people bother to build rolling road dynos for cars? Another few members are also arguing that you don't need to remap after modifications? I think they are crazy. Or are they?  ____________________ British beauty: Triumph Street Triple R; Loony stroker: KR1S; Track fun: GSXR750 L1; Commuter Missile: GSX-S1000F; Cheap project: CBR900RR FireBlade
Remember kids, bikes aren't like lego. You can't easily take a part from one bike and then fit it to another. |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
 Posted: 13:50 - 18 Aug 2008 Post subject: |
 |
|
Hi
On your last point first, most cars run closed loop systems. To a large extent on anything other than very large throttle openings the car doesn't care how its fueling is mapped as it will just redo it from the feedback from the lambda probe anyway.
You can actually determine power from a laptop like that. Change in time time between sparks means you can get a pretty accurate view of acceleration, and plumb in the weight of the car and the like and you can work it out. However for mapping you want consistancy so you would need to use the same stretch of road (and really the same fuel load.
The main thing it sounds like they are mapping is the allowable boost and ignition timing rather than fuel mixture. And for this an accurate way of finding when detonation becomes a problem is needed.
All the best
Keith ____________________ Traxpics, track day and racing photographs - Bimota Forum - Bike performance / thrust graphs for choosing gearing |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
 Posted: 15:23 - 18 Aug 2008 Post subject: |
 |
|
| MarJay wrote: | Apparrently they are setting everything to 14.7 to 1 across the rev range which what really worries me... They can't guarantee that for a different load the fuelling will be the same... |
Except the ECU will probably sort that out for itself. It will know that the last time it was on X throttle position with Y boost (and plenty of the rest of the alphabet in other parameters) that to get the fueling to an acceptable amount it opened the injectors for Z time. It doesn't need to know the load directly as it can infer it from the manifold pressure (on a turbo engine likely to be boost level).
Disconnect the battery for a while and it will probably run badly to start with and improve as the management system maps itself, (probably including up to a predefined and fairly safe level of boost).
Fair chance that its default mapping will be based on aiming for a fairly lean mix while cruising, very lean (and maybe no fuel at all) when the throttle is closed at revs and a rich mixture on large throttle openings. These default mixtures could be played with, but chances are that it would take a bit of driving for the ECU to decide how much fuel it needs to inject for the mixture it wants in those conditions.
| MarJay wrote: | Also, surely any given map will have tolerances that it won't stray beyond. So it will 'learn' to some degree, but it won't learn enough to improve peak power by much, or to get rid of holes in the torque curve for example? |
The fueling isn't really there for peak power. The boost / compression / ignition timing / cam timing / etc are generating the power while the fueling it just set up for that. You gain a small amount of power with a richer mixture and loose a bit for a leaner mixture within reason (and go outside those limits and you will land up with something that barely runs or blows up)
A hole in the torque curve is probably where other components are conflictiing, screwing the mixture up as well. You are not going to be able to fix it on the fueling alone, but the ECU can find that when it has these conditions the mixture has leaned off and it needs to add X% more fuel to compensate.
To an extent the feedback from the lambda probe enables the engine to run on pretty much any old fuel, hence getting away with adding ethanol to petrol (ecu detects the leaner mixture and richens up to compensate, with the effects of the ethanol being within the range it thinks is acceptable).
Playing for really silly amounts of power (at the level that the Subaru flat 4 would already be shrapnell embedded in the walls) then yes you might well want to run stupidly rich mixtures partly to cool down inlet temps and piston temps.
All the best
Keith ____________________ Traxpics, track day and racing photographs - Bimota Forum - Bike performance / thrust graphs for choosing gearing |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
 Posted: 15:52 - 18 Aug 2008 Post subject: |
 |
|
So I could increase the boost that my car produces and I would make more peak power. Simple as that? I'm really disappointed now. I know for example that the PPP ECU runs 1.2 bar max boost, and my standard ECU runs 0.9 bar boost. I've had the other PPP mods done (exhaust, centre de-cat, strengthened intercooler hose) so I could just put my boost up to 1.2bar (if I had the right software and an ODBII cable) and my car won't blow up?
I thought engine tuning was a black art, and it turns out that simply bolting stuff on can work. Damn you closed-loop fuelling systems!!
I'm still unconvinced. I'm certain a re-map will make my car faster. Maybe boost and ignition timing need to be tweaked. ____________________ British beauty: Triumph Street Triple R; Loony stroker: KR1S; Track fun: GSXR750 L1; Commuter Missile: GSX-S1000F; Cheap project: CBR900RR FireBlade
Remember kids, bikes aren't like lego. You can't easily take a part from one bike and then fit it to another. |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Dom_ |
This post is not being displayed .
|
 Dom_ Points Mean Prizes

Joined: 02 Aug 2004 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
 Posted: 16:57 - 18 Aug 2008 Post subject: |
 |
|
| MarJay wrote: | so I could just put my boost up to 1.2bar (if I had the right software and an ODBII cable) and my car won't blow up? |
Probably won't go bang due to the fueling (assuming that it is mapped already up to that boost level, and Subaru didn't just cut the mapping off at a lower level). Might well go bang due to pistons not being strong enough, detonation, blown head gasket, etc.
| MarJay wrote: | I thought engine tuning was a black art, and it turns out that simply bolting stuff on can work.  Damn you closed-loop fuelling systems!! |
Knowing what you are really doing might well get you more power with loads of effort and time spent mapping. As long as the ECU doesn't just ignore it and remap itself.
| MarJay wrote: | I'm still unconvinced. I'm certain a re-map will make my car faster. Maybe boost and ignition timing need to be tweaked. |
Yes it will, but boost is probably only crudely controlled while ignition timing will just land up being retarded based on the feedback from a knock sensor to try and avoid detonation. Ram the boost through the roof and the car will probably detect the detonation and retard the ignition timing drastically, possibly loosing you power.
Dom_ - trouble is that although the load migth be closer to real life, it is also basically impossible to accuratly repeat. Decent dyno means that they can vary the load at the wheels (basically run a brake on the roller). Down side of the dyno is getting a decent enough cold air flow through the intercooler, etc.
All the best
Keith ____________________ Traxpics, track day and racing photographs - Bimota Forum - Bike performance / thrust graphs for choosing gearing |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
 Posted: 17:49 - 18 Aug 2008 Post subject: |
 |
|
| Kickstart wrote: |
Dom_ - trouble is that although the load might be closer to real life, it is also basically impossible to accuratly repeat. Decent dyno means that they can vary the load at the wheels (basically run a brake on the roller). Down side of the dyno is getting a decent enough cold air flow through the intercooler, etc. |
Intercooler air flow isn't an issue on the impreza anyway. Its a top mounted intercooler so the airflow across it is rubbish.
The PPP (prodrive performance pack) ECU (which is fitted to an essentially unmodified car) ups the boost to 1.2 bar, and the engine internals are OK on that. It apparently ups the power to 260bhp from 215, but I can't find out as the PPP ECU I've got is broken The PPP also includes a strengthened boost hose for the intercooler, a single de cat (there are three, PPP takes it to two) and a Prodrive silencer. I've got the boost hose, silencer and de-cat but I need a remap for more power. With a proper re-map I should (apparently) be ok for 270bhp ish. Lots of (the chavvier) members of the scooby forum fit something like an Apexi boost controller or a variable resistor on the MAF sensor, and apparently it rarely causes engine destruction until they up the boost over 1.2 bar.
Apparently stock WRX internals are OK for 320bhp, but that needs a bigger turbo.
Its all fairly moot anyway as my car is blowing from the up (turbo feed) pipe and so will be underboosting anyway. I do hope to de cat the up pipe (and so fix the hole) and then get a remap at some point in the future. ____________________ British beauty: Triumph Street Triple R; Loony stroker: KR1S; Track fun: GSXR750 L1; Commuter Missile: GSX-S1000F; Cheap project: CBR900RR FireBlade
Remember kids, bikes aren't like lego. You can't easily take a part from one bike and then fit it to another.
Last edited by MarJay on 17:53 - 18 Aug 2008; edited 1 time in total |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| sickpup |
This post is not being displayed .
|
 sickpup Old Timer

Joined: 21 Apr 2004 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| G |
This post is not being displayed .
|
 G The Voice of Reason
Joined: 02 Feb 2002 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| sickpup |
This post is not being displayed .
|
 sickpup Old Timer

Joined: 21 Apr 2004 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| Kickstart |
This post is not being displayed .
|
 Kickstart The Oracle

Joined: 04 Feb 2002 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| MarJay |
This post is not being displayed .
|
 MarJay But it's British!

Joined: 15 Sep 2003 Karma :     
|
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
| sickpup |
This post is not being displayed .
|
 sickpup Old Timer

Joined: 21 Apr 2004 Karma :     
|
 Posted: 22:14 - 18 Aug 2008 Post subject: |
 |
|
I'll give you that, it can be read both ways too used to arguing with you
In my experience bikes set up on Dyno's should be taken for a test run afterwards to make sure they actually work in the real world as it's surprising how often they don't.
Bit like software I suppose, you build it in a lab then hand it over for someone to try and destroy. Trouble with Dyno's is people seem to think they can work miracles when all they are is a tool and you should always double check your work. |
|
| Back to top |
|
You must be logged in to rate posts |
|
 |
Old Thread Alert!
The last post was made 17 years, 236 days ago. Instead of replying here, would creating a new thread be more useful? |
 |
|
|