 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Car Forum / Saab Cars / August 2005"Throttle body" on 9-5 -- why!?
My 2002 9-5 Aero started dying when idling, and the "Check Engine" light came on. Oh boy. Take it to the dealer, describe the problem, and the service manager immediately identifies it as a faulty "throttle body." Tech II confirms it, and says the car is in "limp home" mode. So apparently this is a common failure -- the service guy knew instantly what it was, before even seeing the car, and the car has a special mode to keep functioning when it goes bad. Great. The service manager pops the hood and points out the part. "Oh $#|+," I say, "that's $500." He adds up parts & labor and it comes to $506. Do I know my Saabs or what? :-( The throttle body is apparently an electronic throttle controller. It wouldn't be cool enough to use a plain old mechanical throttle linkage like everybody used to (how plebian), so the 9-5 uses a "fly by wire" arrangement with a potentiometer on the gas pedal controlling the throttle body. Plus some kind of mechanical feedback so you can feel it. Questions: * Why does Saab have to replace the simple, cheap, near-foolproof mechanical linkage with an expensive and failure-prone subsystem? The only reason I can think of is the anti-skid controller, but geeze, wasn't there a simpler and more reliable way to do it? * Why is it necessary at all, since obviously the car can run just fine without it in "limp home" mode!? Except they carefully made the "limp home" mode inconvenient (dying at idle) so you'd have to fix it anyway. Grumble. Gary >* Why does Saab have to replace the simple, cheap, near-foolproof >mechanical linkage with an expensive and failure-prone subsystem? The >only reason I can think of is the anti-skid controller, but geeze, wasn't >there a simpler and more reliable way to do it? It's so the ECU can have control over the throttle. There's various needs for that. You've already spotted the traction control. There's also cruise control and often a torque limit on the lower gears (though that might be done on the boost control). There might well be some emissions related stuff too. >* Why is it necessary at all, since obviously the car can run just fine >without it in "limp home" mode!? Except they carefully made the "limp >home" mode inconvenient (dying at idle) so you'd have to fix it anyway. It doesn't have to be unreliable. I've driven cars with electronic throttles for years, yet the only throttle controls I've ever had fail have been cable operated ones. The real shame is that the only way the dealers seem to have of fixing it, is to bin it and fit a new one. It might be an expensive assembly, but I bet the bit of it that broke is really cheap. Cheers, Colin. > * Why does Saab have to replace the simple, cheap, near-foolproof > mechanical linkage with an expensive and failure-prone subsystem? The > only reason I can think of is the anti-skid controller, but geeze, wasn't > there a simpler and more reliable way to do it? i think the main reason is emissions. this way the ECU reads from the gas pedal what you want to do, compares that with what the car is doing and then opens or closes the throttle in a way to reduce emissions. if you open the throttle faster than the injectors can keep up, then you'll have lean issues resulting in more NOx and if you close the throttle quicker than you can shut the injectors down, you'll have a rich condition resulting in more HCs (this can easily be taken care of with a damper on the throttle closing, my old toyota has this and it's fool proof). personally, i hate throttle-by-wire. when i drive a throttle-by-wire car, i can notice quite a lag between the time i press the go pedal and the time the car starts going. it makes pulling out into traffice a pain. i guess you'd probably get used to it, but i don't drive throttle-by-wire enough to know. there are simpler ways to handle things like torque limiting,anti-skit control, rev limiting, etc. it all falls under messing with the timing. but i guess they'd rather put some cheap servos on there and chare $500 when they go bad. i just wish you had a choice when you bought a car. i'd order mechanical controls on everything from the throttle to the windows to the door locks. i like stuff that i can fix myself. Oh dear. I don't really know where to begin... >i think the main reason is emissions. this way the ECU reads from the >gas pedal what you want to do, compares that with what the car is doing [quoted text clipped - 3 lines] >throttle quicker than you can shut the injectors down, you'll have a >rich condition resulting in more HCs Nope. That's not how it happens at-all. Some damping is, and always has been, a good idea. Nothing to do with the speed of injectors. They open and close in microseconds. It's all about gas-flow. >(this can easily be taken care of >with a damper on the throttle closing, my old toyota has this and it's >fool proof). This has nothing to do with electronic throttle control. When you go to electronic control, you might as well put the damping into the software. Take off the electronic control and you'd still need the same amount of damping. >personally, i hate throttle-by-wire. No. You hate some cars which have throttle-by-wire. Unless you've tried the same car both with and without it, you can't tell if it's the throttle-by-wire you hate. >when i drive a >throttle-by-wire car, i can notice quite a lag between the time i press >the go pedal and the time the car starts going. If that's really the case, then it's because the throttle has had to be heavily damped for some reason. Maybe to meet some new emissions standard or whatever, but nothing to do with the electronic controls. >it makes pulling out >into traffice a pain. i guess you'd probably get used to it, but i >don't drive throttle-by-wire enough to know. there are simpler ways to >handle things like torque limiting,anti-skit control, rev limiting, >etc. it all falls under messing with the timing. but i guess they'd >rather put some cheap servos on there and chare $500 when they go bad. "Messing with the timing" is a very badly flawed way of controlling engine torque, as is cutting sparks or cutting fuel. They all have serious problems with smoothness, efficiency, emissions and engine strain. The best way of reducing torque is to use the same control the driver uses - the throttle. >i just wish you had a choice when you bought a car. i'd order >mechanical controls on everything from the throttle to the windows to >the door locks. i like stuff that i can fix myself. You do have a choice. Just buy a car that was built in the '70s. Cheers, Colin. >>personally, i hate throttle-by-wire. > No. You hate some cars which have throttle-by-wire. Unless you've > tried the same car both with and without it, you can't tell if it's > the throttle-by-wire you hate. I must agree. I regularly drive a c900 and a 2001 9-3, and I find them to be equally responsive to the throttle. John >>>personally, i hate throttle-by-wire. >> [quoted text clipped - 6 lines] > > John Is the 2001 9-3 a standard transmission, John? Probably not or you would not say that. When you release the accelerator it there is a delay before the throttle closes. It is extremely annoying and one of the main reasons I ditched my 2000 9-3SE. On an automatic trans car you would hardly be able to notice it. -Fred W >>>>personally, i hate throttle-by-wire. >>> [quoted text clipped - 14 lines] > >-Fred W Are you sure your 9-3 had an electronic throttle? In the UK at least, the 2000 9-3 LPT had a cable throttle. I had one for about a year before getting a 2001 Aero (which does have an electronic throttle). Both are manual and I didn't notice that problem on either car. Cheers, Colin. >>>>>personally, i hate throttle-by-wire. >>>> [quoted text clipped - 19 lines] > before getting a 2001 Aero (which does have an electronic throttle). > Both are manual and I didn't notice that problem on either car. Yes, definitely. Mine was a '00 9-3SE convertible, which has a FPT (the H.O.T. engine) Perhaps the LPT was a better bet in that regard. I have a feeling the HOT engine (at least that year) would have been better paired up with an automatic as it had that terrible trailing throttle delay and also a tendency to break loose the drive wheels in 1st through 3rd gear under heavy acceleration. Then you get into the whole torque steer thing... I currently own 2 1998 900's. An SE with a 5 speed and an S with Auto. Neither has the annoying tendency. I'm sure it was somehow related to the programming for that engine. -Fred W > Yes, definitely. Mine was a '00 9-3SE convertible, which has a FPT (the > H.O.T. engine) Perhaps the LPT was a better bet in that regard. I have [quoted text clipped - 3 lines] > 3rd gear under heavy acceleration. Then you get into the whole torque > steer thing... On that point I totally agree with you. The HOT packs more than enough torque to launch you into outer space if you're not careful. The TCS sort of helps, but the torque steer is still a killer. John >>Yes, definitely. Mine was a '00 9-3SE convertible, which has a FPT (the >>H.O.T. engine) Perhaps the LPT was a better bet in that regard. I have [quoted text clipped - 9 lines] > > John There was no TCS on mine. That may have helped, but based on my experience with SAAB TCS on '92 and '93 9000s, it is also very annoying. Of course neither of those cars had TCS on/off switches which would have been immeasurably better.  Signature-Fred W Toys for sale, Hey get your toys here: <http://users.adelphia.net/~fredwills/> >> Are you sure your 9-3 had an electronic throttle? In the UK at least, >> the 2000 9-3 LPT had a cable throttle. I had one for about a year [quoted text clipped - 3 lines] >Yes, definitely. Mine was a '00 9-3SE convertible, which has a FPT (the >H.O.T. engine) Ah, all the SEs I've come across have had LPT engines. I don't think the HOT engine option sold at-all well on the SE. >Perhaps the LPT was a better bet in that regard. I have >a feeling the HOT engine (at least that year) would have been better >paired up with an automatic as it had that terrible trailing throttle >delay and also a tendency to break loose the drive wheels in 1st through >3rd gear under heavy acceleration. Then you get into the whole torque >steer thing... I haven't noticed the trailing throttle problem on mine. You're definitely right about the traction problems though. I always thought it was mad to fit an open diff on a 210 BHP FWD car! The other problem on the HOT engine is the turbo lag. It was hardly noticeable on the LPT, but it's quite annoying on the Aero. >I currently own 2 1998 900's. An SE with a 5 speed and an S with Auto. > Neither has the annoying tendency. I'm sure it was somehow related to >the programming for that engine. Could still be. Perhaps there is some emissions requirement there that we don't have here or something. There might even have been something wrong with it. Cheers, Colin. >>>Are you sure your 9-3 had an electronic throttle? In the UK at least, >>>the 2000 9-3 LPT had a cable throttle. I had one for about a year [quoted text clipped - 6 lines] > Ah, all the SEs I've come across have had LPT engines. I don't think > the HOT engine option sold at-all well on the SE. In the US all SEs had the HOT engine. Different market rules I guess... >>Perhaps the LPT was a better bet in that regard. I have >>a feeling the HOT engine (at least that year) would have been better [quoted text clipped - 16 lines] > we don't have here or something. There might even have been something > wrong with it. Good point. Maybe the ECU programming is different for UK vs. US export even on the same engines. After all, we have the EPA looking out for us over here... ;-) Signature-Fred W Toys for sale, Hey get your toys here: <http://users.adelphia.net/~fredwills/> >> I must agree. I regularly drive a c900 and a 2001 9-3, and I find them to be >> equally responsive to the throttle. [quoted text clipped - 3 lines] > Is the 2001 9-3 a standard transmission, John? Probably not or you > would not say that. When you release the accelerator it there is a Yes, both cars are 5-speed. John < snip > > if you open the throttle faster than the injectors can keep up, then > you'll have lean issues resulting in more NOx and if you close the > throttle quicker than you can shut the injectors down, you'll have a > rich condition resulting in more HCs I can't see that reasoning holding water. Injectors regulate the fuel mixture on cycle by cycle basis. For the throttle opening / closure to 'beat' the injection you'd need to have a very very fast right foot. Even at tickover speeds the injectors are functioning about every 30 milliseconds. Graham really i said it wrong. the comptuers in the ecus are pretty slow and they're what doesn't react that fast. there are a series of technical articles at www.autoshop101.com that explains a bunch of this. i don't have the time to search for the one i was thinking of when i wrote my original post, but it's there. >really i said it wrong. the comptuers in the ecus are pretty slow and >they're what doesn't react that fast. there are a series of technical >articles at www.autoshop101.com that explains a bunch of this. i don't >have the time to search for the one i was thinking of when i wrote my >original post, but it's there. That's just as wrong as the slow injector idea, I'm afraid. The ECU can happily react as much as it likes in less than one engine cycle. Anything faster than that is irrelevant. Cheers, Colin. > really i said it wrong. the comptuers in the ecus are pretty slow and > they're what doesn't react that fast. there are a series of technical > articles at www.autoshop101.com that explains a bunch of this. i don't > have the time to search for the one i was thinking of when i wrote my > original post, but it's there. It can respond in a few milliseconds to a detected ping (pre-ignition) and cut boost back to base level. I'm guessing that's fast enough. It's just my opinion, but manual boost controllers belong in the same section as turbine intakes, conical performance "cold" air intake filters and throttle body spacers... Signature-Fred W Toys for sale, Hey get your toys here: <http://users.adelphia.net/~fredwills/> > really i said it wrong. the comptuers in the ecus are pretty slow and > they're what doesn't react that fast. there are a series of technical > articles at www.autoshop101.com that explains a bunch of this. i don't > have the time to search for the one i was thinking of when i wrote my > original post, but it's there. Actually many of the computers used in ECUs are derivatives of the 8051 family originated by Intel and they are certainly not slow at all. I've used 8051s myself ( designed the hardware - amd programmed them ) and know this for a fact. I believe you've been misadvised. Graham |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.