DIY_EFI Digest Sunday, October 31 1999 Volume 04 : Number 614 In this issue: Re: DIY_EFI Digest V4 #613 Limp home mode on GM systems Dallas 80C550 Re: DIY_EFI Digest V4 #613....AE Transients See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- Date: Sun, 31 Oct 1999 08:23:28 -0600 From: nacelp@xxx.com (CSH-HQ) Subject: Re: DIY_EFI Digest V4 #613 Just hook it all up per the factory. The "tough"part is the VSS and just read the archives for that. All the sensors?. TPS, CTS, VSS, MAP, how much simplier do you want?. The ecm powers the fuel pump for about 2 sec when you turn the key on to prime the system, from then on it's the oil pressure switch that powers it. In a roll over it serves as a saftey item, since inverted there is no oil pressure. Don't try just running in limp home mode, or just start it that way, do it right once, and limp home mode will probably get ya home if needed. Grumpy > >From: Mike Comai >Subject: Limp home mode on GM systems >I'm putting a GM '747 system into a 1979 CJ5. I was curious if anyone new >what sensors/systems the ECM uses when it enters limp home mode. I >noticed that on the schematics it looks like the fuel pump is running all >the time when the oil pressure switch is closed(i.e. bypasses the relay) >and the only reason I can think for that would be if the ECM fails or >enters L.H.M., or the relay fails. I just wanted to make sure that I >think of all the angles when I do the install, and I'd appreciate any >more info on L.H.M. or any other odities that aren't easy to see. Thanks, >Mike Comai ------------------------------ Date: Sun, 31 Oct 1999 09:40:13 -0800 From: "John Dammeyer" Subject: Limp home mode on GM systems Hi, First I guess we have to decide what consitutes LHM. For instance, my water temperature gauge input is used to determine when the engine is warm enough to remove cold start enhancement. At 150 degrees all enhancements are gone and up to 150, the difference in engine temperature and 150 is used as a gain enrichment factor. If the gauge fails open it reports 240 degrees so I force it to read 220 and remove enrichment. If it fails short or stops responding I do nothing. I should probably add some sort of warmup timer that sets the temperature to +150 if the temperature sender hasn't changed. If the MAT fails open it reports -60 so the mixture is rich. I haven't tried the engine at -40 in winter so I do not know if I should consider -60 a problem. In either case, if it does fail this way, it just enriches the mixture. Not a real problem. TPS is only used for acceleration and WOT enrichment so if it fails the engine stumbles under acceleration but does run. O2 sensor only impacts the mixture a little bit so can be discarded if it fails. ie: if it reports the wrong values the engine just runs a tinly bit rich or lean but still runs fairly well. MAP sensor This is the biggy. If it fails such that it reports low pressure the mixture will be lean and the engine will have a major drivability problem. If it fails with a high pressure reading the mixture will be excessively rich. In all cases when the MAP sensor is disconnected, the engine stalls. The Barometer sensor is also somewhat critical if it fails with a low pressure reading as the mixture will become very lean. A failure with a high pressure reading defaults to just below sea level (Amsterdam?) so becomes unimportant other than enrichening the mixture somewhat. I currently do not sense if the injectors are functioning or what the CD capacitor voltage is so I can't determine a LHM from that. If the RPM sensors fail, so does the timing and therefore so does the engine. Fuel Pump is determined from presense of pulses and I wouldn't subvert that because of fire hazzard if the ECU thinks LHM is there when vehicle has crashed. Cheers, John Dammeyer >Date: Sat, 30 Oct 1999 22:09:58 -0500 (EST) >From: Mike Comai >Subject: Limp home mode on GM systems > >I'm putting a GM '747 system into a 1979 CJ5. I was curious if anyone new >what sensors/systems the ECM uses when it enters limp home mode. I >noticed that on the schematics it looks like the fuel pump is running all >the time when the oil pressure switch is closed(i.e. bypasses the relay) >and the only reason I can think for that would be if the ECM fails or >enters L.H.M., or the relay fails. I just wanted to make sure that I >think of all the angles when I do the install, and I'd appreciate any >more info on L.H.M. or any other odities that aren't easy to see. Thanks, > ------------------------------ Date: Sun, 31 Oct 1999 09:22:23 -0800 From: "John Dammeyer" Subject: Dallas 80C550 Hi Nick, I've struggled with this one too. What I currently do is increase the pulse width to a maximum value of 13ms for n seconds. I determine this from a change in throttle position (dTPS) where the amount of change above a minimum change is multiplied by a gain factor and then added to the current pulse width. If the TPS goes above a certain value like 75% then I just apply a richer mixture (11:1). The tests are done with a large multi-bladed fan on a 2.4:1 reduction belt drive so the load on the engine varies with the fan speed. From idle to WOT I get a stumble that I haven't been able to get rid of yet. From 3500RPM to WOT the engine just quickly increases speed to 5100RPM. The Fan loads the engine such that the engines maximum speed is at the peak torque value which for a propellor is just perfect. Can you tell me how you are doing the timing for the sequential injection? ie: at idle where are you starting/ ending the injection pulse and where at max power? How are you determining RPM? I'm using the 80C592 and 95% of the code is in Tasking C. Regards, John >Date: Sun, 31 Oct 1999 11:40:06 +1300 >From: "Nicholas Parker" >Subject: Re: DIY_EFI Digest V4 #611 > >Hi List, > >I have written an ecu in assembler for 80c552, but I now have a Dallas >80c550 (8x faster or so). Its for a 4 cyl, two coil, sequential injection >etc. I haven't implemented 'acceleration enrichment' yet. Can somebody >please give me some guidelines, ie: The car I designed the ecu for is a >supercharged Toyota MR2 mk1. so It kind of operates in both modes...that is >both forced induction, and natural aspiration bue to an electromagnetic >supercharger clutch, and a supercharger bypass. > >Thanks, Nick Parker. > ------------------------------ Date: Sun, 31 Oct 1999 14:26:49 EST From: EFISYSTEMS@xxx.com Subject: Re: DIY_EFI Digest V4 #613....AE Transients Hi, You didn't state whether this is speed density or MAF....Man, there is no quick and dirty "good" way to do AE......It's a pretty good example of what to do, look at the Gm stuff as shown on the ftp site or ecm guys hacks....you'll have to stroll down through the code to find the AE stuff but is usually all in one spot....also watch for the IAT(Intake Air Temp) equation also....sorry for the vagueness but AE is alot tougher than steady state.....hth's - -Carl Summers In a message dated 10/31/99 2:19:50 AM Pacific Standard Time, DIY_EFI-Digest-Owner@xxx.edu writes: << Date: Sun, 31 Oct 1999 11:40:06 +1300 From: "Nicholas Parker" Subject: Re: DIY_EFI Digest V4 #611 Hi List, I have written an ecu in assembler for 80c552, but I now have a Dallas 80c550 (8x faster or so). Its for a 4 cyl, two coil, sequential injection etc. I haven't implemented 'acceleration enrichment' yet. Can somebody please give me some guidelines, ie: The car I designed the ecu for is a supercharged Toyota MR2 mk1. so It kind of operates in both modes...that is both forced induction, and natural aspiration bue to an electromagnetic supercharger clutch, and a supercharger bypass. Thanks, Nick Parker. ------------------------------ >> ------------------------------ End of DIY_EFI Digest V4 #614 ***************************** To subscribe to DIY_EFI-Digest, send the command: subscribe diy_efi-digest in the body of a message to "Majordomo@xxx. A non-digest (direct mail) version of this list is also available; to subscribe to that instead, replace "diy_efi-digest" in the command above with "diy_efi".