DIY_EFI Digest Tuesday, January 18 2000 Volume 05 : Number 027 In this issue: Re: Lambda Value Re: Lamda sensor mounting Re: Lambda Value Re: DIY_EFI Digest V5 #25 Re: DIY_EFI Digest V5 #25 Size of (rpm,load) tables See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- Date: Mon, 17 Jan 2000 16:29:07 -0500 From: "Bruce Plecan" Subject: Re: Lambda Value Been covered a thousand times in the archives. Also, I'd at one time poted some dream voltages there for the gm 1-2 wire sensors. You'll understand the term dream after reading the posts. Grumpy | I'm new to this game and I'm aiming to start small with a mixture meter, | driven by an O2 sensor. I have a lot of information on lambda sensor output | values, particularly voltage against lambda value. I know that lambda 1.0 is | equivalent to 14.7:1 air/fuel ratio. What I really need to know is what | lambda values relate to other air/fuel ratios, particularly in the area of | 12.5:1 - 13:1. | Does anybody have the answers, or can anybody point me in the direction of | some relevant web resources? | Gareth Jones ------------------------------ Date: 18 Jan 2000 14:54:53 +1200 From: "Tom Parker" Subject: Re: Lamda sensor mounting Ade + Lamb Chop wrote: >Where in the exhaust system is the best place to mount the sensor? Close to >the head? 6ft down the exhaust? I have a Long center branch exhaust >manifold which doesn't bring the ports together for about 3ft after the >cylinder head. The manifold is also new (and expensive) so I would prefur >to mount is in the exhaust if it doesn't affect the reading. I have a mini also, and mine is mounted just after the branches of the exhaust manifold come together. I use a heated sensor and without the heater it doesn't read at idle, or low cruise. With the heater it seems to work at all power outputs though. - -- Tom Parker - parkert@xxx.nz - http://www.geocities.com/MotorCity/Track/8381/ ------------------------------ Date: Tue, 18 Jan 2000 12:33:49 +1100 From: Andy Wyatt Subject: Re: Lambda Value Regarding Gareth's question about O2 sensors.. I know a very little about EGO sensors, which I learnt from an article in Silicon Chip electronics magazine (in Oz), November 1995. It show a graph of voltage against Lambda for a "typical" sensor. It says with a lambda of 0.94, it will develop a voltage of a bit over 0.9V. The graph isn't very well calibrated, so it ain't very useful, but from what I can tell your typical EGO sensor is only really accurate around stoic. Even then there is hysteresis, which means that the curve you trace out is a different shape when going from rich to lean from the curve going from lean to rich. The graph is VERY steep around stoic point, then it becomes quite flat at lambda values below 0.96 and above 1.05. This is only for a typical EGO sensor, which is only designed to work around stoic, so that EFI cars can keep their emissions down. You can get wideband EGO sensors, which are fairly expensive, from what I hear. If it's any encouragement, though, I've just tuned my ECU by using an EGO sensor which I got from a wrecker (actually it was free, it was on the exhaust manifold of an engine I bought), and though I have no idea of how accurate it is, it seemed to work OK. Have a good one Andy W :-D ------------------------------ Date: Tue, 18 Jan 2000 12:48:57 +1100 From: Andy Wyatt Subject: Re: DIY_EFI Digest V5 #25 Regarding William Brennan's idea about building an ECU. > I am new to the list, but like everyone I want to build a EFI for my > car. I have read through the archives and I just have a few questions. > I am also new to the Microcontroller world (Haven't bought anything yet, > just a lot of reading) and the likes, so please bear with me. It seems > that the MicroChips are the easiest to start learning on. Am I correct > on this? Yeah, I s'pose so. If you're talking about PICs, you're probably right. You can get C compilers, Assemblers, Simulators and even BASIC compilers for the PICs off the net somewhere or other. I'm not sure that a PIC will have the processing power you require, though. If you're gonna use a lookup table (fuel map) you'll need a fair bit of ROM. Most of the smaller PICs only have 512 bytes or so. You may be better off using a 68HC11, which has lots of stuff built in, like several A/D channels, timers and so on, which are all useful on EFI ECUs. > I would like to update the program via my laptop, so which > chip series might be the best? You'll need something with EEPROM on it. You can get development boards for the 68HC11 with 32kB of static RAM, and 32kB of EEPROM, and you just connect them to your PC to reload the ROM. > What books or software hardware etc etc would I need. All the stuff I used for my computer was commonly available. It helps if you know your micro inside out, so start off with some projects involving the micro you're going to use, then play with them until you fully understand the micro. Otherwise you'll just need to learn all about the sensors and actuators you're going to use on your engine (ie what voltage the sensors give out under different conditions, how to drive the injectors etc) and of course you'll have to figure out an algorithm to determine how much fuel to inject, and spark timing as well (unless you're going to use the old ignition system..) Most of this you can find out as you go (as I did), provided you are cluey and take measurements. With my EFI conversion (from factory EFI to my own EFI) the way I got a basic idea of fuel maps was to hook up a datalogger to the factory EFI computer, and drive around, and log the duty cycle under various conditions. > I know a lot about cars, just not the electronic stuff just yet, but I am willing to learn. Any help would be apreciated. A basic understanding of electronics is essential if you're going to build an ECU. A detailed understanding would be very helpful. Even then you'll be continually learning as you discover new things about the bits you're playing with. It's a long project - be warned.. > And if it matters, I am going to put the EFI on a 1971 Chevelle w/ a 498" > Nitrous breathing rat motor for Drag Racing only, and maybe put the EFI > on my TR7 when I drop the 5.2l rover v8 into it...Thanks!! A TR7 with an EFI 5.2L V8 should go allright... Good luck Andy W :-D ------------------------------ Date: Tue, 18 Jan 2000 15:50:09 +1300 From: Simon Quested Subject: Re: DIY_EFI Digest V5 #25 > The knock is caused by a lack of exhaust flow back to > the intake which cools the fresh charge. How does the hot exhaust gas cool the fresh charge.... Or have I got this wrong and you mean that the exhaust gas isn't heating the cool intake charge and the cold intake charge is causing the detonation???? Either way I'd be very surprised if lack of ERG had anything to do with increased detonation. > Oh that makes sense > Grumpy > Not Positive as ever eh Grumpy =-) ------------------------------ Date: Sun, 16 Jan 2000 22:56:02 -0800 From: janet jensen Subject: Size of (rpm,load) tables Thanks. Good information. I wasn't sure if a bigger table would be beneficial or not. Great to hear from somebody who has been there.. Thanks Greg >Well, they been running the Buicks since 86 with the intercooler, and making >good HP. The syclones since 91. I'd say that the resolution is good >enough. >Ya, might be nicer, but as far as necessary, that's your call. To me 19 >would be perfect, @xxx. > The big isue is running the right MAP so you don't run out of >CALIBRATION. >If ya run 14 PSI of boost run a 2bar MAP. If you run 18 use a 3 bar MAP. >That's how I see things >Grumpy ------------------------------ End of DIY_EFI Digest V5 #27 **************************** 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".