DIY_EFI Digest Tuesday, 15 October 1996 Volume 01 : Number 314 In this issue: Re: Air Flow Meter Cap discharge Re: L-Jet Re: Cap discharge Re: Air Flow Meter Re: Volume of fuel injected Re: Volume of fuel injected Re: IGNITION PROJECT Re: IGNITION PROJECT K-Jetronic with lambda oscillations Forwarded: Re: L-Jet Re: Volume of fuel injected Gm Controllers Re: Cap discharge efi-332 web site See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: "Bosch, AN, Andrew, Dr" Date: Mon, 14 Oct 1996 12:09:55 UTC-2 Subject: Re: Air Flow Meter > > I have serious doubts about this bigger flow meter stuff. Before you > embark on fitting another flow meter, I think it would be a good idea to > determine if the flow meter is offering any significant restriction. Don't > just assume that it is. Do a pressure drop test accross it. It does not > make sense that the manufacturer would fit one too small. The vane would > bottom out before full flow was reached. The flow meter is the systems > only load sense so the ecu could not determine fuel requirements beyond > that point. Result? High speed lean out and a big warranty claim. I don't > think they are that stupid! > > As for the throttle body being too small and the inlet runners as well, I > think you overlook a significant number of facts. Inlet tract size (length > & diameter) is based on gas velocity requirements, cylinder volume and > sound wave propergation. It was demonstrated 25 years ago that high power > output did not equate to the biggest possible valves, inlet passages and > throttle size. In my not so humble opinion, I think you are embarking on > the degradation of your low speed performance and drivability. There is no > point in having an inlet tract that will pass x+y cfm when the engine will > barely consume x cfm at full noise. Get my point? Hi again Sorry, I wasn't very specific. The reason for all this "A/F meter too small" stuff is that this motor is fairly modified: with Weber 45 sidedraughts it makes 170 KW (~230 HP). The stock fuel injection can't match that, so I've been looking at ways of getting more fuel (variable rate of gain fuel pressure regulator or extra injectors and controllers) and/ or more air. Ultimately, what I really would like to do is get an after-market FI system with individual throttle body injection Andrew Dr A. N. Bosch Physiology Department/ Sports Science Institute University of Cape Town Medical School P. O. Box 115 Newlands 7700 South Africa ------------------------------ From: Mark Pitts Date: Mon, 14 Oct 1996 11:27:53 +-100 Subject: Cap discharge Any body know where to start looking for a cap discharge schematic, or any other info for that matter. Mark ------------------------------ From: Frank Mallory Date: Mon, 14 Oct 1996 08:36:38 -0700 Subject: Re: L-Jet RABBITT_Andrew@xxx.au wrote: > > >hot-wire is the most accurate but also the most expensive (read > >hi-tech), reverse for manifold pressure (MAP)) > > I disagree, Honda is using MAP for their ULEV Accord, and saying > things like 'accurate A/F ratio control' in the same breath. I hear that Mercedes is abandoning hot film due to reliability problems, and going back to hot wire. Does Honda use an oxygen sensor with the ULEV? ------------------------------ From: peter paul fenske Date: Mon, 14 Oct 1996 07:22:21 -0700 Subject: Re: Cap discharge Hi Mark Not too much help but from memory heathkit used to make a CD box. Also check Radio electronics mag many years ago. Cd boxes have gone a bit into history. Generally a switching supply is used to generate 300 volts or so D.C. Then a scr triggered by points is used to transfer this voltage to the primary of the ignition coil. GL: peter At 11:27 AM 10/14/96 +-100, you wrote: >Any body know where to start looking for a cap discharge schematic, or any other info for that matter. > >Mark > > > ------------------------------ From: Jim Hampton <104253.664@xxx.COM> Date: Mon, 14 Oct 1996 13:13:22 -0400 Subject: Re: Air Flow Meter There is a system available which uses throttle bodies which are replacements for Weber carb bodies. It is a bit expensive and not exactly DIY. All the parts except the controller are available from TWM Induction in Goleta, California, for about $4000 US. Their phone number is (805)967-9478. They will send a catalog for $5. If you come up with a more economical system please mail me direct with info as I will not be on the mailing list for long. I am hoping to do much the same thing with an engine using Weber downdraft throttle body replacements. Jim Hampton 104253.664@xxx.com ------------------------------ From: Sandy Date: Mon, 14 Oct 1996 10:34:01 -0700 Subject: Re: Volume of fuel injected Their is a couple if easy drive circuits if you want to drive the high impedance injectors, one that is real easy is using a logic level FET, that with a diode is almost all that is needed. If you want to drive the low impedance injectors, you can use the cherry part, or the national lm1949 with a tip121. Look on the national web page for datasheets and SAMPLES of the lm 1949. Sandy At 09:41 PM 10/13/96 -0400, you wrote: >I'm very interested the design of the injector drive circuit. >Does anyone have a design they would like to share? >Has anyone seen the design Ed Lansinger had in his September 95 Circuit >Cellar INK >article? (i.e. TIP120 power Darlingtons) If so, what do you think about that? > ------------------------------ From: Stephen Dubovsky Date: Mon, 14 Oct 1996 15:18:39 -0400 Subject: Re: Volume of fuel injected If your reffering to the drivers they used when injecting the FZR600 motorcycle engine (from ckt cellar inc), I think they are pretty lousy (although they do work). If you cant find anything else i could help sketch a better one. The use of 'better' is relative. Do you want an ultra simple ckt, something a little more complicated but w/ lower injector hold current, or something really high performance that has very little power loss? Does anyone know the inductance of an injector? hold current? At 09:41 PM 10/13/96 -0400, you wrote: >I'm very interested the design of the injector drive circuit. >Does anyone have a design they would like to share? >Has anyone seen the design Ed Lansinger had in his September 95 Circuit >Cellar INK >article? (i.e. TIP120 power Darlingtons) If so, what do you think about that? > ------------------------------ From: Mark Mason Date: Mon, 14 Oct 1996 16:46:04 -0400 Subject: Re: IGNITION PROJECT I'm currently working on a system similar to the one described in the previous messages under the "Re:IGNITION PROJECT" subject heading. I have a couple questions for anyone with some experience or knowledge in this area. 1) Ignition advance on my controller will be compensated when a change in temperature occurs. Should I sense coolant temperature or air temperature? I know I have to reduce the amount of advance as the temperature rises. Does anyone have a graph/equation for the rule of thumb advance compensation versus temperature? 2) Should I bother with temperature compensation or just use the knock sensor as the "feedback"? Then the uP could learn the change in advance required and store it so after some running time it would look up the compensation value. 3) What is the best method to reduce engine speed when a rev limiter is utilized? Retarding the timing is one suggestion but is it fast enough to protect an 500 HP engine revving 6000 RPM when a sudden big blast of Nitrous Oxide is shot into it and the tires break loose? I think removing the spark from every other cylinder would be a bad idea also ( cylinder wall wash among other things ). Any Ideas? Any help or suggestions would be much appreciated! Mark. ------------------------------ From: jac@xxx.us Date: Mon, 14 Oct 96 17:00:51 PDT Subject: Re: IGNITION PROJECT In my system the rate of change in advance is limited by the response of the MAP sensor and one A/D conversion cycle. The system clculates the advance for the ignition B just after loading the timer for ignition A. It is always 1/2 revolution behind the measured pressure. This is pretty fast. I expect that most systems would react as rapidly. I doubt that the MAP has a time constant this short, no matter how fast the throttle is opened. Consider an additional input indicating WOT + Nitro and program an exact advance if the input is present. As a rev limiter it is effective to drop out all of the advance, but because my units go into airplanes, there is never a possibility of "clutch in - no load" situation. Can you cut out the Nitro and advance and run rich with just air to limit revs? - ------------------------------------- john carroll jac@xxx.us - ---------------Original Message--------------- I'm currently working on a system similar to the one described in the previous messages under the "Re:IGNITION PROJECT" subject heading. I have a couple questions for anyone with some experience or knowledge in this area. 1) Ignition advance on my controller will be compensated when a change in temperature occurs. Should I sense coolant temperature or air temperature? I know I have to reduce the amount of advance as the temperature rises. Does anyone have a graph/equation for the rule of thumb advance compensation versus temperature? 2) Should I bother with temperature compensation or just use the knock sensor as the "feedback"? Then the uP could learn the change in advance required and store it so after some running time it would look up the compensation value. 3) What is the best method to reduce engine speed when a rev limiter is utilized? Retarding the timing is one suggestion but is it fast enough to protect an 500 HP engine revving 6000 RPM when a sudden big blast of Nitrous Oxide is shot into it and the tires break loose? I think removing the spark from every other cylinder would be a bad idea also ( cylinder wall wash among other things ). Any Ideas? Any help or suggestions would be much appreciated! Mark. - ----------End of Original Message---------- - ------------------------------------- john carroll jac@xxx.us ------------------------------ From: barabap@xxx.org (pietro barabaschi) Date: Mon, 14 Oct 1996 17:48:31 +1000 Subject: K-Jetronic with lambda oscillations Hello, i'm new to the mailing list therefore please excuse me if i make a question which has been already treated in the past. I have a ferrari308QV which runs with a K-Jetronic with lambda fuel injection system. the problem i notice is that when the lambda is connected (with the computer in closed loop) the air/fuel mixture is oscillating resulting in a corresponding oscillation of the engine speed. at first i though that the sensor was gone. changed the sensor: nothing changes. measuring the voltage at the sensor i see that it jumps between lean and rich rapidly with a peiod of ~1s. the frequency valve correspondingly follows. the problem goes away when i disconnect the sensor and the computer goes in open-loop automatically. one of the problems is that i have no real technical book that describes the controller employed by the system. i can only assume that it is a PI controller but i do not know is the controller is completely digital or if there's an analog part as well. now the questions: - -does anybody know what may be faulty? - -any bosch technical literature on the K-Jetronic computer? if yes where can i find it. - -is it possible to change the control gains in the K-Jetronic? I appreciate any response . regards. Pietro Pietro Barabaschi ITER-San Diego JWS 11025 N.Torrey Pines Rd.- LA JOLLA-CA 92037 USA [Tel:(1)(619)622-5156] [Fax:(1)(619)452-8163] ------------------------------ From: RABBITT_Andrew@xxx.au Date: Tue, 15 Oct 1996 08:54:28 est Subject: Forwarded: Re: L-Jet Honda would be using some form of HEGO, and I'd not be surprised if they're doing some DSP to determine cylinder to cylinder A/F ratio variations also. re: Hot Film meters, BMW have the Bosch Hot Film MAF across (most) of their range here in Oz, and I assume everywhere else too. Many other European manufacturers are using this meter too, and on recently introduced models. I don't think it's losing much ground to any hotwire MAF From: (Frank Mallory) frankm@xxx.com:smtp Date: ## 10/14/96 08:36 ## RABBITT_Andrew@xxx.au wrote: > > >hot-wire is the most accurate but also the most expensive (read > >hi-tech), reverse for manifold pressure (MAP)) > > I disagree, Honda is using MAP for their ULEV Accord, and saying > things like 'accurate A/F ratio control' in the same breath. I hear that Mercedes is abandoning hot film due to reliability problems, and going back to hot wire. Does Honda use an oxygen sensor with the ULEV? ------------------------------ From: Sandy Date: Mon, 14 Oct 1996 22:18:14 -0700 Subject: Re: Volume of fuel injected > Does anyone know the inductance of an injector? hold current? The saturated injectors draw 1 amp, so they should be about 12 Ohms, and the low impedance injectors need to be 4 amp open, 1 amp hold. Look at the EFI332 archives for a ton of discussions. Also on the FTP site you will 'eventually' find some schematics that I did. They tested very well, and if using the lm1949's they seem to work very well at driving an ignition coil when using an IGBT. I have laid out a universal circuit board that will hold 4 drivers that can work with either injection or ignition. Rev 1.00 is floating around, but it has since been updated, altho I have not posted it... Sandy ------------------------------ From: Jennifer Rose Date: Mon, 14 Oct 1996 22:33:42 -0700 Subject: Gm Controllers Hi Peter Thanks for the infomation. My project is 74 chevy pu with a TPI unit out of an 86 Camaro. The pu has been running for 2 years with the TPI unit. The main problem is hot restarts on a warm day. Wanted to remove the cold start injector and replace software from a 89 calpak. Other big problem is the negative back pressure EGR valve sets a code 32 under loads. Also wanted to change curves to better suit a pu. Any help would be wonderful. Thanks Vance ------------------------------ From: William Boulton Date: Tue, 15 Oct 1996 16:43:00 +1000 (EST) Subject: Re: Cap discharge Mark, In July of 1975, Electronics Australia magazine did project for a CDI. I purchased a kit a couple of years later from Dick Smith Electronics and have been using it ever since (not 1 vehicle). I'll give you the details and you may be able to (at least) get a copy of the original documentation from DSE even though the kit has not been current for a number of years. Kit #: K-3280 Snail Mail Addr: Dick Smith Electronics PO Box 321 North Ryde NSW 2113 Australia The original documentation did not describe how to wind the transformer since it came pre-wound in the kit. If they can't help you, the only other course open would be to get a copy of the project article from EA at about A$7.50. Their address (snail mail) is; The Secretary Electronics Australia PO Box 199 Alexandria NSW 2015 Australia Some points I have learned about CDI systems. 1 They don't like lean mixtures. Spark duration too short. 2 They work much better in combustion chambers with good swirl & distribution. My unit was quite remarkable on a '77 Honda Civic but very ho-hum on a '78 2850 Torana (sorta like 3.? litre Chev 6 of the same period). 3 Slight misfire at very light throttle (slight over run). Don't know why but I've seen it on other units too. 4 Great cold starting. 400v on the primary will start anything. 5 They will reliably fire the worst fouled plugs you will ever see. I had one plug so fouled I couldn't dig the crap out and it kept on firing. 6 Great on engine that rev like stink. No HEI coil will saturate in under 3 ms while a CDI will charge its capacitor to 400v in far less time and they draw stuff all current. That's why some small bike engines use them as do the OZ 5l Group-A cars. So do Mercury outboards. Thats it, bye Bill Boulton On Mon, 14 Oct 1996, Mark Pitts wrote: > Any body know where to start looking for a cap discharge schematic, or any other info for that matter. > > Mark > > ------------------------------ From: William Boulton Date: Tue, 15 Oct 1996 17:21:15 +1000 (EST) Subject: efi-332 web site Hey, is any one else having difficulty connecting to the efi332 web site. I've trying for 2 weeks and nothing at all. Here is the address I've been using. Tell me if it is wrong. http://www.cim.swin.edu.au/~aden/web-docs/efi332/332_index.html This is getting bothersome. Thanks Bill Boulton. ------------------------------ End of DIY_EFI Digest V1 #314 ***************************** 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".