DIY_EFI Digest Friday, 28 June 1996 Volume 01 : Number 182 In this issue: Re: GM Mass Airflow Sensor Re: Missing teeth Re: AC Delco ECM MAP system Re: AC Delco ECM Re: AC Delco ECM Re: AC Delco ECM Re: AC Delco ECM Re: MAP system RE: AC Delco ECM Re: DIY_EFI Digest V1 #181 Forwarded: Re: AC Delco ECM Problems DELCO ECM Re: GM Mass Airflow Sensor Corvair Efi See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: RABBITT_Andrew@xxx.au Date: Thu, 27 Jun 1996 17:04:58 est Subject: Re: GM Mass Airflow Sensor >>>I'm curious if anybody has a clue as to the upper (frequency output) limit of the currently used GM mass-airflow sensor.<<< does 9600 Hz make any sense? ------------------------------ From: walter.kaufmann@xxx.ch (Walter Kaufmann) Date: Thu, 27 Jun 1996 11:57:05 +0200 Subject: Re: Missing teeth >What I want to know is how exactly do thay know when they are at that datum >mark? For instance, when first starting the engine do they compare the time >between pulses and when they get a much larger time compared to previous >ones they think oh I'm at the datum! I just can't quite get my head around >it at the moment. > >It's probably quite staight forward and logical but I can't see it. > >Thanks in advance, > >Dave Johnson > >email:- dajohnson@xxx.com I think, you are right. But for more details read the "TPU" manual for the Motorola 68332. There is an example for 'missing teeth' and for 'additional teeth'. Walter TPU: Time Processor Unit ------------------------------ From: Jens Knickmeyer Date: Thu, 27 Jun 1996 12:44:59 +0200 (MET DST) Subject: Re: AC Delco ECM wrote: > > I know of no available 6801 disassemblers. I got a programmer at work to > help me making a crude 6801 disassembler. It does the job, but requires some > manual tweaking to get the resulting code right. I guess I need to fix > it...... Scott, a 6809 disassembler is available at http://www.ionet.net/~sriddle/6809dasm.c I have a modified version which is set up to 6801 disassembling, email me directly if you are interested... I am currently searching for a 68HC11 disass, but it seems as if there is no one available. I plan to use the 6809 disas mentioned above, but I 1. have no complete and correct code table for the HC11 and 2. I am too lazy to hack that table in, knowing that there must be such tables somewhere. So, if anyone has such a table around, please send it to me. Table format does not matter as long as there is a short description on it. Thanks in advance, Jens ------------------------------ From: "Jim Staff" Date: Thu, 27 Jun 1996 12:16:19 GMT Subject: MAP system The question all you great theorists are here to answer. In Map based system I measure pressure, however I need to put the unit on the outside of a pipe not to disturb air flow. However this placement gives me only the pressure exerted on the walls of the pipe and not the pressure in the tube. _______ | | | | <-Sensor | | - ----------------------------+-----+------------------------------------------- ^ | <--------------------- Desired air stream Measured Pressure - ------------------------------------------------------------------------------ What is the formula to correct for this problem. My physics teachers couldn't tell me what it was but they knew that in principle it wasn't an impossible problem. In fact some great scientists based thier lives work on this So I know it can be done, however I bet it uses calculus and evil variables that I can't control. So be kind..... Also every MAP sensor I've ever seen is used like this. I checked it out in a ford tempo GX and an Olsmobile firenza.. My new system for an injector is much better, I calculate virtual gearing and I track the mpg.. I'm good-- well decent.. ------------------------------ From: marchil@xxx.net (Alain Marchildon) Date: Thu, 27 Jun 1996 13:10:14 GMT Subject: Re: AC Delco ECM > >Peter- > >Wish I had some info to contribute for you, but I'm a real newbie when it >comes to ECM's. > >I did want to tell you I think it's great that you're sharing all you're >learning about these. I hope to adapt a MAF based system to my '65 turbo >Corvair, so I can do away with the old sidedraft Carter. > >I have a complete EFI setup from a 2.8 V-6 out of a '85 Firebird. I >plan on using a Hobbs switch to enrich the mixture under boost. I have >good fabricating skills and a shop to make the manifold adapters as needed. > >How is the code physically extracted from the ECM? I mean where do you >attach, at the ribbon cable going to the sub board? Where do you get >disassembler software to look at the code? > >Thanks for all your help > >Martin Scarr > > Hi Martin The Hobbs switch is baisicly an on off switch. I would not go for the Hobbs switch to enrich the mixture I am using a FMU = Fuel management unit this is a fuel pressure regulator that you add after the stock unit and it has a vacuum hose to sense the boost pressure in the manifold,. what you want it to do is raise the fuel rail pressure from about 40PSI to close to 95PSI but do it gradually with increase boost pressure this way you will have a linear fuel enrichment . See ya // Marchil@xxx.net // // Alain Marchildon // // 1984 Mazda RX-7 GSL-SE Soon with Vortech supercharger "My toy // // 1989 Audi 90 "Family car" // ------------------------------ From: Martin Scarr Date: Thu, 27 Jun 1996 06:25:52 -0700 (PDT) Subject: Re: AC Delco ECM On Thu, 27 Jun 1996 RABBITT_Andrew@xxx.au wrote: > >>>You need (or at least should use) knock sensor handling on your > turbo-equipped Corvair.<<< > > Knock sensing on an air-cooled engine, now there's a challenge! Why? I've seen several Corvairs with HEI and ESC that work quite well. Martin Scarr ------------------------------ From: Frank Parker Date: Thu, 27 Jun 1996 10:40:59 -0600 (EDT) Subject: Re: AC Delco ECM > > > On Thu, 27 Jun 1996 RABBITT_Andrew@xxx.au wrote: > > > >>>You need (or at least should use) knock sensor handling on your > > turbo-equipped Corvair.<<< > > > > Knock sensing on an air-cooled engine, now there's a challenge! > > Why? I've seen several Corvairs with HEI and ESC that work quite well. > > Martin Scarr > I agree the GM system works OK, but I and others have seen a lot of false retard in GM system. It is very sensitive to proper location of sensor and use old analog filtering. The J&S system is MUCH better. It uses digital signal processing and can be set to retard only the knocking cylinder or to retard all the same. Can use either the high Z GM sensor or the low Z Bosch sensor. Plus I built a LED display of the amt of retard from the unit. On the post of a few minutes ago about a FMU for your projected turbo. They work very well if you do not ask them to add fuel over a large boost range. I have used the Cartech one for 2 projects and works well. They are about 6-7 to 1 on ratio, so fp up 6psi for every psi of boost. I mapped mine on bench with math calculations for density ratio from turbo and trimmed it pretty close. I have one for sale if anyone interested. frank parker ------------------------------ From: Wayne Braun Date: Thu, 27 Jun 1996 08:38:08 -0600 Subject: Re: AC Delco ECM At 10:56 PM 6/26/96 -0400, you wrote: Snip..... > > >I know of no available 6801 disassemblers. I got a programmer at work to >help me making a crude 6801 disassembler. It does the job, but requires some >manual tweaking to get the resulting code right. I guess I need to fix >it...... > >Scot Sealander FIScot@xxx.com > If I remember right, the 6801 is one of the micros that the Dunfield disassembler will recognise and disassemble. I have no idea how good this disassembler is but I have been quite tempted along the way to buy a copy just to try it out. You may want to have a look at WWW.dunfield.com . Another thing is that Dave Dunfield is available by email and every now and then he shows up in some of the embedded micro news groups. Just a thought. Regards Wayne Braun ------------------------------ From: "David M Parrish" Date: Thu, 27 Jun 1996 11:25:12 +0000 Subject: Re: MAP system > From: "Jim Staff" > The question all you great theorists are here to answer. In Map based system I > measure pressure, however I need to put the unit on the outside of a pipe not to > disturb air flow. However this placement gives me only the pressure exerted on > the walls of the pipe and not the pressure in the tube. That's exactly what you want. (I think.) When the sensor is ported into the wall, you're measuring static pressure, which should be nearly the same all through the volume. (Except for possible differences due to the viscosity of air near the surface. I'm not sure that effect is great enough to worry about.) If the port is open into the air stream, you're measuring what's called dynamic pressure, which measures both the static pressure and the pressure exerted by the movement of the air. The air speed indicator in an airplane is a sensitive pressure guage with one side ported to static and the other to dynamic, so it measures only pressure created by the velocity of the air. I'd think the added factor of air velocity would be more a problem than a solution in a MAP system. - --- David Parrish Going with the flow... ------------------------------ From: Peter Shoebridge Date: Thu, 27 Jun 1996 09:14:07 -0600 Subject: RE: AC Delco ECM - ------ =_NextPart_000_01BB6408.FF81E740 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit I have a 6801/6803 disassembler. Not sure of the origin but it works well. Email me directly if any one is interested. Peter "Zed's dead baby, Zed's dead", Varoommmmm snip I know of no available 6801 disassemblers. I got a programmer at work to help me making a crude 6801 disassembler. It does the job, but requires some manual tweaking to get the resulting code right. I guess I need to fix it...... Scot Sealander FIScot@xxx.com snip - ------ =_NextPart_000_01BB6408.FF81E740 Content-Type: application/ms-tnef Content-Transfer-Encoding: base64 eJ8+IgkPAQaQCAAEAAAAAAABAAEAAQeQBgAIAAAA5AQAAAAAAADoAAEIgAcAGAAAAElQTS5NaWNy b3NvZnQgTWFpbC5Ob3RlADEIAQ2ABAACAAAAAgACAAEEkAYAZAEAAAEAAAAMAAAAAwAAMAIAAAAL AA8OAAAAAAIB/w8BAAAAYwAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAGRpeV9lZmlAY291bG9t Yi5lbmcub2hpby1zdGF0ZS5lZHUAU01UUABkaXlfZWZpQGNvdWxvbWIuZW5nLm9oaW8tc3RhdGUu ZWR1AAAeAAIwAQAAAAUAAABTTVRQAAAAAB4AAzABAAAAIwAAAGRpeV9lZmlAY291bG9tYi5lbmcu b2hpby1zdGF0ZS5lZHUAAAMAFQwBAAAAAwD+DwYAAAAeAAEwAQAAACUAAAAnZGl5X2VmaUBjb3Vs b21iLmVuZy5vaGlvLXN0YXRlLmVkdScAAAAAAgELMAEAAAAoAAAAU01UUDpESVlfRUZJQENPVUxP TUIuRU5HLk9ISU8tU1RBVEUuRURVAAMAADkAAAAACwBAOgEAAAACAfYPAQAAAAQAAAAAAAAC2k0B BIABABEAAABSRTogQUMgRGVsY28gRUNNAHEEAQWAAwAOAAAAzAcGABsACQAOAAcABAAWAQEggAMA DgAAAMwHBgAbAAkADAAXAAQAJAEBCYABACEAAAAwOTZDNUQ0NkNDQ0VDRjExQjM5MjAwQUEwMDM4 NUI4NQAvBwEDkAYAqAMAABQAAAALACMAAAAAAAMAJgAAAAAACwApAAAAAAADAC4AAAAAAAMANgAA AAAAQAA5AABEIEk7ZLsBHgBwAAEAAAARAAAAUkU6IEFDIERlbGNvIEVDTQAAAAACAXEAAQAAABYA AAABu2Q7SRVGXWwKzswRz7OSAKoAOFuFAAAeAB4MAQAAAAUAAABTTVRQAAAAAB4AHwwBAAAAEwAA AHBldGVyQGFzcGVucmVzLmNvbQAAAwAGEHj6/xMDAAcQWQEAAB4ACBABAAAAZQAAAElIQVZFQTY4 MDEvNjgwM0RJU0FTU0VNQkxFUk5PVFNVUkVPRlRIRU9SSUdJTkJVVElUV09SS1NXRUxMRU1BSUxN RURJUkVDVExZSUZBTllPTkVJU0lOVEVSRVNURURQRVRFUiIAAAAAAgEJEAEAAAAkAgAAIAIAAIAD AABMWkZ17Co91P8ACgEPAhUCpAPkBesCgwBQEwNUAgBjaArAc2V07jIGAAbDAoMyA8YHEwKDEjMT D2Y0D3poZWzRAyBEbGcCgH0KgAjPxQnZOxefMjU1AoAKgYMNsQtgbmcxMDMUIBcLChLyDAFjAEAg SSABEcB2ZSBhIDY4iDAxLx0RMyBkBACqYQQQZQbQbASQLgewGm8FQHMIcBzQb2YgLnQWIB8QBRBn C4AgYqp1BUBpBUB3BbBrBCD2dxYxHnBFAMADEQeAHaHhF6BjdGx5IDAfMABw/yIwAiAc0AQAIDAC MASQB5DZI1BkLgqFCoVQEgAEkGkKhSJaCYAnBCANsGFGZB/wAaB5LCAlmCL1JoBWCsBvA3AnwiPs CvTwbGkzNg3wG28ewAMA7nALRhQiDAFwA2AjUCIAxyRGCoUcgGtubwfgHyH7LZAc4HYhUQGgHkAd Ax2r9nMecBxxZx6hHPAsAQnAvmEnwASQHOAgVB9AbwqFrRYhcCGCAMBrC4BnHOF9BQB1DbAuzx5h HHAFQGTGbweRH1Jqb2ImgCACWRegcXUh0QQgcwNwZfsKhQOBdQdAH0Ag0DL0MdCPMDASAB9DI3F1 bHQzEucFoDOhH6FodC/0ClAEEY8cgCLQCYA4kmZpeAqF7SBALjyjI+xTBaAFQAZgjwdAAHAEgTAA IEZJPeL0QGEG8C4FoCgPKR8qLxdBiywNFsEARPADABAQAAAAAAMAERAAAAAAQAAHMMBz+wo7ZLsB QAAIMMBz+wo7ZLsBHgA9AAEAAAAFAAAAUkU6IAAAAAADAA00/TcAAJ/t - ------ =_NextPart_000_01BB6408.FF81E740-- ------------------------------ From: talltom Date: Thu, 27 Jun 1996 11:55:52 -0700 Subject: Re: DIY_EFI Digest V1 #181 Alright! Now I can get into this! For starters most of these models were tbi. So is it easy to change to tpi? How? Why 2.0 to 5.7? Why not tubocharged 8.2? Parameters are parameters, if air valves, injectors, map, etc are sized correctly what difference does size make? While we're at it, any suggestions as to how to get around the pulsations inherent in kinky cams? Also need to know the ug level of getting one of these to work as a map type system, as I think maf's are a pain in the butt.(to much plumbing, cost, unreliable, space, not to mention the interface, mapping, sizing etc probs.) Knock sensors in the head? I thought they went by the crank to be closer to where the harmonic is transmitted from the cyl? How do you intend to keep valve noise from making your tune crazy with knock sensor in head? > From: pfenske@xxx.ca (peter paul fenske) > Date: Wed, 26 Jun 1996 08:52:02 -0700 > Subject: AC Delco ECM > > Hello Gang > Just some more info on GM ECM. > FYI the more desirable ECMs are those used in port operation > from about 1986 to 1992. The last #s are 165 and 730 on the > PN. These two ECMs can use a 16K and a 32K respectably. > On the weekend picked up two 165s, one complete tpi harness, and > a firebird aircleaner assembly for 50$ can. 730s a bit harder to > find though. > There are usually several K of unused prom space, as well as several > unused inputs and outputs. GM sensors and actuators are just as cheap. > About the only hindrance so far is untangling the IO. > > So if anybody has any info or clues please let me know > > If someone would like to run a BMW or whatever with one of these let > me know. There are virually hundreds of calibrations for these ECMs. > Anything from 5.7L to 2.0L. > > What I have managed to untangle so far are Spark Control, Idle control, > Fan control, Speed limiter function. Untested as yet are checksum control > and some fueling. If anyone has info to correlate please let me know. > > With the price and availability these ECMS would be the answer for > many EFIers. Also the electronics is far superior to what can be > made and believe me I have access to good PCB manufacturing. > > WEll GL all: peter pfenske@xxx.ca > ------------------------------ From: RABBITT_Andrew@xxx.au Date: Fri, 28 Jun 1996 10:39:32 est Subject: Forwarded: Re: AC Delco ECM Knock sensing is looking for high amplitude signals at certain frequencies within a defined crank angle window to determine whether knocking is occuring. With the larger piston clearances etc. on air cooled engines, the increase in structure-borne noise must make this difficult task extremely difficult. From: c.pugsley@xxx.au:smtp Date: ## 06/27/96 18:28 ## > > >>>You need (or at least should use) knock sensor handling on your > turbo-equipped Corvair.<<< > > Knock sensing on an air-cooled engine, now there's a challenge! > Excuse my ignorance, but why is this difficult? Craig. pugsley@xxx.au ------------------------------ From: pfenske@xxx.ca (peter paul fenske) Date: Thu, 27 Jun 1996 19:23:43 -0700 Subject: Problems DELCO ECM >Alright! Now I can get into this! For starters most of these models were tbi. First off most calibrations are sized in this range for PFI. 8.2s are TBI. I do not have enough info the completely rewrite the code. I doubt anybody outside of GM powertrain does. And they like to keep their jobs. Maybe one guy though he's too smart to open his mouth like me. >easy to change to tpi? How? Why 2.0 to 5.7? Why not tubocharged 8.2? Parameters are There are a lot of trouble codes built in. Exceed some parm and the thing shuts down. Goes into backup not a pretty place. >parameters, if air valves, injectors, map, etc are sized correctly what difference does >size make? While we're at it, any suggestions as to how to get around the pulsations FYI Well GM currently does synchronous reads for MAF and MAP. The old ECMs don't do this. By sampling at the same time and place you can get a pretty good idea . Also filtering. A low pass filter to be exact. It takes five lines of code to make a second order for the MAP. Also a long vacum line to the Map does the same thing. True you lose transient. >inherent in kinky cams? Also need to know the ug level of getting one of these to work Well a modern 0.500 " lift roller cam will pull about 14" of vacum. This is adequate for Map and low 12s. Ask linginfilter He has gone much better with Map. You can always go N-alpha if you are using a 0.800" cam. WOT is about all you will ever use. >as a map type system, as I think maf's are a pain in the butt.(to much plumbing, cost, >unreliable, space, not to mention the interface, mapping, sizing etc probs.) > Knock sensors in the head? I thought they went by the crank to be closer to where >the harmonic is transmitted from the cyl? How do you intend to keep valve noise from >making your tune crazy with knock sensor in head? > True knock sensor location is a science. Thats why the people on the net are worried. However it is possible to adjust the sensitivity both mechanically and electrically. A tech 1 can be used to check for knock activity. Also remember it took GM billions and many millions of manhours to make this work. I most certainly don't have all the answers. I merely wished to get some activity started. Boy did I ever GL all: peter ------------------------------ From: Darrell Norquay Date: Thu, 27 Jun 96 21:01 PDT Subject: Re: GM Mass Airflow Sensor At 11:18 PM 6/26/96 -0500, you wrote: >I'm curious if anybody has a clue as to the upper (frequency output) >limit of the currently used GM mass-airflow sensor. The sensor in >question is the one used on LT1/LT4's, Vortec's, etc... We've tested >vehicles with up to 480 HP and the MAF sensor output is still rising >with airlow demands. >Brock Fraser Sounds like just the unit I'm looking for. Got part #'s, car models, years used etc. to facilitate a junkyard search? I had no idea GM was still using a frequency output MAF. BTW, as far as air flow measurement, you could always dangle it out the window of your favorite ride, and drive down the nearest interstate at 130+ MPH like Peter Wales did! ;] Seriously, this may be a simple way to check how high the output goes. Just go until the output equals your best previous reading, and keep going... (Could always see if you can get a surplus JATO bottle if your ride won't go fast enough :O) Disclaimer: This should only be attempted by professional drivers on closed roads... regards dn ------------------------------ From: pfenske@xxx.ca (peter paul fenske) Date: Thu, 27 Jun 1996 23:17:28 -0700 Subject: Corvair Efi Hi Martin Well had a chance to think about this. First the mechanical side. You will have to fab a long runner manifold for your flat six with the runners into a plenum. This is necessary to use a GM throttle body with TPS and IAC. Also you will need to mount injectors and aim at the proper angle at the intake valve where the manifold mounts to the head. Then you will have to fabricate fuel rails to feed and hold injectors in place. Also you will have to mount bypass regulator. You will need: a 280z or jetta frame mounted fuel pump, mount a return line to your gas tank filler, a knock sensor new type with carbon pellet, not old hi z type, and aldl connector. Also will have to deal with VATS. We might do this in code. Not sure yet. I hope you are a good fabricator Martin. Now as to ecm. Tec, DFI, and Accel would make life easy for you. Much more user friendly than GM. Although I've seen the electronics and GMs much better. Can't go wrong with a few billions for R&D. Also much cheaper. And neater? Now plan B is to rob one of the GNs of their C3 along with the attendant electronics. These guys have a software package called Chipster which takes care of a lot of your problems Martin. Also a lot of the GN lookup is readily available. Albeit the component price is a bit high. A 87 GN motor with all is about 2500$ here in Vancouver. The electronics might be ok. I believe the same ECM is used in other 3.8 buicks. The MAF may be hard to come by cheaply. But you do get sequential and wastegate control. Now to the 1227730 GM ecm. I don't absolutely know all the code. If you get additional help this can probably be accomplished. There are two ways to do this. One is to let the ECM do all the fueling for non boost operation. Then we merely have to fudge the VE maps with the Wot corrector to get the right A-F ratio. Then if you add a fuel rail pressure modifier(fmd) you should have it licked. If you use 5.7 injectors and limit boost to 8 PSI. Mind you we still have vats and no egr to contend with. Solveable but tacky. Also have to watch the Map sensor to avoid an error code. A boost check valve should do it. The more difficult way is to get the 730 to do everything. I do not have the resources to do this. In a year maybe. There is at least 20K of code. You would have to add a two bar Map. This entails rewriting the Map routines of which I am yet unsure of. Also some code must be added. If the runtime is tight we may have to remove some code. Don't need egr or tcc I trust. I believe a lot of people much more competent than me have been working on this for a long time. Maybe they will help you in your endeavor Martin. As to my projects have done a few gm tpi to gm conversions. Pretty safe as long as you install everything. Now have to do a LT1 to 4 runner Yikes this thing wants to talk to 8 other computers and it gets hornery if missing friends. Also built a few proto fuel controllers. Non I would install in a car. I am working on one that I would install in a car. Borrowing a lot from the General on this one. Well tnx everyone for Help: peter ------------------------------ End of DIY_EFI Digest V1 #182 ***************************** 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".