DIY_EFI Digest Wednesday, 17 July 1996 Volume 01 : Number 200 In this issue: Re: RE: EFI power supply Re: RE: EFI Old ALDL Synchronous format Re: power supply fwd: Re: Where is ACH-04-08 ? Thanks. Re: RE: EFI Re: power supply Re: power supply Re: oil control [none] Re: RE: EFI Re: power supply See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: Alexander.M.Lichstein@xxx. Lichstein) Date: 16 Jul 96 05:37:15 EDT Subject: Re: RE: EFI Anyone out there have any solid simple cheap ideas for a 5V power supply for an EFI system. It should have good filtering and be able to supply constant output down as low as 8V and as high as 14V. I hear that transient spikes as high as 200V can be seen after the voltage regulator on an engine - so it should be immune to these as well. Any ideas? I am using a DATEL DC-DC converter which works rock solid through all cases. Unfortunately it is $130. The equivalent PICO unit is of equal cost. I would like a solution for under $25. Is this possible? Thanks in advance. BTW I think I need to be able to drive about 1Amp or so. - - Z ------------------------------ From: Krister Wikstrom Date: Tue, 16 Jul 1996 15:35:31 +0200 Subject: power supply >Anyone out there have any solid simple cheap ideas for a 5V power supply for an I'm making a simple ignition controller and would also like to get some opinions on power supply for automotive application. I have made a simple one like this (hope it is readable): --L---D-------[7805]----- +5V | | | | | C Z EC C Z | | | | | - - - - - The first L and C should filter most high frequencies, the diode and the large EC capasitor prevents sudden drops in voltage, and the first zener ( I have 22V) should ease up voltage spikes. The last zener is just in case, and it actually sits in the CPU board along with some tantal. capacitors. Should this be enough? Krister Wikstrom kwi@xxx.fi ------------------------------ From: walter.kaufmann@xxx.ch (Walter Kaufmann) Date: Tue, 16 Jul 1996 14:26:18 +0200 Subject: Re: RE: EFI >Anyone out there have any solid simple cheap ideas for a 5V power supply for an >EFI system. It should have good filtering and be able to supply constant >output down as low as 8V and as high as 14V. I hear that transient spikes as >high as 200V can be seen after the voltage regulator on an engine - so it >should be immune to these as well. Any ideas? I am using a DATEL DC-DC >converter which works rock solid through all cases. Unfortunately it is $130. >The equivalent PICO unit is of equal cost. I would like a solution for under >$25. Is this possible? Thanks in advance. BTW I think I need to be able to >drive about 1Amp or so. > >- Z The transient problem is known as load dump and is discripted in the data book 'protection devices' from SGS. A device like a zener-diode, named transil (SGS), recommended to cut the overvoltage.... I like 'National Semiconductor'...., they have a voltage regulator in a 11-pin case and 3 outputs: 5V/500mA (or so) for CPU, 5V/100mA for sensors and 5V/20mA (batteriebackup) for memory. Some switching regulator named 'simple switcher' do the job too. 5V/1A in an 8 pin dip without cooling and a coil and a fast diode and two capacitors. (They have some switcher for more than 1A..) See http://www.nsc.com/ for more information Some work but I think, it is cheaper as a DC-DC converter. Any question? Walter ------------------------------ From: Gary W Harris Date: Tue, 16 Jul 96 07:53:00 PDT Subject: Old ALDL Synchronous format Hi all Someone a bit ago requested the information on the old GM ALDL synchronous data format. Here is what I know about it: ; The data frame consists of a start pattern of 9 ones, (the frame ; sync pattern) followed by the actual data. Each data word is ; 9 bits, but the first bit is always a 0. Thus, the leading bit is ; stripped away, leaving the 8-bit data byte. MSB is transmitted first. ; The total frame length is 19 data bytes (?). ; ; ; The following is a list of the bytes in order of reception: ; Note-- the address is controlled by table @ 0d4c5h for type 4E ECU ; an interesting aside--can change RAM/ROM location that is ; monitored by altering address in EPROM table ; ; A/D conversion values based on 256 counts = 5.000 volts (not 255 counts!) ; Thus, 1 bit = 19.53 mv. ; ; ID_HIGH high byte of eprom ID#, location 0d002h ; ID_LOW low byte of eprom ID#, location 0d003h ; IAC current idle air count ; COOLENT coolent temp, deg C= (0.75 x value) - 40 ; MPH miles per hour ; MAP manifold air pressure KPA = (0.3125 x value) + 20 ; RPM RPM = 25 x value ; TPS throttle position, % = (100 x value) / 255 ; INT integrator value ; O2 o2 sensor value, mv = 4.50 x value ; ERR1 error codes byte 1 ; ERR2 error codes byte 2 ; ERR3 error codes byte 3 ; AFSTAT air/fuel status byte ; BATT battery voltage, Volts = value x 0.195 ; IOSTAT descrete IO status byte ; KNOCK knock counter (rolls over) ; BLM block learn mode value ; ALDL aldl counter (rolling--frame cntr?) ; ; ; ; ; Use 2 EPA channels to perform the serial input function. ; EPA0 is set-up to detect a falling edge on PORT 1.0, then ; EPA1 is used to sample the bit value in the middle of the ; the BAUD period. ; ; ; 6.250 mS (=1/160 baud) ; |<------------------------------->| ; | | ; ;---------+ +\\\\\\\\\\\\\\\\\\\\+------+ +\\\\\\ ; | | data here (1 or 0) | | | next data bit ; +-----+\\\\\\\\\\\\\\\\\\\\+ +-----+\\\\\\ ; ; | | | |<--->|<---1.65 mS high always ; | | | ; | | | ; | | +--sample bit here, t=2.400 ms ; |<--->| from falling edge bit start ; | ; +--200 uS start time (low always) ; ; Cheers, Gary ------------------------------ From: "Oliver Scholz" Date: Tue, 16 Jul 1996 17:12:26 +0200 Subject: Re: power supply I used this setup (and it works for my microcontroller application, although I am not completely confident...): > --R-D-----------[7805]----- +5V > | | | > S C C > | | | > - - - The resistor is fairly small (4.7 Ohms) and depends on the max. current of the circuit (voltage drop across R !). The Diode prevents negative spikes from reaching the circuit. Use a fast diode here! S is a Suppressor Diode (P6KE15) that clips fast positive spikes to about 15V. The capacitor C is fairly large (1000 uF) and together with R forms a low pass filter to suppress variations in the supply line. The 7805 also has two 100nF capacitors to suppress oscillation and to filter fast spikes. The output C buffers the circuit. Any improvements will be appreciated... Best regards, - -Oliver ------------------------------ From: Date: Tue, 16 Jul 96 10:48:00 -0500 (CDT) Subject: fwd: Re: Where is ACH-04-08 ? Thanks. Comments By: Steve Ravet@Prj Eng@PCPD Hou Originally To: Originally From: Helmut Forren Original Date: 7/14/96 2:06 AM Comments: There is an accelerometer listed in the EFI reference that I maintain, the ACH-04-08. Someone wrote me the following information, which indicates that this part is probably not suitable for our application. I'm also adding the info to the efi ref list. - --steve Steve Ravet sravet@xxx.com Baby you're a genius when it comes to cooking up some chili sauce... - -------------------------[Original Message]-------------------------- Thanks for keeping information like you do on the ACH-04-08 accelerometers. I did get detailed information and found out that they have an AC response only, no DC response. This means that they are good for measuring "vibration" and "impacts", but not for reading steady-state forces, like the lateral force on a car going around a corner, accelerating, or braking. These are the applications that I need, so the part's useless for me. Also, there is a 25% cross axis sensitivity on one axis. This means that if there's a 1g vibration on the other axis, this one will read a quarter g. That's pretty bad performance in accelerometer-land. You could only use this part in applications where cross axis sensitivity does not matter. If you don't mind, please add this info to the page for others to access. Helmut Forren President Forefront Inc. ------------------------------ From: Sandy Date: Tue, 16 Jul 1996 09:37:11 -0700 Subject: Re: RE: EFI Look at the National Semi book, I am using the LM2940, and the LM2925, which are both Automotive rated, can handle work down to a small amount over the regulated output (less then 1 Volt) are spike and reverse protected. You will have to do a bit more to clean up some of the other Automotive noise, but that should not be too bad. Sandy At 05:37 AM 7/16/96 EDT, you wrote: >Anyone out there have any solid simple cheap ideas for a 5V power supply for an >EFI system. It should have good filtering and be able to supply constant >output down as low as 8V and as high as 14V. I hear that transient spikes as >high as 200V can be seen after the voltage regulator on an engine - so it >should be immune to these as well. Any ideas? I am using a DATEL DC-DC >converter which works rock solid through all cases. Unfortunately it is $130. >The equivalent PICO unit is of equal cost. I would like a solution for under >$25. Is this possible? Thanks in advance. BTW I think I need to be able to >drive about 1Amp or so. > >- Z > ------------------------------ From: "Tony Bryant" Date: Wed, 17 Jul 1996 08:32:34 +1200 Subject: Re: power supply > --L---D-------[7805]----- +5V > | | | | | > C Z EC C Z > | | | | | > - - - - - > > Should this be enough? > Works fine, until you try to start with a partially charged battery. If you are drawing any decent current through it (>500mA), you need an exceptionally large capacitor (6800uF isn't enough for 600mA). The 7805 NEEDS to be on a decent heat sink. I resorted to using a small battery backup in addition, which holds the voltage up across spikes, and the gaps in the ignition switch. Also your input stage capacitor is going to be sucked dry in a matter of microseconds, by the starter motor when you turn it over. Try moving the diode before the cap. If I do it again, I'll be using a switch mode DC-DC convertor chip. *********************************************************** * "Insanity is the only sane response to an insane world" * * >> bryantt@xxx.nz << * *********************************************************** ------------------------------ From: Terry Sare Date: Tue, 16 Jul 1996 17:35:54 -0400 Subject: Re: power supply Krister, Unless you have current limiting installed in line with either zener, you will toast em' with the first spike that comes along. If not, the coil will act as a fuse. Use a transorb or some form of surge protection zener such a a 1.5KEXX (see Digikey catalog). I learned the hard way, with lighting! Use bidirectional devices if available. National makes a line of 3 legged regulators for automotive use. Terry ______________________________ Reply Separator ________________________________ _ Subject: power supply Author: owner-diy_efi-outgoing@xxx.edu at dell_unix Date: 7/16/96 3:35 PM >Anyone out there have any solid simple cheap ideas for a 5V power supply for a n I'm making a simple ignition controller and would also like to get some opinions on power supply for automotive application. I have made a simple one like this (hope it is readable): --L---D-------[7805]----- +5V | | | | | C Z EC C Z | | | | | - - - - - The first L and C should filter most high frequencies, the diode and the large EC capasitor prevents sudden drops in voltage, and the first zener ( I have 22V) should ease up voltage spikes. The last zener is just in case, and it actually sits in the CPU board along with some tantal. capacitors. Should this be enough? Krister Wikstrom kwi@xxx.fi ------------------------------ From: ptimmerm@xxx.GOV Date: Tue, 16 Jul 1996 15:01:41 -0700 Subject: Re: oil control My reading points to the critical nature of the oil return lines to the pan. If this is a home installation, have you checked you oil return is good by running it disconnected. Should come out like a chocholate milk shake. There must be a gravity flow, yet not come in below the oil line in the pan. Best wishes. paul timmerman ------------------------------ From: "Mr. Slate" Date: Tue, 16 Jul 1996 16:20:56 -0600 (MDT) Subject: [none] I realize this request it is not quite in keeping with the theme of do-it-yourself fuel injection, but here goes anyway. I am working on developing a fuel injection system as part of a university project. The system will be used to develop an alternatively fueled version of a production vehicle. The fuel system will be tested and calibrated using an engine dynomometer. I am looking for commerically avaliable engine managment systems (Haltech, Motec, etc.). I would like control of the usual parameters (fuel, spark, etc.) for a V-6 engine with DIS. Ideally, I would like a system which can work in conjunction with the OEM gasoline computer. This would allow me to maintain the stock diagnostics and OBD-II compatability. If anyone could supply company names and phone numbers or addresses I would be most appreciative. Thanks, Neall *************************************************************** Neall Booth nbooth@xxx.ca Mechanical Engineering LPGV Project (Project Leader) University of Alberta HEV Project (Mech. Sys. Leader) *************************************************************** ------------------------------ From: Darrell Norquay Date: Tue, 16 Jul 96 21:20 PDT Subject: Re: RE: EFI At 05:37 AM 7/16/96 EDT, you wrote: >Anyone out there have any solid simple cheap ideas for a 5V power supply for an >EFI system? It should have good filtering and be able to supply constant >output down as low as 8V and as high as 14V. I hear that transient spikes as >high as 200V can be seen after the voltage regulator on an engine - so it >should be immune to these as well. I think I need to be able to drive about >1 Amp or so. >I would like a solution for under $25. Is this possible? You bet. Depending on how handy you are with electronics, Linear Technologies, Maxim, and National all make chips to build up a switching power supply for around $15 for the current output you need. Extremely simple. They will handle inputs from 6V up to around 60V, and for protection against load dump and other spikes you can just use a MOV and either a fuse or polyswitch electronic fuse in series with the input. You may be able to make the unit even cheaper by phoning Maxim or National and requesting a couple of samples. Parts to consider are: Maxim MAX726-5 2A @ 5V fixed/adjustable National LM2575-5.0 1A @ 5V fixed Linear Technologies LT1076 2A @ 5V fixed or adjustable All you need is the chip, 3 caps, one resistor (or in the case of the adjustable units, a couple more resistors), a diode, and an inductor. LT and National parts, as well as inductors and low ESR Caps (important!) are all available from DIGI-Key. I'v used these in several projects now, and they really couldn't be easier to use. There are a few tricks to using switching regulators, so be sure to read the data sheets and app notes thoroughly before applying solder. If you need more info, drop me a line. regards dn dnorquay@xxx.com ------------------------------ From: Darrell Norquay Date: Tue, 16 Jul 96 21:20 PDT Subject: Re: power supply At 03:35 PM 7/16/96 +0200, Krister wrote: >I have made a simple one like this (hope it is readable): > > --L---D-------[7805]----- +5V > | | | | | > C Z EC C Z > | | | | | > - - - - - > >The first L and C should filter most high frequencies, the diode and the >large EC capasitor prevents sudden drops in voltage, and the first zener >( I have 22V) should ease up voltage spikes. The last zener is just in case, >and it actually sits in the CPU board along with some tantal. capacitors. This circuit may suffer from several problems. First of all, you have no means to limit the current into the first Zener. If a spike comes along with sufficient voltage to cause the zener to conduct,, it will likely self destruct immediately. Zeners usually fail shorted, and since you have no fuse or other device to protect the circuit it will likely make big smoke! Use a low value wirewound resistor in series with the input, along with a fuse. The resistor value should be chosen to drop around 3V at full supply current, and it should have a power rating around double it's dissipation at full operating current. This Zener should also be a 5W device. You could also use a MOV (Metal Oxide Varistor) or Tranzorb type device, they are usually rated at higher current than standard Zener diodes. If you use a resistor in place of the inductor, you will still have good filtering, and a resistor is a lot cheaper than an inductor. It may also be a good idea to place a reverse biased diode from + to ground to prevent negative spikes from getting through, although the Zener will also do this to some extent. The second zener is OK for a Fail-Safe, don't limit it's current as you want it to fail in the shorted mode. This may protect your expensive processor chip from damage in case of a catastrophic failure. Use at least a 5.6 V unit, though, to prevent drawing excessive current at normal operating voltage. Even with the power supply filtered as you have shown, you may still have problems with ignition spikes coming in on the power leads, the ground lead, and sensor inputs. This may cause random processor crashes. The solution to this is to run all input and output wires through feedthrough capacitors, grounded directly to a full metal case. Mount the processor away from the engine, and run power and ground leads directly back to the battery. regards dn dnorquay@xxx.com ------------------------------ End of DIY_EFI Digest V1 #200 ***************************** 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".