DIY_EFI Digest Thursday, 1 August 1996 Volume 01 : Number 217 In this issue: Re: Circuits for magnetic pickups Re: Circuits for magnetic pickups Re: Proper Mixture. Re: Circuits for magnetic pickups Throttle switch Re: Proper Mixture. [none] Re: Color Tune source Re: Proper Mixture. Specialized Audi salvage yards Re: Circuits for magnetic pickups Re: Proper Mixture. Re: Proper Mixture. Re: Proper Mixture. Re: Circuits for magnetic pickups Re: Color Tune source Re: ColorTune Source Re: Circuits for magnetic pickups See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: Tim Drury Date: Wed, 31 Jul 1996 09:27:58 -0400 Subject: Re: Circuits for magnetic pickups I used the circuit shown in http://spbted.gtri.gatech.edu/hpe/ignitor.gif. It is a +5V only circuit with adjustable threshold. The diode on the input filters the negative voltages (tooth moving away from pickup). If your having problems with noise try using both ends of the pickup in a differential-pair configuration and feed them into the + and - inputs of the opamp. Any noise will show up on both lines and be cancelled out in the opamp. - -tim ------------------------------ From: Donald Whisnant Date: Wed, 31 Jul 1996 09:41:35 -0400 Subject: Re: Circuits for magnetic pickups >> From: "Tony Bryant" >> Date: Wed, 31 Jul 1996 08:43:51 +1200 >> Subject: Circuits for magnetic pickups >> >> I'm using the following circuit for my inductive pickups: >> >> >> >> The requirements I had when designing this circuit were: >> 1) circuit is stable, when pickup is not picking up (i.e. it needs >> hysteresis) >> 2) I can pickup both approaching & receeding teeth (i.e the output >> should be a squared up version of the pickup current) >> 3) 5V only supply >> >> My question is what are other people using? > > From: Gary W Harris > Date: Tue, 30 Jul 96 17:26:00 PDT > Subject: Re: Circuits for magnetic pickups > > You might also look at using the National LM1815 Adaptive Variable Reluctance > Sensor Amplifier chip. It is inexpensive (well, $5.70 from DigiKey) and > operates over an input voltage range of 100mV to 120V! Check it out! > > Gary Tony... The LM1815 that Gary mentioned is good as well as Harris Semiconductor's HIP9020AP... The Harris part was specifically designed as a Vehicle Speed Sensor Buffer chip -- used to pickup the inductive pickup on the output gear of the transmission ... Plus it has an on-chip pulse frequency doubler and some selectable dividers as well... Works with 100mv to 130V ... And is about a $3 (US) part (I think -- as I got all of mine as samples) ... I've tried both the National LM1815 and Harris HIP9020AP and both are very good for this application --- They take a lot of heartache out of the design process... Also, you may want to try a filter cap on the signal. The HIP9020 with the VSS sensor on the transmission requires this! (I can tell you from experience!) -- I had so much noise before putting on the cap that while starting the car (in park of course), the ecm (as it was told by my circuit) thaught the car was traveling at 60-65 MPH!.. This totally confused the computer -- as it must have been there "saying" something like: we're starting the car, the car isn't running, we are in park, we're traveling 60 MPH... hmmm .... It didn't want to start right ... Here is the front end I used with the HIP9020 --- it works like a champ!! : 40K res Pickup >----^^^----*-----> To HIP9020 | --- 0.1uF rated at 250V --- | Gnd Pickup >--- Gnd The resistor is needed as a current limitor and it together with the capacitor form an integrator (i.e. filter) that will get the noise out.... I mention this because the Harris datasheets vaguely mention something about a capacitor, but none of their drawings show it in there... Good luck... Donald Whisnant dewhisna@xxx.com ------------------------------ From: Donald Whisnant Date: Wed, 31 Jul 1996 09:59:10 -0400 Subject: Re: Proper Mixture. > From: Markus Strobl > Date: Tue, 30 Jul 1996 13:24:07 -0500 > Subject: Re: Proper Mixture. > > I wonder if this has changed in the last few years? when I tune my Z28, > I tune it by adjusting fuel pressure until I get about 890-920 mV from > the O2 sensors. This corresponds to about 13:1 A/F ratio. > > Interesting note about GM PCMs: If the fuel pressure (FP) is increased > above 40 PSI, the block learn for idle and cruise will start going to > less than 128 (effectively leaning the mixture). However WOT block learn > will *never* go under 128, and the WOT mixture will be rich. > > On the other hand if FP is decreased below 40PSI the block learn will > increase for idle and cruise to compensate but *also for WOT*, keeping > WOT O2 readings in the 890-920 mV range. > > Man, I'd love to see that source code! IF someone at delco is listening > send it to me and I'll be your life long friend! :-) > While we are talking on the subject of "block learns" and fuel pressures, there is something I've been wondering about with this engine I've been working with... I've recently replaced the 85 3.8L V6 electronic controlled carborated with a 94 4.3L V6 CMFI injection engine in my 85 Grand Prix -- And this 4.3L (because it is CMFI) requires about 60 PSI of fuel pressure. Obviously I had to put a fuel pump on the car (as well as increase the size of the return line)... When the tank starts getting low (i.e. just entering the "yellow" zone on the guage), if I hang a turn too hard, the fuel sloshes to one size and for just a split second, the fuel pump sucks air... Once this air makes it to the engine, it causes a breif hesitation and loss of power -- the lower the fuel level, the worse it is... Once, it was so low (about 2+ gallons in tank!) that it shut off totally when I made a turn at an intersection... ... My philosophy about this situation (and the reason I didn't have it on the old carb system) was that the fuel bowl in the carb preventated gas flow interruption --- plus I assume that the tanks on the fuel injected systems have a well just below the pickup (with the pickup extending into this well) to prevent this... In a way, this is neat, as it is like a "reserve" --- when it starts running out of gas, you definitely know it, but it won't leave you stranded! -- as long as you don't take turns hard, you can make it to a gas station .... My question is what effect does this have on the PCM and/or the engine? For a little instance (unless the poppet valve just doesn't open at all because of the air) the engine (at least a cylinder or too) would be running very lean -- which over time, can't be good on it .. Also, how will the PCM deal with it on the block learn??? Donald Whisnant dewhisna@xxx.com ------------------------------ From: Donald Whisnant Date: Wed, 31 Jul 1996 11:05:06 -0400 Subject: Re: Circuits for magnetic pickups >> From: "Tony Bryant" >> Date: Wed, 31 Jul 1996 08:43:51 +1200 >> Subject: Circuits for magnetic pickups >> >> I'm using the following circuit for my inductive pickups: >> >> >> >> The requirements I had when designing this circuit were: >> 1) circuit is stable, when pickup is not picking up (i.e. it needs >> hysteresis) >> 2) I can pickup both approaching & receeding teeth (i.e the output >> should be a squared up version of the pickup current) >> 3) 5V only supply >> >> My question is what are other people using? > > From: Gary W Harris > Date: Tue, 30 Jul 96 17:26:00 PDT > Subject: Re: Circuits for magnetic pickups > > You might also look at using the National LM1815 Adaptive Variable Reluctance > Sensor Amplifier chip. It is inexpensive (well, $5.70 from DigiKey) and > operates over an input voltage range of 100mV to 120V! Check it out! > > Gary Tony... The LM1815 that Gary mentioned is good as well as Harris Semiconductor's HIP9020AP... The Harris part was specifically designed as a Vehicle Speed Sensor Buffer chip -- used to pickup the inductive pickup on the output gear of the transmission ... Plus it has an on-chip pulse frequency doubler and some selectable dividers as well... Works with 100mv to 130V ... And is about a $3 (US) part (I think -- as I got all of mine as samples) ... I've tried both the National LM1815 and Harris HIP9020AP and both are very good for this application --- They take a lot of heartache out of the design process... Also, you may want to try a filter cap on the signal. The HIP9020 with the VSS sensor on the transmission requires this! (I can tell you from experience!) -- I had so much noise before putting on the cap that while starting the car (in park of course), the ecm (as it was told by my circuit) thaught the car was traveling at 60-65 MPH!.. This totally confused the computer -- as it must have been there "saying" something like: we're starting the car, the car isn't running, we are in park, we're traveling 60 MPH... hmmm .... It didn't want to start right ... Here is the front end I used with the HIP9020 --- it works like a champ!! : 40K res Pickup >----^^^----*-----> To HIP9020 | --- 0.1uF rated at 250V --- | Gnd Pickup >--- Gnd The resistor is needed as a current limitor and it together with the capacitor form an integrator (i.e. filter) that will get the noise out.... I mention this because the Harris datasheets vaguely mention something about a capacitor, but none of their drawings show it in there... Good luck... Donald Whisnant dewhisna@xxx.com ------------------------------ From: Jens Knickmeyer Date: Wed, 31 Jul 1996 19:12:25 +0200 (MET DST) Subject: Throttle switch Hi all! In my Polo-G40, I found the idle throttle switch to open _before_ the throttle plate moved. When I had a friend of mine pressing down the accellerator pedal, I heared the click of the switch and shortly after that, the throttle started to open. I always thought it should be the other way round: the throttle opens a bit and then the switch opens its contacts? The reason for my question is a problem during warm-up. The engine starts perfectly, runs good during post-start enrichment, but then suddenly acts a bit reluctantly. When I press down the accellerator pedal only a little bit, the engine revs 100rpm below its normal idle rpm. Accellerating is weird, the engine tends to stall. These effects vanish when the engine is warmed up. I wonder if the idle switch is adjusted correctly. The VW service men said yes, but they tell a lot of nonsense, so I thought I'd better ask the list members. Any help is apprciated, thanks in advance. Jens ('92 VW Polo-G40) - ------------------------------------ Jens Knickmeyer Technische Universitaet Braunschweig Mikroprozessorlabor 38106 Braunschweig knick@xxx.de - ------------------------------------ ------------------------------ From: agc@xxx. Chichak) Date: Wed, 31 Jul 1996 11:24:04 -0600 Subject: Re: Proper Mixture. At 21:56 7/29/96, John Dammeyer wrote: >Tom Cloud mentioned that he had an Oxygen sensor that he monitored under >cruise. It's no big deal to add a 12 bit A/D to my processor and read the >voltage value. Question is.... how do I interpret this? > >I can see if there is zero O2 left then I have complete combustion for the >amount of O2 in the intake stroke but if I have a too rich mixture wouldn't >that happen anyway? > My bro put an O2 sensor into his Mini and wired it to a recalibrated high impedance panel meter. He was having some problems with misfire at high RPM and figured he was leaning out. The meter immediately told him he was WAAAAAAAAY too rich. He adjusted it to run properly and great googly moogly Powerrrrr. Big problem now was it was too lean to idle. Out came the needle file... Biggest problem using a A/D would be slowing the bloody thing down. On a BS you would definitely see every exhaust pulse, and if you used it for closed loop control you would adjust the hell out of the system. But then of course you, of all people, know about analogue and software filtering and control. On a BS you would have to take a look at the area of the exhaust port and area of the sensor intruding into the gas stream. The sensor has to be close enough to the head to heat up quickly and stay hot, and small enough to not impede the flow excessively. You could alter the shape of the exhaust manifold to have a larger area around the sensor but watch out for a drop in exhaust gas speed. Since BS uses an F head, you will have a problem with fuel wetting of the inlet manifold and the blobby charge response that this gives (as well as updraft manifolds and heads with 180 degree bends). The O2 sensor should give you the feedback you need but you will need to do some heavy filtering. You may have to run it a bit rich to have it work properly. A - -- Andrei Chichak | Information Systems agc@xxx.ca | University of Alberta Hospitals (403) 492 - 4431 (work) | CSB 8-120 (403) 492 - 3090 (fax) | 8440 112 Street Edmonton, Alberta http://cooper-s.uah.ualberta.ca | CANADA T6G 2B7 ------------------------------ From: Don Robinson Date: Wed, 31 Jul 1996 12:10:16 -0600 Subject: [none] To: diy_efi@xxx.edu Subject: Subscription change request Newsgroups: local.mlist.diy_efi Organization: Oasix, Inc. X-Newsreader: TIN [version 1.2 PL2] I didn't find any trace of this list being served by a list server, so please excuse me if there really is a server address. Let me know if there is. Please change my list subcription address from: donr@xxx.com to donr@xxx.com Thanks! DER =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Don Robinson (303) 440-7997 x128 Internet: donr@xxx.com Oasix Inc., Boulder, Colorado UUCP: ...uunet!oasix!donr * I'd rather be roddin',brewin',skiin',shootin',huntin',offroadin',campin'... * ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Wed, 31 Jul 1996 14:18:48 -0500 Subject: Re: Color Tune source Colortune Engine Analyzer $ 44.99 from The Eastwood Co. 800-345-1178 (Canada 800-820-9042) FAX 610-644-0560 http://www.eastwood.com tom cloud ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Wed, 31 Jul 1996 14:44:54 -0500 Subject: Re: Proper Mixture. >At 21:56 7/29/96, John Dammeyer wrote: >>Tom Cloud mentioned that he had an Oxygen sensor that he monitored under >>cruise. It's no big deal to add a 12 bit A/D to my processor and read the >>voltage value. Question is.... how do I interpret this? >> >>I can see if there is zero O2 left then I have complete combustion for the >>amount of O2 in the intake stroke but if I have a too rich mixture wouldn't >>that happen anyway? >> > > The DPM that I just posted (from Hosfelt) reads nicely -- no noise. But then, it isn't going to allow you to feed it to a micro processor. It is true that you will get pulses with each blast of gas from each cylinder, but a simple integrator (read R-C network) ought to fix that. As far as knowing what it means: I would really like to see a data sheet on a Bosch (or other) O2 sensor. The data I have says that the curve is extremely nonlinear, flattening out at about .5 V / 14.7:1 A/F (stoichiometric). Higher voltages is richer mixture (lower A/F), etc. with 1 volt about max and 0 volt being what you'd expect before the sensor gets hot -- except with 10 to 100 M-ohm Z-in the DPM or A/D will always see something, even cold. From my observation: the sensor impedance must drop after it gets hot as my DPM stops wandering. ------------------------------ From: marchil@xxx.net (Alain Marchildon) Date: Wed, 31 Jul 1996 21:43:01 GMT Subject: Specialized Audi salvage yards Hi this is not a DIY question but almost. I am looking for some Audi parts and the salvage yards in eastern Canada do not have them and to buy them new the price is way to high, so my question is are there any salvage yards in the US that specialize only in Audi automobiles any information is welcomed . Thank You // Marchil@xxx.net // // Alain Marchildon // // 1984 Mazda RX-7 GSL-SE Soon with Vortech supercharger "My toy // // 1989 Audi 90 "Family car" // ------------------------------ From: "Stuart Woolford" Date: Thu, 1 Aug 1996 10:10:50 +1100 Subject: Re: Circuits for magnetic pickups > You might also look at using the National LM1815 Adaptive Variable Reluctance > Sensor Amplifier chip. It is inexpensive (well, $5.70 from DigiKey) and > operates over an input voltage range of 100mV to 120V! Check it out! or the phillips sensor for this application, but I cann't remember it's code just not - $30NZ per ($20US) with all the circuitry on board, etc, uses magneto-resistance, and a magnet - can even give you direction in the flasher model (sooo useful on crank.., meant for ABS, I suspect..) - 2 wire 12 volt connection, current mode data, etc. very nice. but then Tony knows about that one... just thought others may wish to know.. If anyone asks, I will dig up the data.. - ---------------------------------------------------------------------- Stuart Woolford, stuart@xxx.nz >>>>In VI Where Available<<<< - ---------------------------------------------------------------------- ------------------------------ From: Frank F Parker Date: Wed, 31 Jul 1996 19:20:29 -0400 (EDT) Subject: Re: Proper Mixture. > A simple and effective indicator for O2 sensor is a DPM (digital > panel meter) from Hosfelt Electronics. It is $14.95, their part > number 39-165 (the part no. on the data says PM-128). It uses an > Intersil ICL7106 DMM chip with a fairly standard wiring. There is > one problem -- they have connected the supply voltage common with > the input common, which can cause ground loop problems -- I solve > this by using a 9 volt battery (actually I was lazy and that was > just the fastest way for me to get it up and running). The 9 > volt battery lasts about 6 months. One could use a simple resistor > and zener or LM-317 to run it off the 12 - 14 volt battery system > (I don't think it wise to run it straight off the auto's battery > system). This is a good meter but I am pretty sure the power supply ground and the measuring circuit ground can not be the same. Thus can not run on car voltage and still measure o2 volt which has the car gnd as common. I have also used a 9 volt battery. That's why they are cheap. frank > ------------------------------ From: FIScot@xxx.com Date: Wed, 31 Jul 1996 22:54:33 -0400 Subject: Re: Proper Mixture. In a message dated 96-07-31 10:42:31 EDT, you write: >My question is what effect does this have on the PCM and/or the engine? >For a little instance (unless the poppet valve just doesn't open at all >because of the air) the engine (at least a cylinder or too) would be >running very lean -- which over time, can't be good on it .. Also, how >will the PCM deal with it on the block learn??? I would not worry about what the ECM is doing..... What about running the pump dry? The ECM won't change dramatically in a couple seconds, and if you are on/off the throttle, probably won't change at all. I would think that running the pump dry might shorten it's life. The Mike Knell book on Chevrolet TPI and TBI Engine Swapping had a neat idea to stop engine from dying in this situation. He made a little reservoir box to hold fuel that fed the high pressure pump. A low pressure pump fed the box from the tank. Requires two fuel pumps, but solves the problem if you can't get a baffled fuel tank. Scot Sealander FIScot@xxx.com ------------------------------ From: Alexander.M.Lichstein@xxx. Lichstein) Date: 31 Jul 96 23:16:20 EDT Subject: Re: Proper Mixture. Everybody should keep in mind that the (so called) lambda sensor is not measuring O2 or A/F at all. What it is is a galvanic cell which (at temperature) will generate a voltage from EACH and EVERY component of the exhaust gases. The voltage contribution of each of these is proportional to their partial pressures, as well as some constants relating to their size (permittivity of the sensor), and their electronegative potential. It so happens that during normal combustion, the combined EMF's of exhaust gases corresponding to 0.45V is about the stoichiometric limit. As you get richer from there, the different HC's present as well as the added CO etc. will contribute unpredictably (at this point) to the voltage produced by the cell. This includes the different fuel components (aromatics, ene's, yne's, anes...) which may be present, as well as the rate of drop-off of NOx production et. al. If we could predict, at every temperature and operating condition of THAT engine what these particular contributions would be, then we could use these voltages in the rich (closer to 1V). At this point in our understanding of combustion, we are unable to predict these complex dynamics in a general sense. It may be that some of you have found that you can repeat values which MAY correspond to exhaust compositions which correspond to desirable A/F ratios under AVERAGE conditions, but these should not be considered "REAL". There is a really dense SAE paper about this topic which is a good read.... number to follow. Don't lose heart though... some of these averaging and "slurring" routines and filters may give an indication and could be useful for set-up as environmental conditions change. Further the sensors are real good at telling you to which extreme you are if things just aren't right. - - Zander Feel free to shoot me down. I need to know more for my project. ------------------------------ From: Sandy Date: Wed, 31 Jul 1996 20:19:40 -0700 Subject: Re: Circuits for magnetic pickups At 09:44 PM 7/30/96 PDT, you wrote: >At 08:43 AM 7/31/96 +1200, Tony Bryant wrote: >>I'm using the following circuit for my inductive pickups: > If anyone get a good pickup going (no extra pulses) that uses the simple op-amp, let me know. I would love to remove the lm1815 from the design. I am far into the routing of the board, but I have a couple of extra amps in the LMC6494 that I am using for O2 buffers. So I you guys get some good thig working, shoot me some mail. Sandy ------------------------------ From: garfield@xxx.com (Garfield) Date: Thu, 01 Aug 1996 02:30:00 GMT Subject: Re: Color Tune source On Wed, 31 Jul 1996 14:18:48 -0500, you wrote: >Colortune Engine Analyzer $ 44.99 from The Eastwood Co. > >800-345-1178 (Canada 800-820-9042) FAX 610-644-0560 > >http://www.eastwood.com > Thanks for all the info you've been posting to efi....but this URL doesn't seem to be right. In fact, I do a lookup using WhoIs on eastwood and don't get anything resembling this address. Garfield ------------------------------ From: Michael Fawke Date: Thu, 01 Aug 1996 15:48:12 +1100 Subject: Re: ColorTune Source On Wed, 31 Jul 1996 14:18:48 -0500, Garfield wrote: >>Colortune Engine Analyzer $ 44.99 from The Eastwood Co. >> >>800-345-1178 (Canada 800-820-9042) FAX 610-644-0560 >> >>http://www.eastwood.com >> >Thanks for all the info you've been posting to efi....but this URL >doesn't seem to be right. In fact, I do a lookup using WhoIs on eastwood >and don't get anything resembling this address. > >Garfield I just had a quick look, it should have been - www.eastwoodco.com Michael Fawke Canberra, Australia. fawkacs@xxx.au ------------------------------ From: "Oliver Scholz" Date: Thu, 1 Aug 1996 09:24:35 +0200 Subject: Re: Circuits for magnetic pickups Hi all, regarding the recent discussion of magnetic pickups I have a question: On older transmissions GM used a PM generator as the Vehicle Speed Sensor. It outputs a sine wave and about 4000 pulses per mile. The later model transmissions (it is Getrags I am particularly interested in) they use a hall sensor of some kind that is triggered by a slotted disc on the axle. That signal is run to the ECM/PCM which drives the speedo at the usual 4000ppm. Now, I would like to use the new transmission on an older car, so I would have to run the Hall sensor signal into a microprocessor that generates the proper speed signal from it. Now the question: Does anybody know the waveform of the signal? Is it sine, square, ??? Which circuit would be recommended to buffer it? Also, how many pulses does it generate per revolution of the drive wheel? Any help would be appreciated! Best regards, - -Oliver ------------------------------ End of DIY_EFI Digest V1 #217 ***************************** 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".