DIY_EFI Digest Friday, 10 May 1996 Volume 01 : Number 135 In this issue: Re: search VVT MAF sensors Re: 8748 add'l injector controller design re: Re: Variable cam timing Re: Injector controller (attached) Re: VVT Re: MAF sensors Re: Variable cam timing Re: search Re: Variable cam timing Re: someone willing to sell cs453 inj drivers Re: MAF sensors Re: Doppler shifted MAF sensors Re: Doppler shifted MAF sensors Re: 8748 add'l injector controller design Re: Injector controller (attached) Reply to: Re: Doppler shifted MAF sensors Re: Injector controller (attached) See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: Frank Parker Date: Thu, 9 May 1996 07:42:05 -0600 (EDT) Subject: Re: search > Subject: search > > Has anyone heard of a company called "Simple Digital Systems" > > I was in search of a way to contact them, phone number, address etc. > > > Thanks > > Steve > Look in Turbo mag. They have ads in there. I'm @xxx. Email back if can not find. fp ------------------------------ From: einarp@xxx.no (Einar Sjaavik) Date: Thu, 9 May 96 13:50:09 CET Subject: VVT >what essentially happens here is >that the cam gear is not spinning at a constant speed, rather varying >sinusoidally. Seems you just described the MGF VVT. The cam does one rev per 2 crank revs as usual. But it's changing the instantaneous speed during one rev. - -- einarp@xxx.no ( Maserati Biturbo Spyder ) ------------------------------ From: Mark Pitts Date: Thu, 9 May 1996 12:18:51 +-100 Subject: MAF sensors Anybody interested in playing with an ultrasonic, dop shift MAF sensor? I saw an idea somewhere, probably a year ago, for a MAF sensor that used = an ultrasonic transducer pair to measure airspeed by dop shift... seemed = like a very simple idea, and had the advantage of being relatively = unintrusive to the air flow. Or for that matter does anybody out there = know of a production version, or any more reserch into the idea??? Mark ------------------------------ From: michael glidewell Date: Thu, 09 May 1996 09:29:50 -0400 Subject: Re: 8748 add'l injector controller design At 07:53 PM 5/6/96 -0700, you wrote: > > I have completed the breadboard of my 8748 controller which drives >two 60 lb/hr injectors (using MC3484 peak/hold drivers). It works great >in my BMW 745i (factory intercooled turbo, 7-8 psi). I am now running >16-17 psi boost with no detonation - in fact, I think my fuel map is a >little too rich! (no oxygen sensor). > I have laid out the artwork for a two-sided PC board using Eagle >2.6, but have not had it fabricated yet (the first two boards are the >expensive ones - about $250-300 ;) It's not a very flexible design (map >is stored in EPROM), but it _is_ simple, with a low component count, >and the software is only about 100 bytes long (plus the 256 byte map). >The only expensive item (other than the PC board ;) is the 0-30 psig >pressure sensor with 0.5-4.5 volt output - cost me $43 from OMEGA. > Anyway, if anyone's interested I can post a ZIP file with the >schematic and the artwork (printable on a HP laser printer). It's only >about 60Kb compressed. Usual disclaimers about blowing up CPU's >and engines at your own risk, etc. > Next revision is to add a multiplexer to the A/D input (from the >pressure sensor) so that an oxygen sensor may be used for closed-loop >control. >-Charles \ Sounds very lilke a very ambitous project! Is the 745i an L-jetronic or Motronic? What did you actually replace on the stock unit? Ihave a 528i L-jet which I would like to improve. - ---------------------------------------------- Michael Glidewell Boston, Massachusetts glide@xxx.net KE1CP BMW 528i BMWCCA http://www.shore.net/~glide/ ------------------------------ From: ehernan3@xxx.com (Edward Hernandez (R)) Date: Thu, 9 May 1996 10:29:33 +0500 Subject: re: Re: Variable cam timing Sorry guys, I haven't been able to access email for a while. Apparently, I left some of you slobering: "Ed, are you going to say anything else about this, or are you going to let your remark just lie there???? :-)" "Come on Ed, don't tease us like that. I remember a mechanical type described in Motor Trend a while back. The chain end of the camshaft was threaded, and the sprocket screwed onto the thread. There was an oil driven piston that pressed against the sprocket. As the piston moved the sprocket back and forth, it caused the camshaft to be advanced or retarded via the threads. It was a very coarse thread." Nope, you missed my hint completely. You describe the "common" way of phasing cams that most OEMs have played with. Some have made it to production(Nissan, M-B, Ford, et al). "Sounds like you guys are cookin sumpin' up over there at Ford..." You are right, except that the beanies told us to can it and put it up on the shelf for later use. Happens too often over here. No more hints, except to say the first post was very, very close to what we tried and succeeded in achieving. Anything that controls the camshaft as a whole component cannot be described as having infinitely variable duration on an engine where the cam controls the valves on more than one cylinder. Changing the duration during a lift event for one cylinder will compromise the valve events on all other cylinders, unless perhaps you have a two cylinder engine which the K-series aren't. Hydraulic bleed systems(and systems that phase the entire cam) can be used to achieve real BSFC improvements, not penalties. The way to do this is to delay intake closing at the low rpm. This kills torque, so the throttle must be opened further which reduces pumping losses and improves BSFC. Well designed strategies will base cam phasing on throttle position as well as rpm. In this way, cam phasing can give you the best of both worlds, not just performance improvements. "This would change only the timing, not the shape of the lobe. Controlling the lift and duration of the valve would change the performance much more dramatically. Why not use a solenoid valve to bleed off pressure from hydraulic lifters? The cam could be ground much more radical and the computer could easily bleed." Not necessarily true. Of the four events(EVO, EVC, IVO, IVC, where the initials are Exhaust or Intake, Valve and Closing or Opening), IVC is the most dominant. Substantial gains have been achieved with cam phasing alone. In order to bleed lifters, the solenoid you mention must reside inside the lifter/lash adjuster. If you put it in the oil galley where the lifters reside, the lifters must be designed "loose" in order to respond to varying galley pressures which makes them very noisy(okay, so that's really an OEM concern, not a DIY concern). Still, the computer cannot easily control this as you mentioned unless there is some form of feedback on EACH valve. Variable bleed lifter technology isn't to the point yet where you can count on it without feedback. The manufacturing tolerance required for REPEATABLE bleed rates needed for a system like this are not achievable(for mass production OEMs) yet. And no one has invented true constant viscosity oil yet. The computer would have a tough time, not an easy one. Cam phasing systems just need a cam position sensor in conjuction with a crank position sensor, and these already exist on SEFI engine management systems. Sure, you CAN get more out of a variable profile system, but I'm not sure I would call it dramatically better, espec- ially if your cam phasing device can swing the cam 30 or 40 degrees already. I would imagine the DIY crowd would be interested in trying this in it's simplest form. Oops, I babbled again. Ed Hernandez Ford Motor Company ehernan3@xxx.com PS I hope I didn't forget any carriage returns, dn. ------------------------------ From: Wayne Braun Date: Thu, 9 May 1996 08:35:21 -0600 Subject: Re: Injector controller (attached) At 07:57 PM 5/8/96 -0400, you wrote: >> > Can this be decoded as I do not have MIME compatable tools, >> > Email : dchannon@xxx.au >> >> What exactly does this mean? All you need is a copy of PKUNZIP >> (shareware). Can someone help out David? The unzipped file is almost 2 >> Mb...inefficient plotting software, I guess, but that's too long to >> send! > >Hi! The files are available from: > >ftp://efi332.eng.ohio-state.edu \ >/pub/diy_efi/reader_contributions/inj_controller_cmorris@xxx.com > >just ftp them. > >______________________________________________________________________ > Daniel Ridge | USRA CESDIS > Technical Specialist | Code 930.5 > email: newt@xxx. W274 > tel: 301-286-4403 | Goddard Space Flight Center > fax: 301-286-1777 | Greenbelt, MD. 20771 >_____________________________________|________________________________ > Yes the files are available, but unfortunatly they appear to be in a format that many cannot read/decode. If some one would be so kind as to make this available as a normal .zip file it would be appreciated. BTW: Pkunzip, Winzip, Extract, & Esscode all do not recognise the format that this file is in. Thanks Wayne Braun ------------------------------ From: Bruno! Date: Fri, 10 May 1996 00:49:43 +1000 (EST) Subject: Re: VVT On Thu, 9 May 1996, Einar Sjaavik wrote: > >what essentially happens here is > >that the cam gear is not spinning at a constant speed, rather varying > >sinusoidally. > > Seems you just described the MGF VVT. The cam does one rev per 2 crank > revs as usual. But it's changing the instantaneous speed during one rev. That's right. How does the MGF VVT work? Bruno. (u933234@xxx.au) ------------------------------ From: Bruno! Date: Fri, 10 May 1996 01:04:53 +1000 (EST) Subject: Re: MAF sensors On Thu, 9 May 1996, Mark Pitts wrote: > Anybody interested in playing with an ultrasonic, dop shift MAF sensor? What about hacking one together? I just happened to do a bit of reading into mass flow, and the idea behind these is quite simple, like as you said, working on the doppler shift. it looks something like below. \tx1\rx2\ ================== ================= - ------> air flow ========================= =========== \rx1\tx2\ one sound wave is opposed by the flow (pair #2), reducing speed in moving air, while the other is assisted (pair #1). you should be able to find quite a few journal articles, books, under the subjects: anemomtery, gas flow, flow measurement, etc. Bruno. (u933234@xxx.au) "A sewer rat might taste like pumpkin pie, but I wouldn't know cos I don't eat the filthy muthaf***a" ------------------------------ From: "Kenneth C. King" Date: Thu, 9 May 1996 10:15:22 -0500 (CDT) Subject: Re: Variable cam timing On Mon, 6 May 1996, Bruno! wrote: > Hi everyone, > I've been thinking about the subject of variable cam timing, which seems > to come and go on this group. The idea of electromagnetically actuated > valves may be prohibitive in the amount of power it would use from the > electrical system, what about a synchro-type system where the amount of > slip between the driving motor (driven from the crank) and the driven > motor (connected to the timing chain / cam gear itself) can be varied, > giving a variable cam timing technique? it could be extended into having > different timing for intake & exhaust, for dohc engines. greetings: what i have been thinking about sounds fairly doable. add a link (to lengthen) the timing chain. add a pair of idlers (pushing in, one spring on both of them) and mount both of them on a movable base. this base can be moved by oil pressure, centripital (spin, like timing) or electrical. to get the variable timing, just slide the base to one side (making the (cam) (cam) one idler take up all the slack) to get | / \ | more timing, and run it the other to o| |o -idlers- o| |o retard it. normal operation might be ()/ \() in the middle... (start retarded, for crank crank more torque down low, thru the middle to advanced for high rpm horsepower). note that the ascii drawing shows the crank cog moving side to side. i claim artistic license. :) so, what do y'all think? how much advance/retard could one get, and how would such a timing adjustment change what kind of grind one selected? later, kc - -- "ooooh, crumbs!"if the world is nite, shine my life like a lite"live your life with PASSION"hey waiter, there's a transvestite in my soup"hey mister, are you tall?"all alone in the nite"son of a son of a sailor"John DeArmond fanclub #13 "he's dead, jim"he's not dead, he's electroencephalographically challenged" kc ------------------------------ From: cal@xxx.com (Cal Smith) Date: Thu, 9 May 1996 09:47:05 -0600 Subject: Re: search Steve Asked about Simple Digital Systems: (403)274-0154 in Calgary, Alberta, Canada. For me, their's is the system to beat. Some would complain that their system is too limited at 8 bits. Compared to the complexity of correctly setting up the tables of a 16 or 32 bit system, the SDS unit is _much_ easier to tune and the difference in output between SDS and say, Electromotive or TEC or MOTEC is minimal; probably insignificant for our purposes (no one here is doing a DIY-F1, right? :-) ). Granted, I'm not an expert at _making_ my own EFI ECU. I have researched the commercial units pretty heavily, though. For <$1000 U.S. you get everything, including software and a great in-cockpit controller that lets you make adjustments in real-time and gives you major function readouts in real-time: manifold pressure, coolant temperature, ignition advance (including knock-retard), rpm (every 250rpm), and intake air temperature. The system can run open- or closed-loop. Yes, you can do staged injectors, which seems to be the thread du-jour. :-) You can't do the really trick stuff like traction control, etc. but you do get one hell of an EFI/Ignition system. Rather than explain it any further, why don't you call the number and order the info packet. They should include the installation/setup manual and it makes for a pretty good EFI read in and of itself. Not to knock the list but SDS has given me a lot more information and answered a lot more questions for me than did the members of this list when I asked for help (save it be for the advice offered by James Weiler--Thanks James!). Cal Smith ------------------------------ From: ehernan3@xxx.com (Edward Hernandez (R)) Date: Thu, 9 May 1996 11:46:11 +0500 Subject: Re: Variable cam timing "note that the ascii drawing shows the crank cog moving side to side. i claim artistic license. :) so, what do y'all think? how much advance/retard could one get, and how would such a timing adjustment change what kind of grind one selected?" You can calculate how much this can phase the cam. It's pretty simple, it's been tried, and it's how I would do it on a SB Chevy or other cam in block motor, DIY style. The drawback is increased wear on the chain guides, but you could concievably do it with a cable to the driver's compartment! Okay, so you'd really want both hands on the wheel, but that was just to illustrate how simple and doable this is. ------------------------------ From: WERNER_HAUSSMANN@xxx.com Date: Thu, 9 May 96 09:51:05 -0600 Subject: Re: someone willing to sell cs453 inj drivers Hello David Tell us more about your EFI system.It sounds like a good deal.You wet my appetite but let's hear more before I get too enthusiastic. 1. What method is used to measure air mass? Does it measure air mass directly or does it calculate air mass from air pressure? Does it take into account altitude (we're at 5000-11000 ft)? Can it use more than one type of sensor for measuring air mass? 2. Can it run both at Lambda = 1 for cruising, and with enrichment for acceleration, say Lambda = .88? 3. Any provision for "accelerator pump"? Start enrichment? cold running? 4. Is it fuel injection only or is there some output to drive ignition? ( that would be real nice) 5. Will it control my Fuel pump? 6. Any provision for trottle during warm up? What type? will it run the Ford trottle adjust motor? (thats +/_dc signal) 7. Is it analog or digital? or a hybrid? What adjustments are possible? how? and does it interpolate or does it use the map values directly? 8. How does it sense crank position? or does it not need position, only rpm? Anyway you get the idea Do Tell us more Pleeeeeaase..., I don't want to build my own if you will do it for me for a small charge. Werner ______________________________ Reply Separator _________________________________ Subject: someone willing to sell cs453 inj drivers Author: Non-HP-owner-diy-efi (owner-diy_efi@xxx.edu) at HP-ColSprings,mimegw2 Date: 5/8/96 11:48 PM I have recently went into production of my first EFI system and I am using the Cherry 4/1 amp part (cs453). I would be willing to sell them to the net EFI people in small quantities since I have plenty at the moment. Let me know if there is any interest. If you want some send $6.25 each plus 6.25% tax if in Illinois and $2.00 shipping to: SGD Electronics 2902 N Sangamon Rd Decatur Il 62521 Hope this is of some help. By the way, my efi system is only $345 if any one is interested in that. I don't want this to sound like an advertisement, but am just trying to fill a need that has been requested. David J. Doddek |pantera@xxx.com Owner SGD Electronics & Development Engr for Caterpillar |h 309 685-7965 Formula SAE Team Sidewinder 94-95 |w 309 578-2931 89 T-bird SC, 69 Fairlane w/SGD EFI |fx 217 428-4686 74 Pantera w/Electromitive Tec-II Twin turbos and Nitros | Hey, If you are going to go fast, go REEEAAL FAST. | ------------------------------ From: thoffman@xxx.net (Tim Hoffman) Date: Thu, 09 May 1996 16:57:47 GMT Subject: Re: MAF sensors On Thu, 9 May 1996 12:18:51 +-100, you wrote: >Anybody interested in playing with an ultrasonic, dop shift MAF sensor? > >I saw an idea somewhere, probably a year ago, for a MAF sensor that used = >an ultrasonic transducer pair to measure airspeed by dop shift... seemed = >like a very simple idea, and had the advantage of being relatively = >unintrusive to the air flow. Or for that matter does anybody out there = >know of a production version, or any more reserch into the idea??? > >Mark > I believe all of Mitsubishi's air flow sensors are of this design. I know for sure that the Eclipse/Talon/Laser and the 3000GT/Stealth use them. In the back of "How to Tune and Modify Bosch Fuel Injection" there's a short section on Mits. injection that describes the ultrasonic sensors there. Look for more info on Karmon Vortex sensors... Tim Please Note New Address: thoffman@xxx.net ------------------------------ From: TA Ratke Date: Thu, 9 May 1996 11:07:05 -0600 (MDT) Subject: Re: Doppler shifted MAF sensors Mark, I saw that very topic discussed in some detail in an old SAE Automotive Engineering mag, I believe it was Oct. '81. Interesting idea, but I see it as being a real bear to calibrate. The MAF appeared to be VERY sensitive to turbulence etc within the airstream. In addition, the acoustic properties of the intake system must be accounted for and filtered. If I remember correctly, it worked quite well once all the bugs were worked out of the system. I say good luck! I'll be _VERY_ impressed if a DIY'er can get a system like that working. Todd- On Thu, 9 May 1996, Mark Pitts wrote: > Anybody interested in playing with an ultrasonic, dop shift MAF sensor? > I saw an idea somewhere, probably a year ago, for a MAF sensor that used > an ultrasonic transducer pair to measure airspeed by dop shift... seemed > like a very simple idea, and had the advantage of being relatively > unintrusive to the air flow. Or for that matter does anybody out there > know of a production version, or any more reserch into the idea??? > > Mark ||||||||||||||||||||||||||||||||||||||||||||| Todd Ratke e-mail: tratke@xxx.ca -or- ratke@xxx.ca ||||||||||||||||||||||||||||||||||||||||||||| ------------------------------ From: Bohdan.L.Bodnar@xxx.com Date: Thu, 9 May 1996 12:23:14 -0500 Subject: Re: Doppler shifted MAF sensors This idea appears to be somewhat similar to the Karman Vortex MAF sensor used on some Mitsubishi vehicles (e.g., Dodge Colt)... Time for more coffee... Bohdan Bodnar Lucent Technologies, Inc. ------------------------------ From: cmorris@xxx.com (Charles) Date: Thu, 9 May 1996 15:52:20 -0700 Subject: Re: 8748 add'l injector controller design You wrote: > >> I have completed the breadboard of my 8748 controller which drives two 60 lb/hr injectors (using MC3484 peak/hold drivers). >Sounds very lilke a very ambitous project! Is the 745i an L-jetronic or Motronic? What did you actually replace on the stock unit? Ihave a 528i L-jet which I would like to improve. >---------------------------------------------- >Michael Glidewell Boston, Massachusetts glide@xxx.net This was designed as a _supplementary injector_ controller only - the stock setup is still all there and unaltered. It's an L-Jetronic. (Actually, since I increased the engine size from 3.2 to 3.5 liters, I raised the fuel pressure from 2.5 to 3.0 bar and tightened the airflow meter spring about five notches. The meter can "read" the bigger engine without raising the fuel pressure, but it will run out of range sooner if the spring is not tightened). A more ambitious project would be to replace the stock ECU with my own design... but I'm not feeling up to it! The original design works well enough that I don't want to re-invent the wheel... What don't you like about your 528i system? Feel free to e-mail off the list (or on). - -Charles ------------------------------ From: cmorris@xxx.com (Charles) Date: Thu, 9 May 1996 15:55:43 -0700 Subject: Re: Injector controller (attached) You wrote: > > Yes the files are available, but unfortunatly they appear to be in a format that many cannot read/decode. If some one would be so kind as to make this available as a normal .zip file it would be appreciated. > >BTW: Pkunzip, Winzip, Extract, & Esscode all do not recognise the format that this file is in. Strange! I encoded it with PKZIP, and used Netcom's NetCruiser mail utility to send it as an attachment. I'm not extremely knowledgeable about PC's (although I have designed embedded 80C86 systems ;) so if someone can sort this out, please do so! - -Charles ------------------------------ From: RABBITT_Andrew@xxx.au Date: Fri, 10 May 1996 07:50:40 est Subject: Reply to: Re: Doppler shifted MAF sensors Kaman Vortex and Doppler Shift MAF sensors are not actually MAF sensors in that they only measure airflow volume, therefore you will need temperature and pressure (altitude) compensation which gets messy. The Thermal Mass Flow meters (Hotwires) are true mass flow devices requiring no compensation. But anyway, if you want to go this route, take a look at Laser Doppler Velocimetry. Andrew Rabbitt oecar1@xxx.au From: Bohdan.L.Bodnar@xxx.com:smtp Date: ## 05/09/96 12:23 ## This idea appears to be somewhat similar to the Karman Vortex MAF sensor used on some Mitsubishi vehicles (e.g., Dodge Colt)... Time for more coffee... Bohdan Bodnar Lucent Technologies, Inc. ------------------------------ From: David Piper Date: Fri, 10 May 1996 00:34:19 -0400 Subject: Re: Injector controller (attached) At 03:55 PM 5/9/96 -0700, you wrote: >You wrote: > >> >> Yes the files are available, but unfortunatly they appear to >be in a format that many cannot read/decode. If some one would be so >kind as to make this available as a normal .zip file it would be >appreciated. >> >>BTW: Pkunzip, Winzip, Extract, & Esscode all do not recognise the >format that this file is in. > > Strange! I encoded it with PKZIP, and used Netcom's NetCruiser mail >utility to send it as an attachment. I'm not extremely knowledgeable >about PC's (although I have designed embedded 80C86 systems ;) >so if someone can sort this out, please do so! >-Charles > WinVn can decode MIME files. TurboDave Less Maintenance, More Performance. ------------------------------ End of DIY_EFI Digest V1 #135 ***************************** 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".