DIY_EFI Digest Tuesday, 8 October 1996 Volume 01 : Number 304 In this issue: FW: DIY Traction Control? FW: drive-by-wire RE: RE: drive-by-wire FW: drive-by-wire Air Pressure Transducer EFI Vendor List Re: Air Pressure Transducer RE: DIY_EFI Digest V1 #303 Oval Pistons Re: RE: Re: Using PC HW Re: Air Pressure Transducer Re: FW: drive-by-wire RE: Re: Using PC HW Re: Re: FW: DIY Traction Control? RE: Re: Using PC HW RE: DIY_EFI Digest V1 #303 68332 vs others, what makes a good EFI controller. FW: DIY_EFI Digest V1 #303 FW: FW: DIY Traction Control? FW: DIY_EFI Digest V1 #303 See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: Mark Pitts Date: Mon, 7 Oct 1996 11:54:35 +-100 Subject: FW: DIY Traction Control? Fully, completely and utterly seconded! All the comments bellow. 'cos somdays..... you just got to let the tail end hang out! Mark - ---------- From: MaxBoost@xxx.com] Sent: Saturday, October 05, 1996 7:51 AM To: diy_efi@xxx.edu Subject: DIY Traction Control? How about a DIY traction control system? Relatively simple comparison of wheel speeds (most cars have abs these days, so parts are readily avail). Wire into the injector ground circuit back to the ecm and drop cylinders as needed. Use a pic processor and some drivers to interupt the ground circuit. Also need an on off switch on the dash as well as a pot to adjust the percentage of slip allowed. (got to have the switch so you can turn it off and roast the tires :) Max. ------------------------------ From: Mark Pitts Date: Mon, 7 Oct 1996 10:15:08 +-100 Subject: FW: drive-by-wire So all these autos in the films is fairly accurate then... what do you = guys out there in DIY EFI land prefer? (Manual or automatic = transmission) (apart from putting the box under processor control.) Mark (there is a real question going with all of this eventually... just = getting some feelings at the moment) - ---------- From: Daniel Ridge[SMTP:newt@xxx.gov] I understand that only 11% of new cars are sold with manuals. The VW "drivers wanted" series of ads that has been running here lately has featured _only_ manual trans. cars. - -Dan ------------------------------ From: Mark Pitts Date: Mon, 7 Oct 1996 11:52:21 +-100 Subject: RE: RE: drive-by-wire At my tender age of 26 anything over 2L would be more or less = uninsurable for me. I'm luck, my Spitfire counts as a classic, and so = get classic insurance rates (they only think it comes out on sunny = weekends). Seems to be a different world over there, as petrol here is (quick = calculation) is about $3.80 per gallon, we need to get into the 30 to 40 = mpg on cruise, and 20 odd round town. All the same, we still belt round = the lanes at 70 plus, and 90-95 is normal cruise (usually about 3800-400 = rpm). I just think its a different culture and attitude to engines.=20 Remeber F1 a few years back? 1000 horses from tiny engines that blew up, = and even then we didnt give them enough fuel to finish a race! Indy on the other hand has always been geared to bigger lumps, and fuel = stops, from what I can work out. (sorry..I do like Indy, I just cant watch it, 'cept if I'm lucky, so I = dont get to 'follow it') apart from the year Nigel Mansel went over = there... when our press went nuts on it. Mark - ---------- From: Markus Strobl[SMTP:eusmsrt@xxx.se] Having grown up in Europe I don't understand how anyone would want a = auto (or slush box as I call them) instead of a stick.=20 So it seems Europeans have much more brains when it comes to = tranmissions but why do European cars have those small pathetic sub-3 liter engines? = AFAIK Europe never had V8 engines in 'normal' cars. Only European V8 I know of = outside expensive british sportscars or luxary cars would be the opel diplomat = back in the 60s? =20 > Steve Ravet =20 Markus '96 Camaro Z28. 6-speed (of course). ------------------------------ From: Mark Pitts Date: Mon, 7 Oct 1996 10:11:42 +-100 Subject: FW: drive-by-wire The bike guys seemed to manage OK... cant remember what motor hadit, but it seemed to work! Mark - ---------- From: Daniel Ridge[SMTP:newt@xxx.gov] PS: we were talking a while ago about non-circular pistons. Wouldn't non-circular piston rings be difficult? ------------------------------ From: "Chuck Thigpen" Date: Mon, 7 Oct 1996 07:37:05 +0000 Subject: Air Pressure Transducer Does anyone know of a source for a fairly low cost Air Pressure Transducer that measures positive and negative gauge pressure? It doesn't have to measure extreme pressures. I'm looking for a transducer that will measure boost pressures of up to 30 psi and vacuum down to - 14.7 psi. It would also be helpful if the transducer was linear in measurement. Any help on any transducers would be greatly appreciated. Chuck Thigpen 90 300ZX 2+2 92 Honda CBR600F2 (Racer) 89 Yamaha FZR400 (Racer) ------------------------------ From: "Orin Harding" Date: Mon, 7 Oct 96 13:54:13 UT Subject: EFI Vendor List Several weeks ago someone asked for a re-posting of the EFI vendor list. Here it is and if you have any additions or corrections please pass them to me. Thanks After Market EFI Systems 1. Electromotive, Inc., 14004-J Willard Road, Chantilly, VA 22021 (703) 378-2444 - (TEC II) 2. Haltech USA, Texas 3. Digital Fuel Injection (DFI), 37732 Hills Tech Drive, Farmington Hills, MI 48024 Makes system sold by Mr. Gasket as the Accel system 4. Emtech (Australia) 5. NOS/EFI Tech 6. Edelbrock 7. Holley, 11955 E. Nine Mile Rd., Warren, MI 48089-2003 8. Mr. Gasket, 8700 Brookpark Rd., Cleveland, OH 44129 (216) 398-8300 Contact: Mark Hamel x488 (Accel/DFI) 9. Microtech (Australia) 10. Racetech Engineering, Bay G, 1007 55th Ave., NE, Calgary, Alberta, Canada T2E 6W1 (403) 274-0154. (Simple Digital Systems - SDS EM-1) 11. Advanced Engine Management Systems, 22 Melrich Rd., Bayswater, Victoria, Australia, 3153 Phone: +61 3 9761 3161, Fax: +61 3 9761 3162 (2D AUS$895, US$671.00 and 3D AUS$1395.00, US$1046.00) 12. Advanced Automotive Electronics Limited, Van Nuys House, Scotlands Drive, Farnham Common, Slough SL2-3ES, England Phone/Fax: 01753-642019 Contact: Brian Warburton 13. EFI (Haltech), Unit 7, 130 Taren Point Road, Taren Point 2229, Sydney, Australia Phone: (02) 525 2400, Fax: (02) 525 2991 14. Plus 1 Micro, Inc., PO Box 1781, Lawrenceville, GA 30246-1781 (InterACQ) Manufacturer Electromotive Haltech DFI/Accel Model TEC II 8E Cost w/sensors US$1,200 US$1,500 US$1,300 PC Interface yes yes yes Software cost US$200-500 incl incl Program offline yes no Map fuel yes yes yes Inj. drivers 4 8 4 Seq. inj. avail($) no avail($) Idle control IACM no IACM Data Logging avail($) yes yes # fuel maps by RPM 8 16 Prog. points ea. RPM 8 8 Total prog. points 64 256 Warm-up Map yes yes yes Speed Density yes yes yes Mass air flow avail($) no no EGO - Closed loop yes yes Ignition control yes yes Ign. amp (IE: MSD) needed no yes Ign Map points 64 128 Detonation sensor yes yes Manufacturer Advanced Automotive Electronics Model Ign. Mgt. Sys Clubmans Sys. Eng. Mgt. Sys. Cost w/sensors UK£395 UK£495 UK£755 PC Interface yes yes yes Software cost Program offline Map fuel n/a yes yes Inj. drivers n/a 1 (2 +£) 4 (Up to 8 +£) Sequential Injection n/a no yes Idle control n/a IACM Data Logging yes yes yes # fuel maps by RPM n/a 16 16 Prog. points ea. RPM n/a 24 24 Total prog. points n/a 384 384 Warm-up Map n/a Speed Density n/a yes yes Mass air flow n/a EGO - Closed loop n/a no yes Ignition control yes yes yes Ign. amp (IE: MSD) needed no no no Ign Map points 384 384 384 Detonation sensor no no yes Manufacturer Holley SDS Model Projection EM-1 Cost w/sensors US$600-1000 PC Interface no no Software cost n/a n/a Program offline Map fuel no Inj. drivers Sequential Injection Idle control yes Data Logging no # fuel maps by RPM Prog. points ea. RPM Total prog. points Warm-up Map Speed Density yes Mass air flow EGO - Closed loop Avail ($) Ignition control no Ign. amp (IE: MSD) needed n/a Ign Map points n/a Detonation sensor no Manufacturer Advanced Engine Management Model 2D 3D Cost w/sensors PC Interface no no Software cost n/a n/a Program offline Map fuel Inj. drivers Sequential Injection Idle control Data Logging # fuel maps by RPM Prog. points ea. RPM Total prog. points Warm-up Map Speed Density Mass air flow EGO - Closed loop Ignition control Ign. amp (IE: MSD) needed Ign Map points Detonation sensor ------------------------------ From: walter.kaufmann@xxx.ch (Walter Kaufmann) Date: Mon, 7 Oct 1996 17:13:02 +0200 Subject: Re: Air Pressure Transducer >Does anyone know of a source for a fairly low cost Air Pressure >Transducer that measures positive and negative gauge pressure? There is a Motorola MPX 4250 (I am not sure, I don't have it here). It is for apsolute pressure, -1 bar to 1 bar overpressure ( +/- 14.7 psi??) The transducer is linearised and temperature compensated. The price here in Switzerland is 75 CHF, so it would cost 50 USD or maybe cheaper. Walter ------------------------------ From: ph%wn.bbc@xxx.uk (Heneghan,Paul) Date: 07 Oct 96 17:18:12 GMT Subject: RE: DIY_EFI Digest V1 #303 From: "Chaxel Frederic (QUALITE)" >I get some TOP signals from a captor. This information >is used to calculate the engine RPM and to calculate the >Ignition time (by the way of a table). >But of course I get the RPM at T time and used it for >ignition at T+DT time (1 engine round after). > >My solution is actually to use RPM(T-1) and RPM(T) to calculate >RPM(T+1) : > RPM(T+1)=RPM(T)-RPM(T-1)+RMP(T)=2*RMP(T)-RPM(T-1) > >It's a simple linear approximation (and simple to program >in assembly language). > >This give good results (much more better than using RPM(T+1)= >RPM(T)) > >Maybe somebody know a better solution (Kalman Filter >for instance, or ...) ? You could try RPM(t+1) = 3 x RPM(t) - 3 x RPM(t-1) + RPM(t-2) This will work for curves (quadratics only) and straight lines. If you need higher order (cubics etc.), let me know. The extra complexity might not be worth it though. >Sorry for my dirty English ! Looks pretty good to me (you should see my French). Paul paul.heneghan@xxx.uk ------------------------------ From: DJohn77284@xxx.com Date: Mon, 7 Oct 1996 14:42:10 -0400 Subject: Oval Pistons I think for oval pistons you have to read Honda. Back in the late 70's Honda had a 4-stroke NR500 motorcycle racer. It was a V4 that thought it was a V8. That is, by using oval pistons, 8 valves per cylinder, two plugs per cylinder and two conrods per cylinder, the breathing capacity was increased to that of a V8 of similar size. The governing racing body had previously put a restriction of 4 cylinders maximum on the 500cc class, hence the elegant engineering. I think the rev limit was somewhere like 22 - 23,000rpm but despite their efforts, Honda couldn't compete with the two strokes. After that, while Honda diverted it's racing efforts to two strokes, the comercial division brought out a road going version of the racer, called NR750. Only 700 were ever made and the cost was way over the top at #38,000. I remember seeing one at a motorcycle exhibition and it didn't look worth the money - loads of plastic on the outside with talk of loads of titanium on the inside. If you want to see one in the flesh I think the best place is the Suzuka 8 Hours where the track marshals use them as run-abouts. Regards, Dave. ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Mon, 7 Oct 1996 14:42:42 -0500 Subject: Re: >Does anyone know about aftermarket MAP sensors for 5.0 Ford FI Speed >Density motors? I know all the Mass Air arguments, but it seems the SD >motor runs much better than its stock MA counterparts. Possibly the >restriction in the stock MA is too great to overcome a non-restricted SD >motor? The question is, is there any way to tweak an SD motor? Comments >would be appreciated. > >On a similar note, has anyone attempted to redesign the Ford 5.0 FI intake >at home, experimenting/fabricating with different runner configurations, >dual throttle bodies, etc? If so, descriptions of your attempts/successes >would be appreciated as well. > >Tony Iomi I'd like to hear about this, also -- my truck is speed density. I think I'd be just as well off to leave it that way -- if I could make mods (like cam, etc). That's one reason Mike's 'calibrator' seems such a good deal -- if I could get one. Tom Cloud ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Mon, 7 Oct 1996 15:01:03 -0500 Subject: RE: Re: Using PC HW >I'll agree 'a car is no place for pc's' It can be done (and is) but why make >extra trouble. [ snip ] > ..... The 68332 is a magnatude better for EFI, _AND_ it has a nice >rich instruction set, including the TPU and other nice thing. If you want >built in A/D they have a version with that too. But the idea is the same, a >PC mudder board will be about the worst thing that you could base a system on. [ snip ] >>I've just *begun* using a Philips 8xC552 uController, and the thought >>of the time and effort and extra hardware in using a PC to control >>a car makes me want to PUKE!!!!! Not to mention how FECKIN' BIG'N >>UGLY it would be!!!!! Forget it!!! All this wasted bandwidth is makin >>me vomit!!! [ snip ] well, at least we have strongly held opinions. mine is that using a CP/M derivative (i.e. MS/PCDOS) and Intel derivative processor buys one the most bang for the buck. now, it's got some downsides. the Motorola instruction set is more elegant -- so what! there's more software for the PC. but, a PC mudder bored wasn't designed for process control (EFI is a process control problem). it's certainly more flexible (from some points of view) than ANY available 68xxx system available (more hardware via plugin cards, more software, cheaper, etc., etc.) but it's B-I-G ! it's got connectors (the bane of any high rel app). it's got so damm much VLSI ASIC's it's almost impossible to get total control of it. so watta i think? i think the PC as a platform looks great. i'll probably use a 68xxx. actually, if i could get Wesley's 'calibrator', i'd use the eec-iv and go drivin'. Tom Cloud ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Mon, 7 Oct 1996 15:06:45 -0500 Subject: Re: Air Pressure Transducer >Does anyone know of a source for a fairly low cost Air Pressure >Transducer that measures positive and negative gauge pressure? >It doesn't have to measure extreme pressures. I'm looking for a >transducer that will measure boost pressures of up to 30 psi and >vacuum down to - 14.7 psi. It would also be helpful if the >transducer was linear in measurement. Any help on any transducers >would be greatly appreciated. >Chuck Thigpen have you looked at the (recent) recurring thread(s) re MAP ?? there's posts friday and today from meself and Todd Knighton re Motorola pressure xducers available from electronics suppliers for about $25 that will read absolute. why can't you convert that to gauge? a range of 30 psig to -15 psig is 3 atmospheres. if you're worried about the actual correlation between barometric (i.e. changing atmospheric) and manifold pressure, use two -- one to sense atmos and the other for manifold -- or use a differential version. Tom Cloud ------------------------------ From: cloud@xxx.edu (tom cloud) Date: Mon, 7 Oct 1996 15:13:52 -0500 Subject: Re: FW: drive-by-wire >So all these autos in the films is fairly accurate then... what do you guys out there in DIY EFI land prefer? (Manual or automatic transmission) > >(apart from putting the box under processor control.) [ snip ] para mi? i've always driven a standard ... when i could i got an auto for the SO. when i finally got to where i felt like i could buy my first new vehicle i got an F-150 with 302 & E4OD auto. love it. i always drove manual because it was less expensive to buy and to maintain -- still know that to be true. but, the auto is nice. i guess my next will be auto also -- just takes more hp to get the same 'fun' as a stick. Tom Cloud ------------------------------ From: "Woodd, Michael" Date: Tue, 8 Oct 1996 08:33:00 +1200 Subject: RE: Re: Using PC HW I belong to the EFI332 list too, dont worry. But I came along too late to really consider getting the setup through the bulk buy-in - I missed the parts buy and I don't know how easy/expensive it would be to get them any other way. I have downloaded all the schematics etc, and maybe later on if I find the 552 has not enough speed then I'll look at it. My understanding of the 332 is the processor cycle is CLK/2 - is that correct? so at 20 MHz, it would be 0.1uS? Also, what about the TPU timer resolution? My problem with the 552 is that for a 16000rpm motorcycle engine, I don't know if the 552 is fast enough with a CLK/12 processor cycle and timer resolution. I think it will be for a mapped 4cyl wasted spark ignition system, but not for timed sequential EFI. Anyhow, for those who expressed an interest in the AN number, it is Philips AN 418, it is to do with Timer 2 Counter/Timer Capture/ Compare functions etc, but it describes an example of how to use these 552 features to implement coil and injector timing. It doesn't describe any algorithms to calculate these values or anything, but the simplest system should be easy to implement once I get used to programming it. I wont be able to scan it and post it before I leave my current job on Thursday, sorry! Mike Woodd (wooddm@xxx.nz) >I'll agree 'a car is no place for pc's' It can be done (and is) but why make >extra trouble. On the EFI 332 list you could have had the CPU board and [snippety snip] >built in A/D they have a version with that too. But the idea is the same, a >PC mudder board will be about the worst thing that you could base a system on. > >Sandy ------------------------------ From: Sandy Date: Mon, 07 Oct 1996 14:16:08 -0700 Subject: Re: About this ford stuff, does anyone have information on the quad coils on the 4.6L mustangs (other 4.6's too). Sandy ------------------------------ From: dzorde@xxx.au Date: Tue, 08 Oct 96 08:35:02 Subject: Re: FW: DIY Traction Control? I know dropping random cylinders is the way used to limit RPM and regain traction, etc. But I dread everytime I hit the rev limiter in the Chev, it feels like someone is trying to tear the engine out from under the bonnet. And there is a similar feeling in the newer sports cars (ie. relatively powerful cars). Is this normal ?? Dan dzorde@xxx.au ______________________________ Reply Separator _________________________________ Subject: FW: DIY Traction Control? Author: diy_efi@xxx.edu at INTERNET Date: 10/7/96 8:06 PM Fully, completely and utterly seconded! All the comments bellow. 'cos somdays..... you just got to let the tail end hang out! Mark - ---------- From: MaxBoost@xxx.com] Sent: Saturday, October 05, 1996 7:51 AM To: diy_efi@xxx.edu Subject: DIY Traction Control? How about a DIY traction control system? Relatively simple comparison of wheel speeds (most cars have abs these days, so parts are readily avail). Wire into the injector ground circuit back to the ecm and drop cylinders as needed. Use a pic processor and some drivers to interupt the ground circuit. Also need an on off switch on the dash as well as a pot to adjust the percentage of slip allowed. (got to have the switch so you can turn it off and roast the tires :) Max. ------------------------------ From: Dirk Wright Date: Mon, 7 Oct 1996 20:42:18 -0400 (EDT) Subject: RE: Re: Using PC HW On Mon, 7 Oct 1996, tom cloud wrote: but it's B-I-G ! You could always take a swa to the parts of the board you dont' need. Just hack 'em off! :) **************************************************************************** Dirk Wright wright@xxx.gov "I speak for myself and not my employer." 1974 Porsche 914 2.0 "A real hifi glows in the dark and has horns." 1965 Goodman House **************************************************************************** ------------------------------ From: William Boulton Date: Tue, 8 Oct 1996 13:33:38 +1000 (EST) Subject: RE: DIY_EFI Digest V1 #303 On 7 Oct 1996, Heneghan,Paul wrote: > From: "Chaxel Frederic (QUALITE)" > >I get some TOP signals from a captor. This information > >is used to calculate the engine RPM and to calculate the > >Ignition time (by the way of a table). > >But of course I get the RPM at T time and used it for > >ignition at T+DT time (1 engine round after). > > > >My solution is actually to use RPM(T-1) and RPM(T) to calculate > >RPM(T+1) : > > RPM(T+1)=RPM(T)-RPM(T-1)+RMP(T)=2*RMP(T)-RPM(T-1) > > > >It's a simple linear approximation (and simple to program > >in assembly language). > > > >This give good results (much more better than using RPM(T+1)= > >RPM(T)) > > > >Maybe somebody know a better solution (Kalman Filter > >for instance, or ...) ? > > You could try RPM(t+1) = 3 x RPM(t) - 3 x RPM(t-1) + RPM(t-2) > This will work for curves (quadratics only) and straight lines. > If you need higher order (cubics etc.), let me know. > The extra complexity might not be worth it though. > > >Sorry for my dirty English ! > > Looks pretty good to me (you should see my French). > > Paul > paul.heneghan@xxx.uk > > Hi folks, Like to add 2cents to the debate. The best way (IMHO) to achieve any degree of accuracy with timing prediction is to run as many reference points per revolution as possible. I've see the result of 1/rev on a twin and that was a disaster. Engines just do not maintain consistent angular velocity during a revolution. I developed a simulator some time ago to test theory on this and found that good results are obtained with at least 6 reference points per rev. I tried a number of basic equations and settled on a variation of the one Chaxel used. Just multiplied the result by some large fraction just less than 1. Still easy in integer assembler. The problem in predicting timing stems mostly from dA and not dV. Paul, I'd be interested in how you arrived at your equation. RSVP Bill Boulton. ------------------------------ From: "Stuart Woolford" Date: Tue, 8 Oct 1996 17:39:06 +1100 Subject: 68332 vs others, what makes a good EFI controller. > I'll agree 'a car is no place for pc's' It can be done (and is) but why make > extra trouble. On the EFI 332 list you could have had the CPU board and > debugger for a couple of $$, and all the free software you could use. > > _But_ more importantly, get the App note. I would like to get a copy of it > if you can. It will be interesting to see other peoples ideas on doing EFI. > I have a 552' about the biggest problem with them (and all 8051 relatives) > is the nasty structure of the processor (My A/C is on, so feel free to > fire away). The 68332 is a magnatude better for EFI, _AND_ it has a nice > rich instruction set, including the TPU and other nice thing. If you want > built in A/D they have a version with that too. But the idea is the same, a > PC mudder board will be about the worst thing that you could base a system on. No need for AC, but have a look at the new 80296 (a new 80196 + stuff) from intel, it has what they call an EPA (event processor array), which has been around for a while in the '196, it is 16bit, 50MHz, fast, has 40bit multiply/accumulate at 12.5 mips, and costs next to nothing. IMHO the EPA, plus some other internal goodies, way surpases the TPU (which is also useful..). These things have 2k onboard RAM, a 6MB address space, 512 bytes 3-port register, etc, and all the nice programmable chip-select/bus type from the old '186, plus more. Real new, but looking real interesting for those high-grunt applications.... > > > >Flame on full: > >I've just *begun* using a Philips 8xC552 uController, and the thought > >of the time and effort and extra hardware in using a PC to control > >a car makes me want to PUKE!!!!! Not to mention how FECKIN' BIG'N > >UGLY it would be!!!!! Forget it!!! All this wasted bandwidth is makin > >me vomit!!! > {snip} > - ---------------------------------------------------------------------- Stuart Woolford, stuart@xxx.nz >>>>In VI Where Available<<<< - ---------------------------------------------------------------------- ------------------------------ From: Mark Pitts Date: Tue, 8 Oct 1996 09:36:06 +-100 Subject: FW: DIY_EFI Digest V1 #303 Could you copy me on the maths side as well please. Mark - ---------- From: William Boulton[SMTP:boultonw@xxx.au] > >My solution is actually to use RPM(T-1) and RPM(T) to calculate > >RPM(T+1) : > > RPM(T+1)=RPM(T)-RPM(T-1)+RMP(T)=2*RMP(T)-RPM(T-1) > > > >This give good results (much more better than using RPM(T+1)= > >RPM(T)) > > > >Maybe somebody know a better solution (Kalman Filter > >for instance, or ...) ? > > You could try RPM(t+1) = 3 x RPM(t) - 3 x RPM(t-1) + RPM(t-2) Like to add 2cents to the debate. The best way (IMHO) to achieve any degree of accuracy with timing prediction is to run as many reference points per revolution as possible. I've see the result of 1/rev on a twin and that was a disaster. Engines just do not maintain consistent angular velocity during a revolution. I developed a simulator some time ago to test theory on this and found that good results are obtained with at least 6 reference points per rev. I tried a number of basic equations and settled on a variation of the one Chaxel used. Just multiplied the result by some large fraction just less than 1. Still easy in integer assembler. The problem in predicting timing stems mostly from dA and not dV. Paul, I'd be interested in how you arrived at your equation. RSVP Bill Boulton. ------------------------------ From: Mark Pitts Date: Tue, 8 Oct 1996 09:25:36 +-100 Subject: FW: FW: DIY Traction Control? Completely standard... feels like the distributor cap has come lose, and = its firing anything anytime, completely horrible, but aparently its = better if the fuel isn't injected either (cos somtimes it detonates due = to lack of spark) and thats not a nice thing to do to your lump! Mark - ---------- From: dzorde@xxx.au] I know dropping random cylinders is the way used to limit RPM and=20 regain traction, etc. But I dread everytime I hit the rev limiter = in=20 the Chev, it feels like someone is trying to tear the engine out = from=20 under the bonnet. And there is a similar feeling in the newer = sports=20 cars (ie. relatively powerful cars). Is this normal ?? =20 Dan dzorde@xxx.au ------------------------------ From: Mark Pitts Date: Tue, 8 Oct 1996 09:27:49 +-100 Subject: FW: DIY_EFI Digest V1 #303 Thanks for the Math (but for somthing looking THAT disgusting in brackets... go wash your typing fingers now!) Mark - ---------- From: Heneghan,Paul[SMTP:ph%wn.bbc@xxx.uk] >My solution is actually to use RPM(T-1) and RPM(T) to calculate >RPM(T+1) : > RPM(T+1)=RPM(T)-RPM(T-1)+RMP(T)=2*RMP(T)-RPM(T-1) > >It's a simple linear approximation (and simple to program >in assembly language). > You could try RPM(t+1) = 3 x RPM(t) - 3 x RPM(t-1) + RPM(t-2) This will work for curves (quadratics only) and straight lines. If you need higher order (cubics etc.), let me know. The extra complexity might not be worth it though. >Sorry for my dirty English ! Paul paul.heneghan@xxx.uk ------------------------------ End of DIY_EFI Digest V1 #304 ***************************** 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".