From owner-diy_efi Mon Nov 21 23:35:21 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA20742; Mon, 21 Nov 94 23:35:21 GMT Received: from shiva.trl.OZ.AU by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA20737; Mon, 21 Nov 94 18:34:44 -0500 Received: by shiva.trl.OZ.AU id AA12888 (5.65c/IDA-1.4.4 for DIY_EFI@coulomb.eng.ohio-state.edu); Tue, 22 Nov 1994 10:34:15 +1100 From: Craig Pugsley Message-Id: <199411212334.AA12888@shiva.trl.OZ.AU> Subject: Re: FW: How much memory? To: DIY_EFI Date: Tue, 22 Nov 1994 10:34:14 +1100 (EST) In-Reply-To: <199411211949.AA15692@tomcat.al.noaa.gov> from "Ciciora Steve" at Nov 21, 94 12:41:31 pm X-Mailer: ELM [version 2.4 PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2304 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI > My solution to your delima is to use a chip that can be configured in single > chip mode or expanded mode (8051 family, 6811, etc). With something as > complicated as this project, don't fool yourself that you will only build one. > If you try to limit yourself to hardware that will be the development system > and the final product, it could be so unpleasent to work with that it will > loose it's fun. I have a development system that is 10 times as big as > necessary, 10 times as much memoy, 10 times as much power supply, etc. I have > a 6811 in expanded mode with a 32k ram chip (I debug most of my code out of > ram, when it works, transfer it to eprom) and 32k eprom. That's what I was planning to do (use a 'boot' program to load code from the serial port into ram), The main problem I have is to figure out what the final size is likely to be so I can make an initial processor choice. While chips in this family (8051) share code that is very similar, they vary dramatically with IC pinouts (IE Can't just drop in the latest & greatest device without reconfiguring the pinout) > computer). No passenger seat, the dash is ripped out and a big breakout box > (of all the computer wires) is put in it's place. With thermocouple wires and > sensor wires all over the place, and a computer where the passenger seat was, > it gets quite a few stairs when I bring it in to get the emmissions checked! ;-) Also, on my QFP socket question I received two replies: > Also, is there such a thing as a socket for a 'quad flat pack' SMD IC. > I can find them for PLCC packages but not QFP, and QFPs are pretty hard > to wire wrap :-/ Any ideas?. SAMTEC makes QFP to PGC Adapter boards for a variety of QFP Packages. Chip Specific Adapters "SPEC" for a variety of Motorola and Intel Chips, As well as Chip Generic Adaptor "GEN" for various lead count JEDEC and EIAJ QFPs. Simply solder down, and Plug in to a PGA socket! SAMTEC phn 800 726-8329. Depends on what size you need. 3M makes PQFP sockets in 84, 100, 132, 164, and 196 pin counts. I believe AMP makess a line too, but I don't have their catalog handy. The 3M ones use a staggered pin layout, so you really couldn't use them on standard perf board. Craig pugsley@trl.oz.au From owner-diy_efi Mon Nov 28 20:24:51 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA00211; Mon, 28 Nov 94 20:24:51 GMT Received: from localhost.eng.ohio-state.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA00206; Mon, 28 Nov 94 15:24:48 -0500 Message-Id: <9411282024.AA00206@coulomb.eng.ohio-state.edu> To: DIY_EFI Cc: jsg Subject: F/A feedback oscillations Date: Mon, 28 Nov 94 15:24:48 -0500 From: John S Gwynne Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI -------- A friend happen to find a section in a controls book (Feedback Control of Dynamic Systems, 3rd ed. by Franklin, Powell, and Emami-Naeini) that touched upon automotive engine control. It's only eight pages and written for someone not familiar with engines, but what I read seemed reasonable and consistent with other thing I've seen. It's not informative enough that I would recommend getting the book. The author assumes a fairly simplistic model for the control system which I'll try to show below. ----> C2 --->| | | desired F/A --> + --(error)--> C1 --->| + ---> C4--->exhaust F/A | | | ^ (-) ----> C3 --->| | | | | | ----- [sensor] <-------- C5 <--------------- where C1 = Control law C2 = Inlet manifold (fast fuel) = 1/(t_1 s + 1) C3 = Inlet manifold (slow fuel) = 1/(t_2 s + 1) C4 = time delay (engine rotation...) = e^(-s T) C5 = Sensor lag (exhaust gases mixing in the manifold) = 1/(t s + 1) sensor = sensor characteristic; Vout as a function of F/A I imaging this is as simple of a model that one could use to show the desired results and and is of little other value to us. A proportional plus integral (PI) controller (control law = K_p + K_1/s) was rationalized and bode and root locus plots are shown for for the special case of t_1 = 0.02 sec., t_2 = 1 sec., T = 0.2 sec., t=0.1sec., and a linearized sensor model around the stoichiometric point. The analysis continues with a look at feed back gain and and system stability. What is then shown that is of interest is the problems induced by the highly nonlinear sensor response. The bottom line was that because of saturation, the effective sensor gain is much less (when away from stoichiometric) and the system settling time was much longer then desired; on the order of 13 seconds. By increasing the gain the system becomes unstable and the F/A ratio begins to oscillate with increasing "swings" until the saturation reduces the effective sensor gain and the system reaches a limit cycle corresponding to the point where the root locus crosses the imaginary axis. With this approach, the settling time is more like 2 sec. but the output F/A ratio will oscillate with a "swing" dependent on the gain parameters. So what's the bottom line? Well, it would seem that *in part* (1) the oscillation in the feedback is due to the desire to greatly improve the settling time of the system. (2) By controlling the PI gain factors the amplitude of the oscillations are controlled (clearly a function of sensor characteristics and engine speed/load). So, Dale, if we were to add a differentiation term to help the oscillation, would this seem consistent with what you have found in other controllers? (also excluding the terms to induce more/faster oscillations) John S Gwynne Gwynne.1@osu.edu _______________________________________________________________________________ T h e O h i o - S t a t e U n i v e r s i t y ElectroScience Laboratory, 1320 Kinnear Road, Columbus, Ohio 43212, USA Telephone: (614) 292-7981 * Fax: (614) 292-7292 ------------------------------------------------------------------------------- From owner-diy_efi Mon Nov 28 23:22:30 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA04666; Mon, 28 Nov 94 23:22:30 GMT Received: from curly.cc.swin.edu.au by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA04661; Mon, 28 Nov 94 18:22:26 -0500 Received: from romulus.mm.swin.edu.au by curly.cc.swin.edu.au (5.65c/1.34) id AA21509; Tue, 29 Nov 1994 10:21:50 +1100 Received: From MECHMAN/WORKQUEUE by romulus.mm.swin.edu.au via Charon-4.0-VROOM with IPX id 100.941129102133.352; 29 Nov 94 10:21:50 -1100 Message-Id: From: "Andrew Dennison" Organization: Swinburne University To: DIY_EFI Date: Tue, 29 Nov 1994 10:21:25 EST-11 Subject: Re: F/A feedback oscillations Priority: normal X-Mailer: Pegasus Mail v3.1 (R1a) Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI From what I've read about feedback control of engines, I've decided that the best control strategy is NOT to try and do real time F/B control in the traditional way. It may work in steady state (constant load) where you can assume that the system is linear but any transients (accel, decel) will be hard to control and probably require seperate control. I believe that the best approach would be to have a traditional 2D load-speed map (like open loop control) and use something like an adaptive least mean squares algorithm to tweak the points in the map. The adaption algorithm would use a simple dynamic model of the engine (such as the one you included) as a reference model for the adaption process. This adaption process can tollerate quite noisy sensors, with the noise only slowing the convergence to the optimal control strategy. Andrew ------------------------------------ Andrew Dennison - Research Associate The CIM Centre Melbourne, AUSTRALIA Phone: +61 3 214 8296 Fax: +61 3 214 4949 WWW: http://cim.mm.swin.edu.au/welcome.html From owner-diy_efi Tue Nov 29 00:15:37 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA04750; Tue, 29 Nov 94 00:15:37 GMT Received: from mail.swip.net by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA04745; Mon, 28 Nov 94 19:15:32 -0500 From: Lotus_Mail_Exchange_at_Lotus-Stockholm@lotussun.lotus.se Received: by mail.swip.net (8.6.8/3.01) id BAA24499; Tue, 29 Nov 1994 01:16:38 +0100 Received: from cc:Mail by lotussun.lotus.se id AA786099802 Tue, 29 Nov 94 01:03:22 Date: Tue, 29 Nov 94 01:03:22 Message-Id: <9410297860.AA786099802@lotussun.lotus.se> To: DIY_EFI Subject: nternet Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Delivery Failure Report Your document: Re: F/A feedback oscillations could not be delivered to: P?r Will?n@LotusInt because: Error transferring to LOTUS-NORDIC-MAILEX mail.box; Maximum hop count exceeded. Message probably in a routing loop. Routing path: Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,LCC LOTUS NORDIC You can resend the undeliverable document to the recipients listed above by choosing the Send command on the Mail menu. Unless you receive other Delivery Failure Reports, the document was successfully delivered to the other recipients, if any. (strikethrough: ________________________) To: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-INTERNET@ccMailSweden cc: From: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-Internet@ccMailSweden Date: 94-11-29 00.48.00 Subject: Re: F/A feedback oscillations From what I've read about feedback control of engines, I've decided that the best control strategy is NOT to try and do real time F/B control in the traditional way. It may work in steady state (constant load) where you can assume that the system is linear but any transients (accel, decel) will be hard to control and probably require seperate control. I believe that the best approach would be to have a traditional 2D load-speed map (like open loop control) and use something like an adaptive least mean squares algorithm to tweak the points in the map. The adaption algorithm would use a simple dynamic model of the engine (such as the one you included) as a reference model for the adaption process. This adaption process can tollerate quite noisy sensors, with the noise only slowing the convergence to the optimal control strategy. Andrew ------------------------------------ Andrew Dennison - Research Associate The CIM Centre Melbourne, AUSTRALIA Phone: +61 3 214 8296 Fax: +61 3 214 4949 WWW: http://cim.mm.swin.edu.au/welcome.html From owner-diy_efi Tue Nov 29 01:15:26 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA04826; Tue, 29 Nov 94 01:15:26 GMT Received: from mail.swip.net by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA04821; Mon, 28 Nov 94 20:15:20 -0500 From: Lotus_Mail_Exchange_at_Lotus-Stockholm@lotussun.lotus.se Received: by mail.swip.net (8.6.8/3.01) id CAA17121; Tue, 29 Nov 1994 02:16:34 +0100 Received: from cc:Mail by lotussun.lotus.se id AA786103403 Tue, 29 Nov 94 02:03:23 Date: Tue, 29 Nov 94 02:03:23 Message-Id: <9410297861.AA786103403@lotussun.lotus.se> To: DIY_EFI Subject: nternet Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Delivery Failure Report Your document: nternet could not be delivered to: P?r Will?n@LotusInt because: Error transferring to LOTUS-NORDIC-MAILEX mail.box; Maximum hop count exceeded. Message probably in a routing loop. Routing path: Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,LCC LOTUS NORDIC You can resend the undeliverable document to the recipients listed above by choosing the Send command on the Mail menu. Unless you receive other Delivery Failure Reports, the document was successfully delivered to the other recipients, if any. (strikethrough: ________________________) To: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-INTERNET@ccMailSweden cc: From: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-Internet@ccMailSweden Date: 94-11-29 01.48.00 Subject: nternet Delivery Failure Report Your document: Re: F/A feedback oscillations could not be delivered to: P?r Will?n@LotusInt because: Error transferring to LOTUS-NORDIC-MAILEX mail.box; Maximum hop count exceeded. Message probably in a routing loop. Routing path: Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,Lotus-Nordic-MailEx,LCC LOTUS NORDIC,LCC LOTUS NORDIC You can resend the undeliverable document to the recipients listed above by choosing the Send command on the Mail menu. Unless you receive other Delivery Failure Reports, the document was successfully delivered to the other recipients, if any. (strikethrough: ________________________) To: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-INTERNET@ccMailSweden cc: From: DIY_EFI@coulomb.eng.ohio-state.edu at UUCP-Internet@ccMailSweden Date: 94-11-29 00.48.00 Subject: Re: F/A feedback oscillations From what I've read about feedback control of engines, I've decided that the best control strategy is NOT to try and do real time F/B control in the traditional way. It may work in steady state (constant load) where you can assume that the system is linear but any transients (accel, decel) will be hard to control and probably require seperate control. I believe that the best approach would be to have a traditional 2D load-speed map (like open loop control) and use something like an adaptive least mean squares algorithm to tweak the points in the map. The adaption algorithm would use a simple dynamic model of the engine (such as the one you included) as a reference model for the adaption process. This adaption process can tollerate quite noisy sensors, with the noise only slowing the convergence to the optimal control strategy. Andrew ------------------------------------ Andrew Dennison - Research Associate The CIM Centre Melbourne, AUSTRALIA Phone: +61 3 214 8296 Fax: +61 3 214 4949 WWW: http://cim.mm.swin.edu.au/welcome.html From owner-diy_efi Tue Nov 29 02:25:21 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA05260; Tue, 29 Nov 94 02:25:21 GMT Received: from eigen.ee.ualberta.ca by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA05255; Mon, 28 Nov 94 21:25:18 -0500 Message-Id: <9411290225.AA05255@coulomb.eng.ohio-state.edu> Received: by eigen.ee.ualberta.ca (1.37.109.4/15.6) id AA29182; Mon, 28 Nov 94 19:25:15 -0700 From: Dale Ulan Subject: Re: F/A feedback oscillations To: DIY_EFI Date: Mon, 28 Nov 94 19:25:14 MST In-Reply-To: <9411282024.AA00206@coulomb.eng.ohio-state.edu>; from "John S Gwynne" at Nov 28, 94 3:24 pm Mailer: Elm [revision: 70.85] Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI ... > corresponding to the point where the root locus crosses the imaginary > axis. With this approach, the settling time is more like 2 sec. but > the output F/A ratio will oscillate with a "swing" dependent on the > gain parameters. > > So what's the bottom line? Well, it would seem that *in part* (1) the > oscillation in the feedback is due to the desire to greatly improve > the settling time of the system. (2) By controlling the PI gain factors > the amplitude of the oscillations are controlled (clearly a function of > sensor characteristics and engine speed/load). A standard Zirconia cell also mis-behaves if you try to run the sensor right at stoich. I'm not sure of the mechanism, but the cell's output begins to drift around a lot, especially at idle, so although you control the sensor's real output to say, .5 volts, the actual A/F will vary. The sensor appears to be more accurate during the limit cycle oscillation, rather than a steady state. This may be due to the sensor's temperature sensitivity, or its sensitivity to carbon monoxide, which tends to bias the sensor on the rich side. > So, Dale, if we were to add a differentiation term to help the > oscillation, would this seem consistent with what you have found in > other controllers? (also excluding the terms to induce more/faster > oscillations) I've only seen it in the GM controller, but it does appear as though increasing the D term should help stabilize the oscillation, and make it less dependant on the sensor itself, and rather, with the controller and the system. It is worth noting that the oscillation size is fed back, along with the oscillation frequency, to result in new and better P, I, and D coefficients... possibly to servo the frequency to where they wanted it. -Dale From owner-diy_efi Tue Nov 29 02:38:03 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA05355; Tue, 29 Nov 94 02:38:03 GMT Received: from eigen.ee.ualberta.ca by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA05350; Mon, 28 Nov 94 21:38:00 -0500 Message-Id: <9411290238.AA05350@coulomb.eng.ohio-state.edu> Received: by eigen.ee.ualberta.ca (1.37.109.4/15.6) id AA29233; Mon, 28 Nov 94 19:37:57 -0700 From: Dale Ulan Subject: Re: F/A feedback oscillations To: DIY_EFI Date: Mon, 28 Nov 94 19:37:56 MST In-Reply-To: ; from "Andrew Dennison" at Nov 29, 94 10:21 am Mailer: Elm [revision: 70.85] Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI ... > (constant load) where you can assume that the system is linear but > any transients (accel, decel) will be hard to control and probably > require seperate control. Usually, the feedback system is 'zeroed' during transients, because it would cause inaccurate learning. > > I believe that the best approach would be to have a traditional 2D > load-speed map (like open loop control) and use something like an > adaptive least mean squares algorithm to tweak the points in the map. > The adaption algorithm would use a simple dynamic model of the engine > (such as the one you included) as a reference model for the adaption > process. In fact, normally, the speed-load map (or airflow curving map) is left alone in ROM, and a small array (from 16 to 64 values) are modified by the feedback process. If you don't mind reading a paper a few times, SAE paper 860594, Development of a High-Speed High-Precision Learning Control System for Engine Control, by Naoki Tomisawa of JECS, and Hiromasa Kubo and Shoji Furuhashi of Nissan is worth reading. > > This adaption process can tollerate quite noisy sensors, with > the noise only slowing the convergence to the optimal control > strategy. Actually, the trick is to filter out the engine pulsations without degrading the performance of the steady-state control system. If you can get your steady-state control algorithms to work properly during all driving conditions, then you can add feedback to make the system better. Because of the differences in time lags, feedback control is not appropriate for engines in the traditional sense. Engines like feed-forward control (which is basically what present systems do), with a small amount of feedback. The trick is to use proper models for doing feed-forward control, rather than spending your life in a dyno lab. Sensor filtering is not trivial, either. Proper filters have no lag-time, which is not the case with most electronic filters. This is not easy to do, so in modern engine control theory, an observer (or forward model) is built to mathematically simulate the noisy or filtered variable. This is where Extended Kalman Filters and fun stuff like that come in... An example of a neat observer is in a paper (1994), Transient Air Flow Rate Estimation in a Natural Gas Engine Using a Nonlinear Observer by Robert Weeks and John Moskwa. Papers 910258, 930856, and 920289 are also good. The first two are by Elbert Hendricks and others, which are what I call essential reading. The third paper covers operating characteristics of Zirconia sensors. -Dale