A Primer on Computer Engine Control | Ford Explorer Forums - Serious Explorations

  • Register Today It's free!

A Primer on Computer Engine Control

Glacier991

EF Tranny Guru
Moderator Emeritus
Joined
February 8, 2003
Messages
9,824
Reaction score
94
City, State
Sacramento, CA 95827
Year, Model & Trim Level
1992 XLT
Often, when the check engine light comes on, there is a tendency to want to start replacing sensors right and left in hopes it might solve the problem. Too few seem to want to take the time to get the codes read, or better yet look at the other parameters a good code scanner could provide. In hopes of saving some of those folks some money by sparing the replacement of good sensors, I thought I’d put together this little primer on the operation of a computer controlled engine. The principles are generic, and in some cases different manufacturers call certain components by different names, but the items are largely similar in function.

How did we get to where we are today? The simple answer is emissions standards. Remember back in the 60’s with big engines and holley 4 barrels? Well the big engines are slowly coming back to power the behemoth SUV’s, but the carburetors are history. Why? Let’s look at the progression of emissions control devices. The first attempts to reduce emissions was to recirculate exhaust gasses and in some cases to inject air into them to allow more complete combustion. Stricter control over carburetor settings (including the infamous tamper proof cap) arrived, with attendant driveability problems. Ford attempted to improve the carburetor by adding a variable venturi in the early 1980’s, and though an engineering marvel, it’s field experience was less than wonderful. What was needed was better and more precise control of fuel mixture at all phases of engine operation – fuel injection controlled by a computer arrived. The fuel delivery problem was solved by having a computer monitor the engine function and deliver a precisely metered amount of fuel to the cylinder based on that data. It created a near perfect fuel air mix throughout the power curve – something not able to be done using carburetors, so, good bye carburetors.

Although an engine can run acceptably with less than the “perfect” air fuel ratio of 14.7 to 1, that ratio not only produces the fewest emissions, it also happens to the range where the catalytic converter (CAT) can operate at maximum efficiency. The CAT operates poorly, if at all, much out of that range on either side. So to further tweak the mixture control capability of the system Oxygen sensors got added providing even more precise measurement of the process for the computer. First they got added upstream of the CAT for this purpose, later we added them downstream too, to provide some indication of CAT performance, and finally we added them on both banks of cylinders. The continuing trend is to monitor more and more functions in more and more places.

Lets look at the information the system works with and how it gets it. Understanding this can often help narrow or pinpoint failed or out of range components. Mind you some of these sensors can be out of whack and not make much of a problem with driveability, and still set a trouble code on the check engine light, others, well if they are bad the poor engine may barely run.

First off, and rarely a problem, is the crankshaft position sensor. This lets the computer know where the engine is in the cycle, and in some cases provides RPM data as well. This more or less keeps the computer abreast of the mechanical operation of the engine. Next there is the matter of air/fuel. Air coming in is measured by the Mass Airflow Sensor (MAF) there is also intake air temperature (IAT) sensors to measure the temperature of that air as well. Let’s take a minute and think – if the computer is receiving bad data about how much air is being used, what’s the likelihood of the fuel air mix being anywhere close to acceptable? Not high. So when the MAF is bad, you are likely going to experience driveability problems. IAT ? If that’s bad probably you may not notice anything at all in the operation of the vehicle. Similar is the coolant temperature data. On the fuel side, the fuel pressure is measured, and the computer, using data as to engine load and speed and throttle position - from the throttle position sensor (TSP) and using a built in lookup table, provides a specific pulse of open injector time to allow the introduction of fuel. You can imagine what a bad TPS could do. The poor computer will be trying to provide fuel for a non-existent condition of operation. At some point after startup, oxygen sensors come into play. Early ones had no heaters, so it took them a while to become operational. Later ones had heaters added to speed up the process. Once these “wake up” the system switches over to them to “fine tune” the air fuel mixture. It does this by adjusting “short term fuel trim” – changing the injector pulse width. The computer can tell you how much, plus or minus, in percent this is being done. In addition your computer has a set of parameters it has learned and uses to compensate for the overall system, called long term fuel trim. As in short term, this can be plus or minus, and the computer can tell you how much is being applied. Fuel trim numbers are a gem that few DIY’ers ever mine. A significantly high long term fuel trim percentage figure (plus or minus) can give you a clue if you are running chronically rich or lean, and provide valuable clues to system problems before they become really major. Finally, we have kept the EGR on many engines, and added controllers for them (Differential Pressure Feedback EGR sensor (DPFE) and EGR vacuum regulator (EVR)). If these get out of whack, too much exhaust gas can be let into the system at lower rpms, significantly affecting the idle, and the computer’s ability to compensate. Often this will show up on fuel trim numbers in the early phases as the computer tries to fix the problem by adjusting pulse width on the injectors. It is very common for a problem in one part of the EGR system to set other trouble codes as other components try and compensate unsuccessfully for an out of balance system. This is why a code reader alone may not lead right off to a successful diagnosis – it may flag an O2 sensor when it’s the DPFE or EGR or EVR that is the real culprit. A scanner can be a great help in sorting out multiple trouble codes.

There are other systems being monitored, but other than to mention the Idle Adjustment Compensator (IAC) I won’t go into detail on them. The IAC exists to give the computer an ability to add engine speed when a load or other demand has caused it to fall or the computer finds it necessary. Usually this would be something like an A/C compressor load as the AC clutch kicks in, or during cold idle situations.

I’ll end this by saying that for the couple hundred bucks they cost, a code reader/scanner is the most valuable tool you can own. Knowing for example that an oxygen sensor should provide a rapidly varying output between .2 and .7 volts and that it should cross over .45 volts frequently can let you watch and see if you have a bad O2 sensor instead of plunking down $75 or more to replace what may be a perfectly good sensor. Scanners can pay for themselves over time, and make diagnosis much easier. Few people realize that today, the DIY can place in his or her hands, interactive computer capability rivaling that which the dealer has and do it rather inexpensively. In another article, and assuming there is interest, I’ll give you readout numbers that can help if you have a scanner.

Happy Exploring

Chris
 



Join the Elite Explorers for $20 each year or try it out for $5 a month.

Elite Explorer members see no advertisements, no banner ads, no double underlined links,.
Add an avatar, upload photo attachments, and more!
.





Excellent post - I learned a lot! I had no idea about fuel trim.

I don't have a scanner/code reader - and was thinking of purchasing one. It seems like the next step for this thread. I've done a search and found varying opinions on them - what do people use, and do you like/hate the device?
 






I got a cheap code scanner off of eBay that cost me about $15... One of the ones where you have to count the number of times a light flashes to figure out the trouble code. It has saved me a ton though.
 






The light flash scanners are for OBD-I, 1996 and later systems use OBD-II. This new system provides all of the sensor readouts as well as codes - i think. You just need the right tool.
 






Well done Chris.

A very easy to read and follow write up. You get my vote for lots of followups. Like which scanner, normal readings etc. All we need now is the same for the autobox diagnosis, transfer case and abs system and we are set.
 






I use an Auto X Ray scanner for OBD II with enhanced Ford capacity. It works great and has saved my bacon often.

I agree that it is indeed the MVP in my tool chest.
 






A scanner would be nice. My problem is that at least one (BII), if not both of my vehicles will not allow me to moniter engine data with a scanner. According to Autoxray, the BII's will only send trouble codes to the scanner, but not engine data (I can already get trouble codes with $3 voltmeter). Not sure at what point in the progression from early OBD-1 to late OBD-1 that the computer started sending data to scanners. Someone here mentioned that he's used his scanner on a '93 Taurus, but that's the earliest model that I've heard of using a scanner on. I've decided that, instead of buying an expensive scanner, to use a technique called "backprobing" where I use a DVOM to monitor sensors output. Given half a chance, I'd invest in a breakout box to simplify the process, but they are prohibitively expensive. I notice, Glacier, that you also have a '92. Have you had success using a scanner on your '92 Explorer? If I knew for sure that it would work on it, I might reconsider getting a scanner (though I still would have liked to have it work on both).
And speaking of the BII, and since this thread is geared towards a generic understanding of EFI, I'll mention that some vehicles don't use a mass air flow sensor to determine incoming air. (Explorers all do). Air behaves very closely to an ideal gas, and can be described by the equation PV=ZnRT. (P=absolute pressure, V=volume, n=moles/mass of air, R=universal gas constant, T=absolute temperature). A mass air flow system measures n directly. However, one could, without loss of accuracy, measure the pressure, temperature, and volume and still know how much air is coming in. These are called Speed density systems, and are present in Ford's 2.9 L V6 engines of the late '80's and elsewhere in the automotive world. These systems utilize a manifold air pressure (MAP) sensor, and the same IAT. Then, knowing the manifold volume/displacement of the engine, the computer can calculate the amount of air coming in and determine the mixture from that. I've seen a few debates at the ranger station about which is better: speed density or mass air, without a definitive conclusion.
 






Shorty... no I haven't been able to get my 92 to spit out real time data... I use my breakout box for that and I have a couple O2 sensor testers that read out real time data from the sensor. Interesting it works on my 93 Sable - like you said about your friend's 93 Taurus, same same. Be careful backprobing <smile>.

Happy exploring

Chris
 












Wow that was very informative. As I told you earlier I am going to buy an OBD2 scanner .......or maybe a entire engine mapping software package......What do you recommend or use? I am 29 and I have just started working on vehicles newer than the mid 80's the computer controlled systems are getting simpiler with help from great people like you THANKS. Wes
 






Think it's time for a bump on this one...
 






Great Post
Just back from a dyno run, the shop ran a Lambda wideband A/F ratio,Other shops never did this before.
Mine ran a consistant 12.5 at 4.2/5.4 rpm, was sort of explained to me as to what this was but do not really understand it.
Can you elude furthur
Thx
 






Let me start by saying I do not know. Then let me add... Lamda is the engineering symbol used by the SAE to describe the "perfect" air fuel ratio of 14.7 to 1. So it sounds like they were seeing what your average A/F ratio was when averaged out (it SHOULD vary slightly above and below the 14.7 number in a perfect world. 12.5 would be rich...

Now, IF my assumption is corrrect, one might want to know why. Injectors? Fuel delivery issues? O2 sensor issues? Electrical issues (like O2 sensor grounding for example), computer control set that way? Intake air impediment/restriction? I'd probably start with the air filter, and go from there. Others may have better ideas, and may in fact be knowledgeable about that test. My reply is just surmise. DO you ever run your Ex at those rpms (4200/5400?)? Who did it look at other rpms (like 3K and 3500?)?

[Note] I just started using an oscilloscope in auto work, and I'll see if I can capture the more or less sine wave scope trace from an 02 sensor and post it. Might be helpful. (If there was enough interest I'd be happy to do a more comprehesive thread on the various inputs to the computer, with scanner reads and scope traces. [Selfish comment start] These kinds of posts take a lot of time, so don't want to do it for just 4 folks. There seemed to be a lack of interest before so I stopped and devoted my attention to other, possibly more useful threads. It's always nice when you work hard to create a thread you think might help a lot of people and that thread gets nominated as useful......it increases the potential audience so you can help more folks that way. Course every thread writer has his or her areas of interest and they may not be reflective of the general board level of interest in that subject. I find this stuff fascinating. (end of selfish comment).
 






I believe the running rich is do to the way my SCT flasher is programed.
On the printout the A/F ratio between 3600 and 4200 rpm runs about 14, graph is sort of hard to read, between about 4600 and 5000K rpm it is about 12.8, am guessing that is in 3rd gear, one can almost see the shift points in the graph, I think.
At 5K the a/f jumps back up to about 13.5 (going into 4th?) and I make peak H.P at about 5250 196.5 RWHP and 228 RW TQ
Great info on your part, thx
 






AH I didn't realize you'd reprogrammed it. Now it makes sense. I have a son in law who runs ricers on the 1/4 mile. He says they routinely program rich, sometimes down into the 12's.
 






This thread was just made to be here... now it is. Might be helpful to folks just getting accustomed (or tuned into) computers and cars.
 






As usual Chris, it's entertaining and informative to read your posts. I hope you continue to experiment and tinker and share your findings. I know I'm all the smarter for it.

Thanks, Dave.
 






Thanks for the kind words Dave, rest assured I am JUST getting started.
 






Read with Great Interest!!!

First of All: Let me say THANKS!!! Great Post and follow-up comments in thread. I now know a lot more about CCE's (computer controlled engines) than I ever knew. I come from the time when you "tinkered" with carbs and points, with a tachometer to get your engine performance where you wanted it. With the advent of computer controlled systems in the mid to late 70's, they effectively ended my"mechanikin".
Secondly: I have a 95 ford explorer manufactured during the week of Feb 10,1995. 4.0L EFI engine. I cannot find where you would connect an engine diagnostic unit. There are no "stickers" under the hood that say OBD compliant or anything else. Can someone tell me where I could find that information? I am wanting to know so that I can purchase a diagnostic unit that has "burn" capability to tweak out my lovely one-owner 1995 Ford Explorer.
 



Join the Elite Explorers for $20 each year or try it out for $5 a month.

Elite Explorer members see no advertisements, no banner ads, no double underlined links,.
Add an avatar, upload photo attachments, and more!
.





Excellent post Chris! You really know your stuff! The problem I have is that I have an OBD II scan tool that my two last cars, A 1994 Ford Thunderbird with 4.6 liter V8, and 1996 Ford Escort with 1.9 4-cylinder worked with. My '95 Eddie Bauer uses what is called OBD I or EEC V. The current scan tool can't pull the codes because it is a proprietary Ford interface! Interestingly enough, the scan tool's connector plugs right in to the data link connector, it just doesn't read anything! Can I retrieve codes via the sweeping meter principal similar to the hairpin and flashing codes idea in General Motors products previous to OBDII? I am curious to whether or not I can get an adapter that will convert the EEC V connection to the OBD II connection so my current scan tool can read it! I don't have any codes right now although I think I have a sticking IAC because every once in awhile the engine will not maintain idle after being started in hot soak. I know Ford's make this black soot that gets everywhere in the system and the engine has 217-thousand miles on it and I probably need to take that off and clean it. Would be interested to see if you have any ideas to getting my OBD II scan tool to work on the EEC V system.
 






Back
Top