noise."MECHANIC: The Ranger · 3:49am Apr 6th, 2022
Today we’re gonna talk about a Ford Ranger! This one’s from a while back, but the repair process was similar to a truck I’ve been working on recently.
But first. . . .
The vehicle in question is a 1996 Ford Ranger, two wheel drive. Four-liter engine, automatic transmission. For an older Ranger in Michigan, it’s in pretty good shape. There’s no major body rust, and the suspension doesn’t look like it’s ready to fall off. It comes in because it’s got a bad misfire.
OBD-II was a standard for automotive engine communication, the idea being that all automakers would have to use the same connector and the same data format for engine communication so that you didn’t need a different scan tool or different adapters for every car. It also sort of standardized the connector location (it was supposed to be under the dash on the driver’s side). While home computer folks had figured out by then that certain ports should be standardized (such as serial or parallel ports), this was a novel concept to automakers.
It also came with some more stringent emissions testing requirements (that the vehicle had to do) and the standardized code tables we all know and love. Gone were the days where you got vague, unhelpful two-digit codes that varied by manufacturer; now they were standard.
Automakers weren’t always quick to adapt; some saw the advantages of the system when it came to diagnosing their own vehicles, others didn’t. Twenty-five years on, most automakers now fully embrace this system, with the only notable exception I can think of being Mercedes.
Which is a roundabout way of saying that the computer system in this thing is really dumb* and it doesn’t monitor nearly as much as it could or should.
___________________________________________________________
*By ‘dumb,’ I mean it isn’t very comprehensively programmed. Ford’s mindset in the beginning of the OBD-II era was to have techs diagnose the vehicles the way they always had, and the codes the engine set were meant to guide you to a particular test. Many things couldn’t be monitored even though the car’s onboard computer knew what was going on. Ford has since embraced not only more comprehensive on-board tests and scan tool tests than some other automakers (you can do a cranking compression test on a modern Ford with just the scan tool), but also codes with additional symptom codes.
The only code I had in the thing when I first started it was Ford’s unique P1000 code. That tells you that the car has not run all its emissions tests yet, and is not useful for diagnosis. I didn’t need a code to tell me it ran badly; I could hear and feel it for myself.
Go to the misfire monitor, and all the data I can get from that is that all the injectors are okay (electrically). In that era, Ford monitored misfires, but wouldn’t let the tech know what they were.
Now, not all hope is lost on that front; you can go to the generic side of the scan tool and in Mode $06 you can view individual test parameter, but you’ve got to figure out which test number is which test and then do a conversion on the raw number; I instead decided to test-drive it in the hopes it might set a cylinder-specific misfire code, and I could also use the opportunity to monitor other numbers.
Luckily, it did set a specific code, P0303, and I also saw that a lot of the other data was suspect. Forward oxygen sensors readings were wrong, mass airflow was wrong, rear oxygen sensor was flatlines, fuel trims were weird, and a few sensors seemed to have biased voltage. As you might remember, most 5v sensors have a sensing range from .5V to 4.5V (so it can tell the difference between ‘open’ and ‘shorted’ vs. an actual reading; the throttle position sensor was 1.1V at closed throttle . . . it’s typically lower than that (like .7V or .8V).
An inspection under the hood revealed more questions than answers. The engine had a lot of new parts, some of which might have been installed in an attempt to fix this drivability problem. There are wiring repairs, the engine ground has been replaced but now it’s on the A/C compressor rather than the block, various bolts and electrical seals are missing . . .
. . . and to add to the fun, the guy who brought it in just bought it, doesn’t know anything about its history, so there’s no way of knowing why any of the previous repairs were done.
The first, easiest thing to do is put a spark plug in and see what happens. We used to stock certain spark plugs, then stopped as the variety of plugs really exploded. We couldn’t send back the old ones, so we keep hoping that something will show up that takes a set we’ve got in stock, and as it happened we had plugs for this Ranger.
A spark plug didn’t fix it. I verified that it had spark down the wire, and also did a compression test on that cylinder since I was there. Some Ford engines of that era had valve problems.
On the one hand, I’m not closer to knowing what’s wrong with this thing, but I do know what’s not wrong with it, and that’s progress. During this testing procedure, it got three more spark plugs installed, I honestly can’t remember why. I think it picked up a misfire on another cylinder which is why it got one of them, and because it is a waste-spark system, I replaced the plug that shared an ignition coil just to be certain that wasn’t the problem.
One thing I did notice was that neither of the upstream oxygen sensors were reading correctly, and the fuel pulsewidth (how much fuel it injects) seemed high for its idle speed. Sometimes knowing what a sensor should be reading is just based on experience—on this vehicle, the oxygen sensors should read between 100mV and 900mV (roughly) depending on if the engine is running rich or lean; given that I can smell gas in the tailpipe and see black soot spitting out of the tailpipe, I know it’s running rich, yet neither oxygen sensor thinks so. In fact, one of them is stuck at 2mV, so that one’s definitely bad.
While I was at Ciderfest, one of the other techs replaced the oxygen sensors which didn’t fix the truck, and it got put back outside.
Sometime after Thanksgiving, it was time to look at the truck again. I took it for another test drive, and I could see that now both upstream oxygen sensors were reading correctly. They both indicated that the vehicle was running very rich, and they were nearly maxed out. The downstream oxygen sensor took forever to start reading, but when it did, it also showed that the engine was running very rich.
In fact, it set two codes for the upstream oxygen sensors stuck rich (IIRC, that was P1132 and P1152, but I’m working off memory here). Strangely, the vehicle never attempted to change its fuel trim.
On a vehicle like this, the computer really only has two goals. The first goal is to make the vehicle always operate with acceptable emissions, and the second is to protect the catalytic convertor. Fords of this era were notorious for relying more on what the downstream oxygen sensor said than the upstream ones, and would often adjust their fuel trims accordingly. Was it possible that the delayed response from the downstream (which made it indicate lean) cause the computer to over-fuel even though the upstream oxygen sensors were both saying that they were getting too much fuel?
The reason I didn’t like that hypothesis is that on a test drive, the downstream sensor finally started to indicate it was rich, too, and the truck still never attempted to correct its fuel.
One of the things that got replaced on this truck were the fuel injectors. Were they the wrong ones? Were they flowing too much fuel? But why did the truck not want to correct, or at least attempt to correct the issue?
Identifix had a suggestion for diagnosing these codes. Basically, it was to make sure that the barometric pressure, air and coolant temperature readings, and mass airflow readings were all correct. If they were, the next step was to monitor the injector pulsewidth with an oscilloscope and see if it was what the computer said it was--if it wasn’t, it needed a computer. Also verify that the correct injectors are in it, and verify fuel pressure.
I told my manager to order a new Mass Airflow Sensor from NAPA so we could send it back when it didn’t fix the truck. The airflow reading was wrong (in my opinion, there isn’t a hard spec for this), but didn’t seem so wrong that it would be a huge problem.
The next day we got one, I installed it and took the truck for a test drive and it was fixed! to nobody’s surprise it wasn’t fixed.
But, now it was trying to set fuel trim. Sort of.
Now, instead of being stuck at 3-4% positive no matter what the oxygen sensors said, it would drop to -23% for about 30 seconds, then come back up to 3-4% positive, and it kept repeating that.
It was time to put myself into the mind of the computer.
This computer is not very smart. It’s old and doesn’t have a lot of complex problems, nor does it have a lot it can try to do. Was it programmed to react to an extended rich condition by trying to take away a bunch of fuel, then wait to see what happens? And to repeat that as its only strategy, even if it isn’t working?
Was the old Mass Airflow Sensor just far enough off that it screwed up the algorithm, so the computer never tried to take away fuel because the numbers were far enough from what it expected that it never fell into an algorithm that the computer knew to use?
This was possible; newer cars tend to be smarter about this, whereas older ones had limited tools at their disposal and did the best they could.
That still doesn’t tell me what’s wrong with this, but does tell me that so far the diagnostic path has been good; I have taken a vehicle which never attempts any type of fuel control and now it does. It can’t do it, but it’s trying.
Hardly the answer we want to tell the customer, but it’s progress. I’m leaning towards the injectors being the wrong ones, but obviously before I go to all the effort of pulling one, getting the part number off it, cross-listing the part number (because they aren’t Ford injectors) it would be worth getting the fuel pressure, too.
Turns out that the fuel pressure is 100psi. Off the top of my head (not looking up the actual numbers), I know that this engine should have around 30-40psi, and I don’t think the fuel pump should even be able to produce 100psi, which makes me think that the fuel pump is wrong and the fuel pressure regulator can’t deal with it.
My manager thinks that the pump should be able to make that much pressure, and the fuel pressure regulator is bad. He also things that the pressure should be 65psi.
Either way, the pressure should not be 100psi, and the fuel pressure regulator is easier to change.
We were both partially wrong. The factory (or factory-equivalent replacement) fuel pump can indeed produce 100psi, and the fuel pressure spec is 30-45psi. With the new regulator in it, the truck now has 30psi of fuel pressure, and almost immediately upon startup, it’s running well.
One of the downsides of this old Ford is that I can’t reset the fuel trims with the scan tool; I have to disconnect the battery cables and touch them together to clear the memory, and the battery cables are as sketchy as anything else electrical on this thing. I also can’t get a good barometric pressure reading without doing a full-throttle pass, and it’s started to snow and the roads are really slick.
I’m also anticipating that it’s going to take a while to even go into fuel control. Just the same, I take a short test drive and only almost crash three times. Once when a semi goes a bit over the centerline, once when a late-model Fiesta slides through a stop sign, and once when I try and get the Ranger over 30mph.
And, unexpectedly, it does actually go into fuel control and starts adjusting its fuel trims based on what the oxygen sensors are telling it.
One of the things I frequently see is vehicles that have clearly had a problem for a long time, and that always complicates diagnosis. With a customer who’s on top of things, it’s usually a sure bet that all the failures I see occurred simultaneously, and all are related. With this? Who knows? But I can make some guesses.
The Mass Airflow Sensor looked new, and it looked like a cheap part. We have a saying that sometimes ‘new’ means Never Ever Worked. But it was close, and it wouldn’t bias fuel trims so badly that the vehicle would run noticeably bad.
The upstream oxygen sensors probably hadn’t worked for a while, either, and the poor computer had been in failure management mode, just going off its pre-programmed instructions for what it should do when it can’t go into fuel control. I doubt this computer was programmed to be clever enough to recognize that the Mass Airflow Sensor was slightly wrong, but I know, since it eventually set codes for the upstream oxygen sensors, that it was capable of recognizing that something went wrong there.
And then the fuel over pressurization happened, and now this poor computer’s got a hopeless task. It doesn’t know what’s gone wrong, although it probably is aware several things have gone wrong . . .
. . . enough things, perhaps, that it finally just gives up and relies completely on its default programming.
To be fair to Ford, there just wasn’t a lot this computer was programmed to be able to do, and even the most modern computer system on a car can’t fix a mechanical problem—it’ll do its best to adapt, but programming only stretches so far.
Venti pony ^^
I miss my 1998 2 door explorer, and I miss my 2003 expedition, but after this saga, I kinda miss them a little less
I bought an OBDII reader for those dumb things and now my big chunky 1988 G20 has OBDI lol
5648958
How barbatic
5648965
Hey, had a feeling that the tone deaf bard would come back around at some point
5648966
Can’t be Tone Deaf as a pony, even ignoring that’s already one of those eye searing OCs out there.
Probably still loves the wine, though...
We have a 2004(ish) RAV4 in right now. Cranks / no start. It sure tries though. Got fuel, ignition, etc, and the timing is right. Fans run all the time, though. Look at the data for ECT (because the fans are on), and it fluctuates between 60F and 234F. Which would indicate a short to positive or the sensor return. Look at the ECT, IAT, and trans temp simultaneously, they are all doing that. Sometimes they just stay high. Sure enough they share a sensor ground / return. But nothing is wrong with it. Wires all fine, no shorts to each other, ground, or positive. So I pull the ECT wire out of the ECM... still reads that.
The passenger side floor pan is soaking wet. ECM is behind the glove box, bone dry. Sure enough, though, once I cracked it open, signs of water ingress. The tiniest little bit, but damaged bad enough that (me) repairing it is basically not an option.
I once owned a car whose previous owner took the attitude "It’s burning/leaking enough oil that I'm always having to put in fresh. So, why bother changing the oil?"
Theres the classic one where the xompany puts a standard plug in a standard location, then ajusts the trim and build so that the location is secure, then adds a Bluetooth or Wifi option etc to make it easier to do a quick read, then, because the password is a random number for each connector, puts the lable on the connector before installation, so that you have to remove the connector to be able to read the password.
Sounds like the New Holland skidloader that is new to the farm which is now giving an error message-- a certain valve not in expected position. However, I need to find out what the three-letter term means so I can figure out where to start looking... is it a hydraulic valve? Air? Some odd subsystem that does something clever, until it doesn't?
And if it suddenly starts working right, sometimes the best option is to back away slowly and keep eye contact. We have a few Cisco phones with really wacked-out programming like that. They ring at the desk, even though they shouldn't with the settings we have on the programming web site, so no touchie.
venti as a pony was not what i knew i needed
My parents have an 86 ford ranger. The fuel gauge doesn’t work (so they just reset the odometer when they refill it) and something drains the battery (probably the aftermarket subwoofer) so they just unplug the battery when it’s not in use.
Interesting as usual; thanks. :)
i.redd.it/t7fpb9v27ms81.jpg
5648958
Venti pony is happy pony
5648964
I can tell you that I own (currently not running) a GMC Sonoma that is one generation dumber as this Ranger as far as engine management goes. In fact it has the same operating system as your “big chunky 1988 G20.”
In some ways those are easy to diagnose, since there’s not much that the computer can monitor. One protip I can offer, the computer has no way of checking if something it wanted the engine to do actually happened, and it has no rationality checks . . . in the case of my Sonoma, the speedometer cable is broken, and it almost never threw a code for that; as far as the computer programming was concerned, the engine running for hours at various RPMs with no speedometer input probably meant I liked sitting in the driveway revving it.
I don’t know if there’s a cheap OBD-I reader available. I assume you know you can short two of the terminals on the DLC to get it to flash codes, which isn’t as good as seeing data, but it’s better than nothing.
If you didn’t know that, you can short two of the terminals of the DLC to get codes. You do that with the engine off, and IIRC, you’ll always get a code 12, no RPM signal, since the engine isn’t running.
5649937
I've seen a few, but they just have an LED in them to flash those codes, and like you said, you can just short a few pins to do it yourself, heh. My speedo cable is an actual physical cable running from (I assume) the transmission to the speedo, I had to lube the darn thing not too long ago because it was making so much noise...
5648980
Those things are the worst. Especially when it’s a stupidly simple problem with a very expensive fix. We had an Econoline in some years ago that ran fine most of the time but would occasionally fall on its face; the issue was a blown capacitor on the PCM. Probably an easy fix for someone who’s confident in repairing a computer and who knows where to get an appropriate capacitor, but that’s not me.
That’s one I wish I’d taken a picture of; after we got it to do it and I figured out the right search terms to put in Identifix, I was pretty much led right to the issue; apparently, it was a not uncommon problem for that particular computer (I think it was EEC-IV, but I don’t remember for sure).
I lost the airbag computer on my Regal due to flooding. Given the condition of the rest of the car, I wasn’t motivated enough to replace it.
Also I don’t think I ever made a blog post about it, but there was a GM car (I think a Cruze) I worked on a while back that had serious water ingress in the trunk and it drown the DC/DC convertor, fuse box, and a few other components. Easy diagnosis when I knew where to look; they were still underwater.
5648987
I also owned a car like that. On the plus side, it rustproofed itself. On the downside, the oil was constantly so clean it was hard to see on the dipstick. On long road trips, i’d stop at every rest area, dump in a quart, then check the level.
5648990
Connector location is government-mandated (in the US), and automakers pay big fines if it isn’t where it should be, or worse as VW found out.
Here in Michigan, they’re mostly gone, but there were a few cars in the transition years that had a normal (for the automaker) connector where they usually put it, and an OBD-II connector hidden somewhere on the car. IIRC, on the Ford Thunderbird, it was under a trim panel in the center console.
5648994
There’s a constant debate in the technician world (and the people who make things for technicians, like scan tools) about whether you should use the automaker’s name for the thing, or a universal name. There is no consensus. To a Ford guy, BOO is something he knows, but anyone else might have no idea what that is, whereas if you call it the “brake switch,” anyone who knows cars will know what that is and what it does.
One time my manger, who worked at a GM dealership for 20 years, got salty when I called the EBCM (electronic brake control module) an “antilock brake module.” (Which is what it is.)
The very next car I looked at was a Ford; while GM calls the climate control module the HVAC module, on that model, Ford called it the EATC module (Electronic Air Temperature Control).
Since then he hasn’t grumbled too much when I use generic terms for a module.
(The instrument panel that you look at on a vehicle might be called a IPC [instrument panel cluster] on a GM, a HEC [hybrid electronic cluster] on a Ford, a CCN [Cabin Compartment Node] on a Chrysler, a COMBI [who knows?] on a VW/Audi/Mercedes, or a METER on an Asian car, and there are certainly other acronyms I can’t be bothered to remember at the moment.)
5649033
One of my trainers once wisely said that when we’re diagnosing/repairing a check engine light, our goal is to make the computer happy. If it is, don’t ask too many questions.
“This phone does things and we’re not sure why it does things but it works so we don’t worry too much about the rest.”
5649150
Protip: x as a pony is always what you need. Full stop, no exceptions.
5649163
I have not owned a single car with both those problems, but I have owned cars with each of those problems. One of my S-10s got the trip odometer fuel level ‘repair’, and if I wasn’t sure, put gas in it. And I had a van with a mystery battery drain that the previous owner put a race-car style shutoff switch for the battery in; since that seemed to work, I never bothered to fix it. The only minor annoyance was that the radio had no memory as a result, but since I listened to stuff through the AUX port or the CD slot, that didn’t really matter.
5649527
You’re welcome!
5649829
Sadly, it’s not like that here. It keeps trying to spring, but the spring stays ground-bound.
5649938
Yeah, ditto on my Sonoma, and it had a ‘sensor’ on the back of the speedometer that fed that information to the computer.
A couple model years later they went to an electronic pickup on the output shaft. The little gear adaptor box on mine went bad, and it either wasn’t available aftermarket, or I didn’t feel like paying for it.
5649948
Whew! Around here there's a junkyard that I call The Van Graveyard, it's full of old vans just like mine, and I've gotten so many obscure parts from there that I'd never be able to find online...
5649939 lol years and years ago, we had a 2002 Audi A6 Quattro with the 2.7L twin turbo v6. Transmission didn't really work. The customer rather quickly cut it loose, and sold it to us. Those cars always have water ingress in the TCM. No brainer repair, tbh, the TCM is on the bottom of the passenger's floor pan. Moron engineers. What shocked me about this one is that it was only marginally wet on the outside. So I broke open the aluminum case, and found the computer board had become a big ice cube (it was winter).
I probably told you this one before. The reason I remember the car so well is because we turned around and sold it to a very good friend of ours, a machinist, who gave it to his wife. She apparently loves the thing, so we've been keeping it running for the past 10 years or so. It is actually a nice car. Those two turbos give it plenty of pep, it's all wheel drive for snow, and it's very well equipped.
5649949
That’s always nice. One thing I noticed back when I drove wrecker (a couple decades ago) is when we auctioned off the unclaimed vehicles in our lot, vans and pickups always got sold, presumably because junkyards could make money off them.
5649958
Speaking of water ingress, the early 2000s Ford trucks often get water on their in-cab fuse box from windshield leaks, and I’m working on one of those right now. Well, sort of; we’re waiting to see if we can get parts. The junkyard isn’t being helpful, and my manager misunderstood the part I need. And I doubt just the fuse box is going to fix it, but I need the electrical system to work correctly so I can do further diagnosis.
... we have a different clientele; the few that we’ve sold have been ‘well, it still kinda runs, and you’re not going to find a better Ranger for $300.’
I’ve bought four cars for personal use from customers; three of them were literally on the way to the junkyard. Amusingly, the one that theoretically had the most life left in it actually lasted the shortest.
5650154
Someday, I'll find a set of the power mirrors that they had on the 1994 G20... SOMEDAY
5650193
Good luck!
My best find to date (for an obscure automotive part) was the ladder for the back door of a 1973-1992 Suburban. It was offered on a trim package called the ‘Outdoorsman’ for some years, and I found a guy parting one out.
5650355
WELL GUESS WHAT BITCH I FOUND A SET
GOT A DRIVER'S SIDE MIRROR AT THE JUNKYARD
GOT A PASSENGER SIDE ON EBAY
AND THE SWITCH ON EBAY TOO
AND I FOUND THE WIRING DIAGRAM
WE HAVIN POWER MIRRORS FOR 2022 BITCHES
so yea that'll be fun to set up
5650156
That's how I got my Accord. This lady came in with a broken axle. I was suspicious, but changed the axle, and down the road she went. Came back rather quickly, broke the fucking new axle. Like, in the middle of the shaft. Somehow or another, the transmission was locking up whenever it felt like it. Weird symptom, actually. But here we are. She was from Iowa, a long way from home, and just sold us the car. Rebuilt the trans (that's what we do :P ) and here I am. Damn thing has 270k miles on it. Nice clean car though, you've seen it lol.
5650649
Huzzah!
On an old-school van, shouldn’t be too much in the way of wiring. Power and ground, and don’t forget the fuse!
5650667
The idea that a transmission can lock itself up enough to snap an axle, then free itself back up and be driveable just to demolish another axle later is kinda terrifying. Although I suppose that’s better than the axles holding and the front wheels locking up. . . .
I had an old, not-so-great S10 with the 2.8L carbureted engine, and one day I was cruising down the highway minding my own business when it decided it wanted a lower gear. The tach went over 7,000, and I heard the transfer case hit the cab floor. Somehow, everything survived that, and when the RPMs dropped to something more sensible, it shifted back up and acted normal.