noise."Mechanic: Chevy Tahoe with a usually reliable code · 2:05am Jun 28th, 2023
Today we're gonna talk about a Tahoe and a usually reliable code. You know what to do:
Before we begin, I'm gonna give you an important PSA, courtesy of Cerulean Blue. If you have a car or truck that uses pushbutton start, get a protective box that blocks RF signals to keep your keys in whenever you're home. There's a technique to increase the effective range of the key that thieves can use which makes the car think that the key is next to it or in it when it isn't . . . and you can buy a tool that does this online, so it doesn't take a technically-skilled thief to manage this. Just one who can read instructions (and who's willing to order something from a very sketchy website, but then he's probably not using his credit card to order it, so. . .).
If you've been reading these for a while, you know that a lot of times the code that sets isn't always the problem. Ferinstance, just because a car set a code for the mass airflow sensor (P0101) it doesn't mean that the MAF is bad, just that its value doesn't make sense as far as the computer is concerned. It might be caused by a vacuum leak or a clogged exhaust or a stuck-open EGR valve or stuck-open purge valve. . .
But there are a few codes that are always reliable, and ironically they're ones that are based on complicated algorithms. Two which I've never seen be wrong are a P0420 and a P0430—bank 1 and bank 2 catalyst efficiency. At some point when the engine is warm and running at a specific RPM/load, the PCM runs tests on the convertors. What happens in them is chemistry, and the engine knows all the operating parameters, it knows what's going in and knows what should come out, and it depends on the upstream and downstream oxygen sensors passing all their self-checks.
Is it possible for an oxygen sensor to fail in such a way that it passes all its self-tests but also somehow fools the computer into thinking that the catalytic convertor isn't working properly? It might be, but I've never seen it happen. It would be a very strange failure mode.
The same holds true of P0128. The code description varies by manufacturer, but in essence it's a code telling you that the engine didn't warm up as fast as it should. The engine uses various inputs to calculate how quickly the engine should come to operating temperature, and if it doesn't, it sets the code. All the sensors which provide inputs for this calculation have their own self-tests: for example, we've discussed before how on startup, if the car has sat for long enough, it cross-checks all its various temperature sensors which should all be the same; if one isn't, it's probably wrong and the computer will flag a code for it.
Which brings us to the Tahoe.
I'm sure that you're already guessing that it has one of these codes, and you'd be right. The customer's complaint was that the temperature gauge didn't work, and the cooling fan ran all the time. The check engine light (MIL) is on.
Naturally, when I pulled it into the shop, the MIL wasn't on, the fan wasn't running, and the coolant temperature gauge was working. I pulled codes, and the only one was a P0128.
Since at this point, I still believe that the customer heard what he said he heard (and I know that the P0128 will turn the MIL on*), I theorize that on this vehicle, when it sets that code, part of its strategy is to turn the cooling fan on high speed, just in case the coolant temp sensor (CTS) isn't operating correctly and the engine is much hotter than it's reporting. Some GM vehicles do this if they set a coolant temp sensor code.
Some diagnosis indicates that the coolant temperature sensor is reading accurately: I used my FLIR (infra-red camera) on the cylinder head next to where the CTS is mounted and various other places in the cooling system.
Source
______________________
*Some codes will illuminate the MIL intermittently, depending on if the failure is currently happening, or if it happened in the last few drive cycles. Oftentimes they'll clear the MIL if the test passes or isn't run for several drive cycles; since the test for a P0128 relies on the engine temperature starting off below a certain setpoint and needing to run for long enough to reach its operating temperature, if it was taken on a bunch of short trips, it wouldn't have failed those test because they wouldn't have run. I don't have the numbers for a Tahoe, but on a Chrysler Sebring I recently diagnosed, the engine start temperature had to be below 77F (25C) and it had to run for up to 22 minutes before it would fail the test [that was an older, dumber car; 'below 77F' means that in many places it'd never run that test in the summertime.]
The Tahoe wasn't fully warmed-up yet, so I took it for a test drive to see what would happen. P0128 suggests that the thermostat is opening too soon or not fully closed all the time.
Explainer:
Source
Engines are designed to operate at a particular temperature range. If they run too hot, they melt; if they run too cold, they're not as fuel-efficient (or they pollute more, or both). As such, cars are equipped with one (or more!) thermostats which keep the coolant circulating only in the engine until it comes up to a specific temperature, at which point the thermostat(s) open, routing coolant to the radiator.
You also need a thermostat if you want an effective heater, since most heaters rely on hot engine coolant circulating through a heat exchanger to heat the passenger compartment.
Generally, a thermostat uses a calibrated wax pellet to push itself open at a specific temperature. Most engines have thermostats which are calibrated from 180 degrees to 220 degrees (82-104C).
What I was expecting to see was the temperature to rise quickly to a point, and then slow down before it got to operating temperature (180 on this engine), and that's what I saw. By the end of the test drive, it had made it to 176 (80C). The cooling fans did not come on.
The check engine light was probably more on.
The manager's suggestion was to replace the coolant temperature sensor first. We'd worked on this thing before, and we'd replaced the radiator, the water pump, and the thermostat; he wasn't sure if we'd replaced the CTS.
It's easy to get to, and I can swap it fast and not lose too much coolant, so I changed it and took it for another test drive, and it was exactly the same as before.
Actually, that was a lie. It was not the same, it was actually worse than it was before. Last time it made it all the way to 176, this time it struggled to reach 165 (74C).
Since we try to approach these things from a scientific perspective, why might that be? Well, the new sensor might be slightly different than the old one; maybe it's a little biased low—or maybe the other was biased a little high, and this one is more accurate. Maybe the extra new coolant I added to top off what I lost made a slight difference. I don't think it'd be ten degrees (about 5 celsiuses) though.
Either way, unless we've got a defective one and the previous one was also defective (but maybe not as defective), the problem isn't the coolant temperature sensor.
The good news is that there are ways to check it. For example, I could use my FLIR to get a temperature reading of the hose or the sensor body. I could use my infrared gun to do the same (it's got a laser pointer, PEW PEW). Or I could use my fancy thermocouple-on-a-stick and touch the CTS and see what that says. Also I can grab the hose and use the Mark I hand to get an idea; 165 and 200 feel different.
(I often do an initial back-of-hand temperature test on cabin heating vents; on most cars if I can comfortably hold my hand there, it's not blowing hot enough air.)
Easiest tool to grab is the PEW PEW infrared gun, and says that the body of the CTS is about the same as what the CTS says it is (couple degrees different, that's close enough).
So obviously, the thing that needs to be changed is the thermostat.
Did I mention up above that this thing has been here before for cooling problems? If I didn't, I can tell you that it's had a thermostat and a coolant temperature sensor and a radiator put in over the course of the last year.
I didn't think that the CTS was bad, but it was easy to change and didn't make much of a mess. The thermostat's a little more difficult (only a little) and will make more of a mess . . . before I dive into it, might be worth doing a quick check in Identifix to see if there's something stupid I'm overlooking. You know, like a toggle switch in the engine compartment that says "engine heats up on/off."
And in fact, I do find something. It turns out that some aftermarket radiators for these were defective exceed factory specifications.
You might have noticed that I refer to 'degas bottles.' Back in Ye Old Times, extra coolant in the radiator just dumped on the ground. Depending on the model and the legal obligations, by the 80s most automakers had realized this wasn't ideal, so there was a coolant overflow bottle that collected the extra; as an added bonus, when the engine cooled down it would suck in coolant if the radiator had started out low. Assuming that there was coolant in the overflow bottle, anyway.
It was also safer for the customer; you could check the overflow bottle and see if there was coolant in it, rather than having to open the radiator and maybe get a hot coolant shower which I can tell you from personal experience hurts a lot.
These days, most cars use degas bottles. The bottle is pressurized (overflow bottles weren't), and it has some small hoses at the top to let air out. You don't want air in your cooling system . . . and that's the 'degas' part of the name. There's also a big fat hose on the bottom which supplies coolant to the engine. When the engine is hot, and the thermostat is open, the degas bottle will be in circulation with everything else; when the engine isn't hot, it will allow in small amounts of coolant to make up for air in the system, but there shouldn't be much, if any.
In the case of the Tahoe, there's a small hose from the top of the radiator that goes to this bottle. Just enough to let trapped air bubbles out.
Since they don't want coolant circulation through the degas bottle when the engine is heating up (that would be extra cooling area), there's an orifice in the radiator. The hose might be 3/8" (10mm), but the orifice is 5/32" (or 5/64", I don't remember; whichever it is, that's smol [4mm or 2mm).
Some aftermarket radiators don't have that orifice.
Techs on Identifix reported that an effective repair was to drill the appropriate sized hole in a tire valve cap, and then jam that into the degas pipe on the radiator. We've got lots of valve caps (most shops do) and I've got a little tiny drill, so I gave it a shot to see what would happen.
I'll admit, I was kind of surprised that it worked like a charm. Instead of struggling to get to 165, it quickly came right up to temperature and then just sort of hovered around there like it ought to.
I am curious why it was nearly a year before the light came on . . . it could be how the customer uses the vehicle, it rarely is operated in the conditions where it can run the test.
Whatever the case, it was fixed, and all it took was one used valve cap with a carefully-calibrated hole drilled in it.
This hits home. I don't bother escalating anything to a vendor/place unless I have a photo, video or logs.
That said, I love being able to just whip out my phone and photo/video something so I can use it as evidence to investigate further/lodge an issue.
Really like the explanation.
If only they were all that easy...
I've read that car companies could make cars a LOT harder to steal. However, car thieves & repo companies pretty much use the same techniques.
Thieves can't steal your car = you can't get a car loan because the lender can't recover your car if you default on the loan.
LoJack claims 90% recovery rate vs about 12% for vehicles without a LoJack. They also claim it is difficult to jam or remove the device.
Hooray, I get to be the exception! Possibly.
For at least six years, my first car had the MIL on with a P0420, but no shop ever found a problem, and every time it got smogged the emissions were spot on. I replaced both O2 sensors but it didn't make a difference. California requires the MIL to be off to pass smog, but I got lucky: every place I took it to would do something to pass the car, then invariably the light would come back on on the way home. Except the one cool place that decided the gas cap was at fault, they tested the gas cap separately and it passed, so they passed my car. I never did find out what the problem was; I just left a piece of masking tape over the light for years.
But that did come in handy when that car was on its last legs. California has, or at least had, a program where if your car drives but doesn't pass smog they'll pay you $1,000 for it, to get polluting cars off the road. Again, the emissions from the tailpipe were perfectly fine, but the light was on. And since the car was probably worth less than $400, that was a win.
P0420 is always *right*? In what way?
I just had a Ford truck come in with that, and I dismissed it as 'fuck it, more 'we don't like what we're dealing with' OBD II whining. Tell me if that's wrong, and it's about to go really wrong.
(I mean, in comparison, we also had someone the other day night-drop an old F650 where somebody - and I can't prove it was the customer, but I have my suspicions - tried to remove the catalytic converter, and screwed it up because they're jackasses. But managed to disconnect the exhaust from the cat.)
5735243
The truck with the attempted theft of the cat, the cat had been painted orange. My guy (who was a former mechanic, so I tend to believe him when he says something is wrong) looked at that and said 'this cat has burnt out - it's orange!'. Turns out, they started painting the cats orange a few years ago, and my guy is just confused - he's well over 70, and easily confused.
The other week, we had a drunken jackass return a truck which didn't want to drive, because the power steering and power brake reservoir had blown out. Our best guess is that they'd tried to drive it out of a bad bogging, based on how filthy and mud-daubed it was. My ancient customer service rep guy, who theoretically knows far more than I do,being an over-educated dipshit as I am, said 'look, they must have hooked the tube, and all the power fluid poured out!' No, that was the sight-bubble. Which somehow got blown out, probably while they were trying to power the truck out of whatever bog they got it caught in.
5735258
I used to work in a convenience store gas station back in the 70s ( when the government decided on catalytic converters). They couldn't deal with leaded gasoline. It would kill them. Leaded gasoline was cheaper because the lead boosted the octane rating.
So, unleaded cars had a gas tank doohickey so leaded gas pump nozzles wouldn't fit. People being what they are, some people would wreck that doohickey so they could use leaded gas then wonder "Why did my car just fail its emissions test?"
Won’t the valve cap fall out ? How is secured?
5735270
The technical term is likely 'interference fit'... basically it is physically jammed in there so firmly it cannot dislodge itself. There is no effective pressure differential affecting it as both sides of it are within the pressurised system.
This does not need to be some fancy box from some online seller. Just a simple metal key hanger case will do.
All my wheels are waaaay to old for this ('67 to '96) but for a friend I simply used an old metal keybox from work and we hung it where their keyhanger was near the front door. As long as you closed the door on the keybox the remote is effectively shielded from responding to the amplifier/relay devices the tea-leaves are using.
Reminds me of my old '97 Geo Metro. It developed a problem where the engine would start racing for no apparent reason, just zooming along. I'd have to kick it into neutral, and in a few moments it would go back to either normal or a low-RPM state. Took it into the shop, and they couldn't reproduce the error. Took it home, and within a day it was back to doing the same thing. It only started doing that after I'd been driving it for a while, so I didn't think they'd tested it properly.
Took it back to the shop, this time they were able to replicate it, by letting the car run a lot longer. Turns out, it was a failing temperature sensor. The PCM controls the engine RPM based on temperature, running it faster when it's cold, then dropping it into idle speed when the engine warms. The way the sensor was failing only happened when the engine was up to temperature for a while, it worked normally while getting up to temp and holding there for a bit. The failure resulted in bad data in the PCM caused it to run the engine much faster than it typically would when it was cold, I'd kick it into neutral and let the engine race for a bit, then something would reset and drop the engine back down to idle, then it would fail again and race the engine, rinse lather and repeat. The shop replaced the sensor and said it was idling normally after that.
Took the car home, and the very next day the problem recurred.
Took it back to the shop and they ran additional tests. Turned out, the brand and model of sensor they used was listed as fully compatible for my model vehicle, but actually wasn't, and had a similar bad data issue. They went through three different brands before finally finding a sensor that was fully compatible and functioned as expected.
5735327
Good joke.
Which version was it? The 3 cylinder or 4?
Substandard 3rd0-party parts can be a pain; sometimes they are just as good as the manufacturer's, but sometimes they kill thet hing taking them.
5735397
The 4 cyl LSi version. Not quite as good fuel efficiency as the 3cyl; but living in Seattle, I needed the extra cylinder to get me up those hills.
I've managed to get it up to 98mph on the straightaway, admittedly with a good tailwind.
5735448
well, yeah... you had the fourth cylinder. The one I rode in was only the 3... passenger weight affected top speed. A full load of people meant you had to have the pedal to the metal on the highway to get to speed.
5735480
Reminds me of the time I was headed out on a field trip with two classmates, riding along with one of them in her 3cyl Metro, three of us total. Going up one of the hills she had to downshift, and we barely made 25mph (in a 40mph zone) the whole way up.
5735237
Oh, if you have pictures or videos that's very beneficial. Or can duplicate it on demand . . . some customers are really good at descriptions, others not so much.
Thank you!
5735238
I know, right? Wait until I blog about the Rendezvous (I probably won't; I'd rather forget that particular vehicle ever darkened our shop.)
5735243
There's a lot of complicated cost/benefit analysis, as well as reliability and convenience concerns. And on most cars, the most complicated security system can be defeated with a tow truck.
The difficulty of removing a LoJack (or any other tracking system) is based on how well it's installed. I've pulled trackers out of cars before. Some thieves take steps to remove them or block them, others just take their chances. I'd say if you've got something expensive it's probably worth putting some kind of tracker on it.
5735255
As far as I know, the P0420/P0430 doesn't necessarily mean it'll fail emissions (depending on the state, what they look at, etc.), just that it isn't as efficient as it ought to be.
Some states do more than check the MIL, and you'd have failed in one of those states. There are readiness monitors that can be accessed with a scan tool, and one of the monitors is 'catalyst efficiency' (or something like that), and on your car it would say 'not run' or 'failed', depending on if it had run the test.
Cat. efficiency codes often take a long time to run, depending on how you use the car.
My dad scored big with Cash for Clunkers; he traded in a very worn out Aerostar that was only worth its scrap value--the last failure before he traded it in was the windshield wiper motor rusting off the firewall). Since it didn't get good fuel economy and he traded it for a Honda Fit he got whatever the max value was (I think $4000).
5735264
They do the same thing with diesel fuel necks, and I've seen gas cars with diesel in them . . . often if you ask the customer, they'll also complain that the last time they filled their tank, it was really hard to get the nozzle in.
derpicdn.net/img/view/2023/5/11/3123785.gif
5735270
It's jammed in, it's a really tight fit. And if it did somehow come out of the radiator, it'd have to work its way through a rubber hose, and it'd get stopped at the degas bottle, where the opening is smaller.
5735281
Yup, it's a tight fit, and even if it comes out, it'll still work if it's in the hose. And it can't make it into the degas bottle, 'cause that's a smaller opening.
Yeah, a metal keybox would probably do the trick. Especially if the keybox isn't close to the front door.
Speaking of which, maybe I should heighten my automotive security. Although none of my vehicles are new enough to have pushbutton, they're probably vulnerable since the keys are in the ignition and the doors are unlocked. . . .
5735256
Really wrong is open to discussion; all that code tells you is that the magic in the catalytic convertor is gone (or mostly gone); if you've got emissions tests you've got to pass that's a problem; if you don't, it's not. Evap codes are often the same way; my van has had the check engine light on for 64,000 miles so far due to an evap problem which I don't care to fix. It might be costing me a little bit more in fuel, since some of the vapors are escaping that shouldn't . . . or else they aren't, and it's just the sensor that's bad.
That's got to be a huge problem for U-haul and other fleets. Heavier trucks are easier to get under, and sometimes have bigger, more valuable cats. One of our customers had one cut off their shop truck, and besides just the cat being gone (expensive), since it was removed in haste with a sawzall and no concern for later reparability, hundreds of dollars worth of sensors and wiring harnesses also got destroyed.
Back in Ye Olde Times, there was a huge theft ring for airbags, since they were basically untraceable. That got busted when automakers started putting serial numbers on them . . . I'm surprised that they're not doing that with cats. I assume 'cause nobody's mandated it yet.
What kind of truck? I'm not familiar with sight-bubbles on a power steering system; none of the normal cars I know of have those. I have seen them on AC systems in older Japanese cars.
5735544
Interesting! I thought that since the emissions tests always came out fine, the MIL was just lying. I wonder if that lack of efficiency could've been related to the worn piston rings...
And yeah, I got to take advantage of Cash for Clunkers with my second vehicle. Even with all of its problems I never saw it get as low as 17 MPG (or whatever the threshold was,) but who was I to question the government? Thanks for the money, Uncle Sam! (I used to say that I'd never sold a vehicle, I just drove them until the government paid me to get rid of them.)
5735327
I've got one right now that has a sensor issue, although I don't know which one yet. It runs okay except when it all of a sudden decides to flood itself with fuel. That's not coming out of nowhere; some sensor is sending it into a bad routine and I'm guessing it's an oxygen sensor but I don't know just yet.
Further complicating the diagnosis is that the vehicle has some aftermarket modifications, some as-equipped systems disabled, and other issues which probably aren't related but might be.
Sometimes it's a challenge to figure it out. And sometimes fixing one thing causes another problem . . . someday I'll talk about blocking codes.
5735407
The best example I've seen of that was some cheaper aftermarket coil boots for Ford Triton motors. Our parts rep showed them off and I happened to be doing a tune-up on a Ford that took them, so I installed them. It ran fine when it came in; not so good with the new boots (two cylinder-specific misfires). Put another eight boots on it, two or three different cylinders were misfiring. Put the ones we normally used on, no misfires.
They were cheap 'cause they didn't work. I suppose you could do a tune-up, figure out which cylinders are misfiring, replace those and hope for the best . . . but are you really saving money at that point?
5735562
Modified GMC Savanna from a couple years ago with about 100k miles on it; rental trucks don't last long. Our most common 10 ft box truck. They just replaced the reservoir on the downed truck with the version with a dipstick.
5735548
I have recently escalated my similar security up at the farm... I've moved the keys to the passenger sunvisor... so much more secure!
5735563
It totally could be. I don't know what the algorithms are or how sensitive they are. If your state does tailpipe testing, they'd stick a probe up the exhaust pipe and readings would be good or not regardless of what the car thought its convertors were doing. It's also possible that it could pass at idle but not at road speed.
No reason not to take free money when Uncle Sam offers it. I got an unexpected windfall from the IRS one year and when I called their 1800 number to tell them I didn't think I qualified for that particular refund, they told me that the IRS does not make mistakes.
It's been almost 20 years and they haven't asked for the money back, so. . . .
5735607
Might be some weird commercial thing I'm not familiar with, then. We had a customer who normally bought 1 ton box trucks which were mostly the same as what we normally work on. Then he bought a Kodiak with hydraulic brakes and we have no idea how to diagnose the parking brake system on it. It's weird.
5735626
That's a big improvement. Most people think to check the driver's side sunvisor, but not the passenger side.