The Audi Q5 has been an important volume seller for the company in the luxury SUV space. However, the German automaker is now tangling with a difficult recall due to reports of multiple fires in customer vehicles.
According to documents filed by the automaker, the recall concerns 89,417 vehicles sold in the United States, and a further 16,863 vehicles in Canada. Affected models include Audi Q5 Quattro and Q5 Sportback Quattro models built from roughly July 2021 to August 2024 in Audi’s factory in San José Chiapa, Puebla, México.
Audi is recalling the vehicles due to engine oil leaks from the cylinder head. Specifically, screws holding the cylinder head cover on have been found to loosen in production vehicles, allowing engine oil to leak from the top of the engine. The oil may then leak onto hot engine components, posing a fire risk. Audi notes that owners should be wary of unpleasant odors or smoke in the engine compartment, which may be an early warning that an oil leak exists.
Audi first became aware of the problem in May 2023, with a number of customer complaints regarding cylinder head oil leaks from the 2.0-liter TFSI engine in the Audi Q5. As per a timeline filed with NHTSA, Audi engineers initially believed the problem to be down to burrs on screws used to affix the cylinder head cover. However, when manufacturing changes were made to solve this problem, oil leak reports continued to increase. From May to October 2024, Audi began fielding reports of “thermal events”—which is usually corporate speak for fires—while engineers were able to trace the problem down to its true root cause.
Ultimately, the cause of the leak came down to production issues. Audi estimates that for 5% of the total recall population, with the cylinder head cover screws were not properly torqued down by the factory. Thus, over time, the screws may loosen, which compromises the seal of the cylinder head cover and causes the leak. Audi rectified this for future production by reviewing its torquing process and changing its screw supplier.
Thus far, Audi has received a total of 2,539 potentially related oil leak claims in the US market. Meanwhile, it has recorded 11 “thermal events” it believes may be related to the issue. Thus far, no injuries or deaths have been reported regarding this matter.
Thankfully, the repair for the issue is simple in cases where it has been caught prior to any fire or serious damage. Dealership technicians will inspect the cylinder head cover screws and properly torque them or replace them entirely where necessary.
It’s a simple mistake on Audi’s part—just a few loose screws. And yet, when you’re torquing up thousands of screws on thousands of vehicles every day, the results of a simple mistake can be quite large indeed. It’s a reminded for engineers at major automakers that every last screw, nut, and bolt really does matter.
Image credits: Audi
Nice to know I’m not the only one 😀
“ Ultimately, the cause of the leak came down to production issues. Audi estimates that for 5% of the total recall population, with the cylinder head cover screws were not properly torqued down by the factory. ”
That’s not a root cause, that’s a description of the problem.
Why were they not torqued? Why 5%? Is it a production process? Is it a part manufacturing issue? Loose bolt isn’t the answer, and if I owned one of these it would bring me no faith at all that everything else is done up.
Seeing how part of the fix was changing the supplier for that bolt, it means they were out of spec leading to false torque values. If Audi assembles anything like GM, each install gun (effectively an electric ratchet) has an integrated torque sensor. When that sensor sees correct torque, it shuts off. This tells the operator and line computer that everything is in spec.
Remember that the operator (line worker) does hundreds of these specific fasteners each day and is working as fast as possible. They’re trained to trust the tool, as it’s much less error prone than a human the vast majority of the time.
My bet is the supplier was not meeting spec, and the torque value was artificially low for what that ‘screw’ and joint could actually handle.
“Audi rectified this for future production by reviewing its torquing process and changing its screw supplier.”
The target torque would be whatever WV’s standard is for an M6 or whatever. For a bolt preload to relax after the correct torque is applied is weird, unless the clamped material creeps (which is why plastic parts have steel inserts) or for the bolt to have a defect (burrs or poor quality plating) on the clamping face that plastically deform, giving much lower than expected preload for a given torque.
However cross-threading, or a damaged thread would give you a high torque without generating the right preload.
It seems odd to me to change the fastener supplier to fix this, as adjusting QC at the current supplier is easier than resourcing (which ought to trigger revalidating the joint, which is expensive and time consuming).
Typicall the assembly tool would be a multi spindle nut runner, tightening the entire joint in one go. Loads of scope for that to go wrong without anyone noticing.
You’re spot on about compression limiters on plastic parts.
At the end of it, all torque is measured by the toolhead, so the spec is for a given quality fastener in a given quality threaded bore. If either is out of spec, you will end up with lower clamping load due as the tool can only sense torque applied to the fastener.
Switching suppliers tells me that the supplier either could not maintain quality (surprisingly common) or was intentionally trying to pass through known bad fasteners (personally experienced this with Chinese supplied parts – not specifically bolts).
There’s no need to revalidate the joint as general purpose fasteners are supposed to be a spec governed, standardized part. If Supplier A and Supplier B are both following the relevant ISO spec, their parts should be functionally identical for dimensions.
Having to change suppliers to get better fasteners to the same spec disproves that fasteners from different suppliers are functionally identical.
I’d revalidate the joint, especially if the changes were the result of a recall involving risk of flaming death.
Finally, an automaker I can relate to. I’ve been told I have a few screws loose as well.
“It’s a reminded for engineers at major automakers that every last screw, nut, and bolt really does matter.”
And a reminder for proofreaders.
I have volunteered to catch stuff like this. I’m retired and would be happy to make that part of my days.
The Mexico production vehicles are so much worse than the one’s that come from Europe. Oil burning, PCV issues, blown out timing covers and oil pans. I’m glad my GTI was made in Germany. I know it’s still not perfect (I’m looking at you water pump) but hey I might have a chance.
I once worked for a major manufacturer. After the legal issues GM faced as a result of the faulty ignition debacle, in particular because engineering had pointed out the issue before release, all of the engineers at my company had to go through ‘Right Speak’ training. This training pretty much boiled down to: “don’t write anything down which could ever make the company look bad in court”. ‘Thermal event’ was one term we were told to use, instead of fire because, and I quote: “Without witnessing it, how do you know it was a fire and not just that it was very hot?”
We were also told not to write down any of our safety or failure rate concerns, or put them in an email.
Every one of my peers and I walked out of that meeting disgusted. As an engineer, I am continually astounded that we continue to let lawyers run the world.
I work in healthcare and I second this. As soon as lawyers get involved everything goes to shit.
When I was told to use “thermal incident” instead of “fire” I started using the phrase “risk of flaming death” in emails instead.
Sure, it makes the lawyers unhappy, but if I’m reporting on something that might lead to risk of flaming death I want it to be taken very, very seriously, and not swept under the legal carpet.
As a company are we trying not kill people, or trying not to get sued?
If you do the first one properly the second one happens for free.
I’ve had many cars over the years that emitted smoke and unpleasant odors from under the hood.
I’ve been told by the VW/Audi fanboys that the EA888 is god’s own engine and that my disdain for it is misplaced
Long live our EA113. The oil filter housing seals were a pain to deal with but replacing it with an aftermarket metal version solved that issue permanently. They can lecture me for hours about the wonders of timing chains, I don’t care. A timing belts is cheap and easy, I can do it in an afternoon for $150. The timing process on a 888 is a nightmare. Our neighbor up the street finally reaninated his long dead MK7 GTI after it dropped a valve because it felt like it.
Is this the EA888? if so, will Macans be affected by this?
It is but if memory serves me correctly, the Macan is actually built in Germany – meaning different suppliers, different works assembling the engines, and different factories. The Q5 in question is built in Audi’s factory in Mexico with different suppliers. You should be good.
This being a VAG product, it catching on fire may or may not be a desirable outcome, depending on who you ask.
Loose Screws Lose Qs
And when they’re lost, the dealer writes “Q-ANON” on the hood and parks them behind the pizza place.
Where they are promptly incinerated by the lizard peoples space lasers.
Mucho gusto, a new TSB. It’s leakier than old MGBs! And if you want to get fixed, physically, it will be over in a minute (if you wait). So we’ll tell you to come over and you’ll take that trip. Then we’ll put your Q5 on a lift. Then we look for the drip-drip-drip, that’s how we know the bolts have suffered from a…little slip.
Audi called, Ungh, we gonna torque wrench it down and we say UNGH, it’s that recall sound.
My wife teaches middle school and apparently Sublime is having a bit of a renaissance with the children. The kids are alright?
As an OG fan of the band (in its time and place), I’m always a little amused to see Sublime shirts on all these MS/HS kids. Especially the girls.
I want to do a supercut of a select few songs to see how it all goes down in the modern era. I don’t think it works 🙂
This sure is an easier fix than the oil-burning 2.0t with piston ring blow-by. A few torqued or replaced screws without having to disassemble the engine? I’d take that over the alternative!
Trying and failing to mark its territory. Need to make the oil pan bolts loosen.
But the smell of burning oil is the only way I know it still has oil! Since when is that a red flag in a VAG car?
What’s a VAG car without it’s Self Lubricating Chassis System?!