Generic question on longevity of modern automotive engines at high load 2

[LMF5000]

This is a "what-if" question - asked mainly out of curiosity, not because I need to implement any of the scenarios mentioned.

Consider a modern European hatchback designed for fuel efficiency and low price - so low displacement turbocharged engine is the order of the day. Say, 1.0-1.5 liter 4-cylinder petrol engine making 100-200 bhp with turbo. And in a car that weighs about 1200kg. Top speed of around 200 km/h.

Question 1 - imagine the car is driven for several hours daily on the Autobahn, at its top speed. How long would it be expected to last? Which component would fail first? Are cars of this size even designed with enough natural cooling capacity to withstand several hours of full throttle operation?

Question 2 - imagine the same automotive engine, but this time driving a stationary load (like a generator or water pump), or used as a boat inboard engine, or an aircraft engine. In each case assume the engine has sufficient cooling (via cold water supply or oversize radiator), no unnatural axial loading on crankshaft (i.e. propeller thrust loads borne by thrust bearing not directly loaded on crankshaft), and engine spends all its time at 80-100% of rated power. How long will it last this time, and which component fails first?

Reason I'm asking is because modern automotive engines strike me as taking advantage of the fact that full power is used only briefly in a car's typical operation, so they have very impressive specific power figures (over 100hp/liter) - but I can find no data on how durable they are when producing high power for extended periods of time. I'm hoping some automotive engineers can shed light on this question.

P.S. This being my first post, I should probably introduce myself and provide some background. I'm from Malta, have a B.Eng in mechanical engineering and a masters in materials engineering. My current job is package development for a semiconductor assembly plant, main focus being R&D of novel MEMS device assembly processes.

 

[tbuelna]

Some components of a recip auto engine will be life limited by dynamic fatigue load cycles from high speed operation, such as valve springs or conrods. Other components will be life limited by the high combustion pressures and temperatures at high load operation, like pistons or exhaust valves. And other components like fuel injectors may have a service life mostly based on number of cycles, and largely independent of how the engine itself is operated.

 

[MikeHalloran]

1. If the car is sold in Germany, it had better survive the warranty.
2. Similarly for generator or marine propulsion, provided that the installation is properly engineered.

In the specific case of generators, Diesels in particular thrive at sustained power levels of 60 to 90 pct of maximum rating, and develop congestion in various forms at lower power levels.

In the specific case of aircraft propulsion, the maintenance schedules for piston engines are so rigorous that none of the parts have a chance to get old, and even the established manufacturers have no experience with really high-time engines. They don't know how to build one because they've never, ever, been forced to do so.

Emissions- related durability testing has forced evolution of car engines that will last for 100,000 miles, and has forced evolution of design tools and manufacturing techniques that allow rapid development of new engines with similar prescribed durability.

The question you may have meant to ask is, 'What happens after 100,000 miles or equivalent?'.

I don't think anyone really knows. Big ol' 'Merican engines used to cruise well beyond that, given even minimal maintenance, but they can't meet the increasingly crazy emissions regs.


My personal suspicion is that new generations of tiny engines, their power level and effective displacement and internal stresses magnified on demand by turbochargers, will meet the statutory durability requirements. ... and then explode, as every part reaches its carefully calculated fatigue end of life at the same time.

It won't matter which part actually fails first, because collateral damage at high power will take out the remainder of the engine, or there will be no fatigue life left in the surviving remainder. It won't be worth repairing a failed engine, because every part in the junkyard will be within minutes of its end of life.

There won't be a need for junkyards, because everyone will be forced to buy either a brand new engine or a new car when the engine grenades.

<tangent>
Big ol' engine example:
I was not long ago involved in the zero-time rebuild of a pair of Waukesha natural gas engines retrieved from a junkyard (for $25,000 each), which will run at 900 rpm for the next 35 years providing up to 350kWe each, on an unattended oil platform in the GOM. The bare engines weigh 20,000 lbs. each, so they might be a little impractical for a hot rod. They do make a nice chest-thumping bark when you fire 'em up, though.
</tangent>



Mike Halloran
Pembroke Pines, FL, USA

 

[NickJ67]

Interesting question.

A little out of date now but IIRC, Triumph engineers in the 60s were aiming for engine survival of 50 hours at full power.

Against this, 100,000 miles @ 30 mph average would give 3,333 running hours (though dependent on ownership, might never actually see full power use). I also always think it is interesting to compare the running hours at higher average speeds (45 mph average drops the hours to 2,222), which perhaps goes some way to demonstrate why a relatively young, but high mileage car (implying it's done longer runs at higher speeds) is usually a less troublesome buy than an older, low miles car and will run to a higher mileage before failure.

This would also help to explain why big countries (US, Oz, mainland Europe etc) seem to report higher survival mileages than we UK Island Monkeys, where opportunities to sit at high cruising speeds for long periods are more limited.

I realise there are alot of sweeping generalisations here, but my personal experiences in almost 30 years of used car ownership have shown that well maintained high mileage cars remain reliable to very high mileages and it's rarely the engines that fail first.

Another wrinkle is how powerful the engine is in the first place, which obviously goes a long way to dictating how far up its power band it will be operated. I suspect that manufacturers are well aware of this and build their small engines tougher/hp than their big ones.

My first car was a Citroen Dyane with a 602cc flat twin and 29 snorting horses. That was driven with two throttle positions, on or off, and high rpms were needed to maintain progress. It could only manage just over 70 mph flat out so I can safely say that that car did get operated at full power for quite substantial chunks of time. It was admitting to 97k miles when I got it and I added about 5k more. It was still running when sold, but was showing signs of wear!

My current car is a '96 Audi A6 with a 2.5L I5 TDI. It is rated at 140 hp and while it has seen full power use on plenty of occasions, I don't think I've ever managed to hold full power for more than about 90 seconds continuously and then only when overtaking trucks on long hills. It just isn't possible (in UK at least) to hold full power for any longer without running out of road or seriously breaching speed limits. This car is now approaching 256k and engine-wise gives the impression that if I keep doing the scheduled maintenance, it'll double that. The body won't last that long though!

My feeling is that that car engines at least, cold starts, temperature cycling and lack of maintenance kills far more than the actual mechanical wear of hard use. Plenty of engines get scrapped in perfect working order too - I've recycled a few in my time.

Cheers

Nick

 

[BrianPetersen]

I have a Honda CBR125 motorcycle, which is also only operated with two throttle positions (open and closed), necessary for its 12 horsepower to make it go anywhere, and usually pretty close to redline, too. I've seen these with over 60,000 km with only oil changes. (Mine has about 20,000 and runs like new)

But, that engine has much lower maximum revs than, say, a CBR600, even though its cylinder is smaller than those in the 600.

I have heard of a street ridden CBR600 in the southern US with over 260,000 miles on it (> 400,000 km) and it has reportedly never been apart.

I recall one of the UK magazines years ago ran a Honda Fireblade at top speed at MIRA for (I think) 24 hours and then tore down the engine, and found no wear. That engine made about the same power from its 900cc as you are suggesting from a bigger engine and did it with no turbo ... it did it with revs, which is generally tougher on everything.

I'm quite sure that nowadays, engines are designed for continuous operation at maximum power while remaining within the bounds of infinite fatigue life on the materials. Lack of maintenance is what usually kills them.

 

[berkshire]

I asked a very similar question to this in 2006 in
thread108-150335
You can go there to see the answers.
B.E.

You are judged not by what you know, but by what you can do.

 

[MikeHalloran]

Whether it constitutes support for my assertions about grenade engines or not, a search on
{"Ford EcoBoost" problems}
will provide a _lot_ of reading material.
It appears that they fell just a little short of the mark, durability-wise,
and will have to add some iron, or fix some software.







Mike Halloran
Pembroke Pines, FL, USA

 

[LMF5000]

Thank you all for the very useful replies.

Mike, tbuelna - do you have any idea of the longevity of an automotive engine converted to run an experimental aircraft, in terms of hours between overhauls? I'm aware of a lot of air-cooled VW beetle engines powering homebuilt microlights, and I was wondering whether it would be possible to capitalise on modern cheap, lightweight high-power turbocharged engines for an experimental microlight. Aerospace engineers will be well aware of the advantages of turbos at altitude, though with this being an automotive engine I don't know how it would cope with flying at altitude design-wise. And I'm not familiar with maintenance schedules for automotive conversions on experimental aircraft - does the owner/pilot have to come up with a suitable schedule himself, or maybe copy one from an existing similar certified piston engine?

Nick - my first car was a Ford fiesta/fusion. 1.4 liter NA petrol with 80bhp. Used to use full throttle quite often. My current car is an Alfa Romeo MiTo - 1.4 liter turbo petrol with 135bhp and MultiAir (for those not familiar with the term, MultiAir is a hydraulic valve actuation system whereby power control is done by variable intake valve duration [early intake valve closing or late intake valve opening] instead of a throttle body. That way the pumping losses are eliminated and they claim a 10-15% improvement in fuel efficiency and specific power compared to conventional (throttle-bodied) petrol engines). I barely ever have the room to use full throttle for more than 20 seconds at a time, and that's only at 2am when the roads are empty! Same engine with a bigger turbo is offered in the QV model with an output of 177bhp. So curious as to whether such an engine would make a practical experimental aircraft engine or whether it would have durability problems since getting 100hp/l means quite a high BMEP, and you'd have to run it all day at 5000rpm to get all 135 horses...

Brian - with regards to full power operation - so assuming fatigue life is infinite, is the cooling system in the car usually sufficient to permit continuous full throttle operation? Or will the engine slowly start to overheat?

Berkshire - thanks for the link. Seems a reasonable assumption is 2000 hours at full throttle, and Diesels are more suitable than petrols for extended high-power running?

 

[pmrobert]

I know of a gentleman who powered a Cozy Mark IV aircraft with a Mazda rotary. As far as I know, his only semi-catastrophic failures (engine kept running but with no boost) were due to the stock turbo hot side wheel detaching itself from the shaft a couple of times. Apparently those turbos are not very well suited for sustained high load use.

 

[BrianPetersen]

The cooling system in a car these days is designed for continuous full load ... PROVIDED that the car is moving through the air at a speed that is appropriate to the power output! It still needs an appropriate amount of cooling air coming through the radiator.

Aircraft usage demands a whole lot of considerations that are not present in an automotive application. The average aircraft piston engine is air cooled ... it's a whole lot easier to get that to pass FMEA (failure mode effects analysis). They have redundant ignition systems, etc. In my last car, once upon a time, I had a coolant hose spring a leak while on the motorway. While the car came home on a tow truck after that happened, it did in fact come home. If that happens 2 km up in the air while you are crossing mountains or a big lake ... consequently, no comment from me on the advisability or suitability of an automotive engine in an aircraft.

 

[LMF5000]

Brian - Coincidentally, in my previous job doing aircraft maintenance at an FBO, I had the opportunity to work on a Diamond DA42 ("twin star") with twin 170bhp Austro Diesel engines. Water-cooled, turbocharged, 4-cylinder in line - and apparently based (originally) on Mercedes 2-liter Diesels. 85 bhp/l is not bad - I think they're miles ahead of the ancient Lycoming and Continental horizontally opposed engines where a typical one in a Cessna 172 needs 9 liters of displacement for just 170 bhp. The TBO of the Diesels is quite reasonable - 1200 hours or more at this point (Link)

I never understood why aviation laws stipulate dual ignition systems (14 CFR Part 33, Subpart C, § 33.37). It's not like the ignition system is the least reliable part of the engine, at least with modern solid-state systems, in my limited experience. The CFR still permits the use of one fuel injector per cylinder (for engines that use them), one fuel pump, one oil pump, one cooling system, one radiator, one alternator, and one timing belt. So why would they be so hard on the ignition system, and require drilling two holes per cylinder for the spark plugs to make the engine qualify for certification (a requirment that an unmodified automotive engine can't meet), when they still permit single points of failure (like the single oil pump, whose failure would disable the whole engine)?

 

[WGJ]

MikeHalloran - your reference to the Ford Ecoboost engine/s is right on the ball.
Through industry links, I have heard that some aspects of the cooling system have been designed without much thought to manufacturing processes and the rate of degradation of the cooling system internals (casting debris, swarf, products of corrosion), resulting in coolant not necessarily being where you would like it to be.

I had a look at a sectioned display 1.0 Ecoboost - the pistons were reminiscent of racing pistons of not so long ago - very short, almost non-existent skirt. The rods, from what I could see, were nothing special - the I-section was what you'd think typical for a production engine. I gather that nearly all rods these days are laser scribed and snapped across the big end, but I also hear the odd story about b/end bolting causing failures owing to inadequately developed bolt types/torques/torqueing methods.

One has to consider that the Ford Ecoboost is currently at 100bhp/litre and the next iteration will probably be 120bhp/litre. It's not so long ago that these were racing outputs from engines that had little durability capability.

Many years ago, I was involved with the testing of truck engines in the range 254 to 380 cu in and the baseline test was, after a thorough oil consumption bed-in, 55 min at full load, rated speed, followed by 5 min at idle, no load. This was continued for at least 1000 hours and in some cases 1500 hrs.
Of course, there was no thermal cycling going on, as you would get in real use.

Bill

 

[BrianPetersen]

I don't think you can paint all of the "Ecoboost" engines with the same brush. I have heard of issues with the 3.5 V6 Ecoboost in the pickup trucks, but little about the 1.0, 1.6, 2.0 three and four cylinder engines. Most of the issues that I've heard of with the 3.5 EB aren't necessarily mechanical-failure but more to do with real-world issues that didn't turn up during their validation testing like moisture condensing/freezing in the intercooler, etc. (VW is having same trouble with 2009-onward TDI engines and will be using a liquid-cooled, temperature-regulated intercooler in the next generation) Obviously hindsight suggests that this means their validation testing was half-baked and didn't cover enough ground to reveal the faults ...

The lightweight pistons in the 1.0 EB stand to reason. That engine is an inline-three with no counter-rotating balance shaft. Keeping the pistons light keeps vibration in check.

 

[WGJ]

Well, Brian, I'm not painting them all, just recounting conversations I've had with FMC devt engrs here in Europe re the 1.0L engine.

Bill

 

[BrianPetersen]

So it's the 1.0 EB that you are mentioning, when you talk about the cooling system design issues?

The thing about the 1.0 EB that has me concerned, is the timing belt running in oil.

 

[berkshire]

The current favorite engine for conversion from automotive to aviation seems to be the Subaru 4 and 6 cylinder series. I have 2 friends using the 6 cylinder Subaru's in RV7 aircraft, they have 4 years and 2 years of use. The only modification they made was retrofitting larger radiators after experiencing overheating on full throttle climb outs on very hot days.
B.E.

You are judged not by what you know, but by what you can do.

 

[btrueblood]

We used to build stationary products with a load/output similar to a genset, using automotive type engines. We could load them to 80% and expect 2,000 hour + life with mostly successful results. Ford 4-cylinder, Daihatsu 3-cyl., and Nissan 4-cylinder engines I do recall surviving the life test. This was circa. 1990-2000, and the engines were naturally aspirated, not turbocharged.

 

[LMF5000]

Berkshire - can you provide more info on the type of Subaru engine they used? Boxer or inline? Turbo or NA? What dispacement and bhp? Was it derated? I'm assuming this was registered under "experimental" aircraft category to get around the dual-ignition regulation?

 

[berkshire]

LMF5000
I have two friends with RV's using the 6 cylinder boxer layout with an Eggenfellner reduction gearbox. The gearbox is controversial, people either love them, or hate them, and factory support is spotty. I will contact them this Saturday. I am doing a weight and balance for one of them, I will ask the specifics on displacement and horsepower. I have another friend with a Kitfox with a belt drive reduction 4 cylinder inline motor. He has not reported any problems with his, other than that the unit is a bit heavy for his aircraft.
B.E.

You are judged not by what you know, but by what you can do.

 

[LMF5000]

Thanks, I would really appreciate the info if you could get it!

I actually have a friend who built an X-air microlight a few years ago. He spent €10,000 on the aircraft and €30,000 on a two-stroke two-cylinder Rotax (about 40-60bhp I think) to power it. From the little time I spent with it, the engine is not terribly impressive. Quite hard to start (prop spins but engine doesn't fire), though reliable enough once it's actually started. Sadly an automotive power plant wouldn't come anywhere near the light weight of this one due to being 4-stroke and liquid cooled. Probably the best bet would be an air-cooled motorbike engine, but not too convinced one would last long at a constant 10,000rpm...