Durability - what does it mean to you?

We've all seen statements like "the J2Z stock bottom end is good for 700 bhp" or whatever bandied about on this and other forums.

By "good for" I assume everyone means "can take it without breaking". This is obviously a very subjective statement because I doubt anyone has so far been bothered to mod a car to this level and then specifically test it to destruction to find out just how far one could be pushed. Even if they did, how would they test it? What kind of running? Street use? Drag racing? etc. etc. etc.

When you look at all the ways a car can be used, and if you know a about how engines are durability tested in the industry, then a sweeping statement like "good for 700bhp" becomes pretty meaningless (no offense meant to anyone who's ever used it). In industry terms, "durable" means that the engine has passed an exhaustive series of tests specifically designed to mimic (both on the engine dyno and in the vehicle) all the types of use it is likely to encounter in a typical working life. There will be tests designed to target specific systems and compoents (high and low speed valve train, exhaust manifold thermal shock) as well as general durability tests (600 hours alternative beteen full power and full torque for example).

The key thing about all these tests is that at no point (unless you include "abuse" tests like overspeeds) will the engine be tested beyond the statement of requirements for the engine at its given levels of performance. The Supra engine outputs (in stock UK form) 350bhp. It will have been tested to 350bhp and nothing more. The only time that this may happen is if some time in the future it is planned to have a high-output derivative based on the same basic parts. If this is true then the base engine components will be specced so that the high-output version can be made from the same castings, and by swapping in new bearings, etc. As far as we know this was never a plan for the J2Z engine. For this reason I always get twitchy when people say that the basic parts in the GE and GTE variants are the same because you never know if someone decided to save a few Yen by using lower grade bearing shells in the GE.

Another point: Engines get developed through two distinct phases of testing. First comes "testing", during which you expect to break stuff. Every time domething breaks you document it and try to fix it. When you are pretty sure that everything is OK you switch over to "validation". These are the same tests as the first phase, but the expectations are totally different. During validation you do not expect to break anything. if you do then you have to do some fancy footwork to get design changes shoehorned in at the last minute.

Because persistent concerns with an engine design are adressed one by one, there will come a point when there is one last problem to solve. By "solved" I mean it will pass the system-specific or general durability test as prescribed. When this last problem is fixed, development on the engine will STOP.

So the engine goes into the field with a certified level of performance and a signed off level of durability at that level of performance. Most people will be happy with the car and have no problems with normal use, as long as the development team have got a test regime that accurately mimics typical use.

This raises the question of what happens when you start tuning.

My personal (rather extreme) view of what happens to all this testing when you tune an engine in the aftermarket is this: You throw it all in the bin. The minute you extract one more bhp from the engine than it was tested at, you invalidate all the work that was done during its development and therefore you cannot guarantee that it will be durable as the manufacturer defined it. Unless you have access to a lot of test data, or a huge fleet of similarly tuned cars you simply cannot know where the engine's particular foibles lay. Also, if that last difficult-to-fix problem is sensitive to what you are doing, then you may end up breaking engines very quickly.

In case you hadn't guessed already, "durable", to me, means that it will pass all the OEM tests. This partly because I work in the industry and partly because my car is my daily (and only) drive so the idea of it letting me down one day doesn't appeal.

So, agree with that or not as you will (or won't). But having set the stage, it brings me back to the original question: What does durability mean to you? What degree of durability would you be willing to sacrifice for a few extra BHP?

Not intended as a kick off for a flame war. Just interested to hear your thoughts (especially from those with heavily modified cars). It also applies equally to all makes and models.

Discuss

Good stuff

For me durable would mean two or three years without dying. That may sound a lot but then I do about 10,000 miles a year. And on top of that I'm usually cruising for a lot of those - the red line doesn't get much of a look in. So I reckon "good for 600bhp on a stock bottom end" means it won't explode every other time you drive it but it's going to wear bits out a lot faster.

That point about the last problem, that's sort of the weakest link I suppose, such as RX7 rotor tips, Sunny GTi-R gearboxes, MkIII supra head gaskets, Lotus Esprit manifolds (cough)... Nothing immediately springs to mind with the MkIV yet, phew.

Funnily enough when my engine let go years back it was on the road again in a week. One poxy overfuelling problem caused by what appears to be stock parts and it's buggered for three months and counting. So don't think your completely impervious

Anyway, I like having someone around that is, not 'against' as such, but more realistic about modifications because of their job and attitude and requirements of the car. I'll be putting my stock crank pulley back on I can tell you But I'll still be, if I can fix the fricker, strapping on that big single tubro

-Ian

Originally posted by Darren Blake

We've all seen statements like "the J2Z stock bottom end is good for 700 bhp" or whatever bandied about on this and other forums.

 

By "good for" I assume everyone means "can take it without breaking". This is obviously a very subjective statement because I doubt anyone has so far been bothered to mod a car to this level and then specifically test it to destruction to find out just how far one could be pushed. Even if they did, how would they test it? What kind of running? Street use? Drag racing? etc. etc. etc.

 

When you look at all the ways a car can be used, and if you know a about how engines are durability tested in the industry, then a sweeping statement like "good for 700bhp" becomes pretty meaningless (no offense meant to anyone who's ever used it). In industry terms, "durable" means that the engine has passed an exhaustive series of tests specifically designed to mimic (both on the engine dyno and in the vehicle) all the types of use it is likely to encounter in a typical working life. There will be tests designed to target specific systems and compoents (high and low speed valve train, exhaust manifold thermal shock) as well as general durability tests (600 hours alternative beteen full power and full torque for example).

 

The key thing about all these tests is that at no point (unless you include "abuse" tests like overspeeds) will the engine be tested beyond the statement of requirements for the engine at its given levels of performance. The Supra engine outputs (in stock UK form) 350bhp. It will have been tested to 350bhp and nothing more. The only time that this may happen is if some time in the future it is planned to have a high-output derivative based on the same basic parts. If this is true then the base engine components will be specced so that the high-output version can be made from the same castings, and by swapping in new bearings, etc. As far as we know this was never a plan for the J2Z engine. For this reason I always get twitchy when people say that the basic parts in the GE and GTE variants are the same because you never know if someone decided to save a few Yen by using lower grade bearing shells in the GE.

 

Another point: Engines get developed through two distinct phases of testing. First comes "testing", during which you expect to break stuff. Every time domething breaks you document it and try to fix it. When you are pretty sure that everything is OK you switch over to "validation". These are the same tests as the first phase, but the expectations are totally different. During validation you do not expect to break anything. if you do then you have to do some fancy footwork to get design changes shoehorned in at the last minute.

 

Because persistent concerns with an engine design are adressed one by one, there will come a point when there is one last problem to solve. By "solved" I mean it will pass the system-specific or general durability test as prescribed. When this last problem is fixed, development on the engine will STOP.

 

So the engine goes into the field with a certified level of performance and a signed off level of durability at that level of performance. Most people will be happy with the car and have no problems with normal use, as long as the development team have got a test regime that accurately mimics typical use.

 

This raises the question of what happens when you start tuning.

 

My personal (rather extreme) view of what happens to all this testing when you tune an engine in the aftermarket is this: You throw it all in the bin. The minute you extract one more bhp from the engine than it was tested at, you invalidate all the work that was done during its development and therefore you cannot guarantee that it will be durable as the manufacturer defined it. Unless you have access to a lot of test data, or a huge fleet of similarly tuned cars you simply cannot know where the engine's particular foibles lay. Also, if that last difficult-to-fix problem is sensitive to what you are doing, then you may end up breaking engines very quickly.

 

In case you hadn't guessed already, "durable", to me, means that it will pass all the OEM tests. This partly because I work in the industry and partly because my car is my daily (and only) drive so the idea of it letting me down one day doesn't appeal.

 

So, agree with that or not as you will (or won't). But having set the stage, it brings me back to the original question: What does durability mean to you? What degree of durability would you be willing to sacrifice for a few extra BHP?

 

Not intended as a kick off for a flame war. Just interested to hear your thoughts (especially from those with heavily modified cars). It also applies equally to all makes and models.

 

Discuss

If it breaks fix it but make it stronger !!!!

Dude:flame Dev

Darren - really interesting post.

When all the new kit is on I will be pushing the engine way beyond its original design specification. Personally, I would be very happy if the engine lasted for 20K without anything major breaking/melting. This would be (a) driving it most days (b) short bursts of 50-75% acceleration most days © occasionally hammering it, but mostly for fairly short bursts (d) once in a blue moon, hammering it for a more sustained period.

20K should last me 4-5 years.

Well, we will see.

Cliff

Originally posted by Ian C

 

Anyway, I like having someone around that is, not 'against' as such, but more realistic about modifications because of their job and attitude and requirements of the car.

 

-Ian

that'll be me then! For me its everyday transport and i can't really afford problems that modifications 'can' cause.

That often talked about stock 300K mile supra? would it have been so trouble free running 1.1 bar instead of 0.8? difficult to say 100% i spose............

As Ian points out the TT in stock form has no standard 'weak link' and i find the stock TT power is ample TBH so for me its a why tempt fate situation?

I realise we have a lot of members with varying combinations of time, money and experience and that the big mods etc receive (quite rightly) alot of 'airtime' but does this possibly make the average onlooker over confident about their cars 'Durability'

Stock and the Mkiv TT has to be one of the most durable performance cars out there doesn't it?

For me, it's a daily commuting car it has to be 100% reliable (or as good as it gets). All the tuning I've had done is after something has broke but the 1st line has always been reliability.

Having said that I've had a BHG (I think it was gone when I got it though), followed by a bottom end rebuild (dying oil pump) and then turbos' valves and bores all in one go (Mmmmmmmm)

Darren, at what point can we expect the quote button to fail due to Dude pushing it too far?

If anyone goes for a big power upgrade and expects it to last forever like that without extra niggles then they are in for a nasty suprise I reckon. More power = more wear and tear. You've just got to accept that as part of your driving lifestyle. If you do low mileages and you have enough cash knocking around and you don't mind getting your hands dirty then modding is perfectly acceptable. I didn't buy a Supra until I got a job that meant I didn't need a daily commuter

-Ian

Darren, at what point can we expect the quote button to fail due to Dude pushing it too far?

You cant keep a good man down !!!!! i tried to explain to everyonr why i kept 'Tipping off' at bike track days , its simple you need to find the limit and then push it !!!!

Dude:flame Dev

Originally posted by Darren Blake..............Durability - what does it mean to you?

It takes a lickin' ,and keeps on tickin' !

Originally posted by matt

It takes a lickin' ,and keeps on tickin' !

So its an edible watch then ????

Dude:blink:

Durability to me:

Simple, it depend on the vehicle in question.

Motocross bike: 1-2hrs maybe 5 if ridden gently.

A daily driver like the Soarer: It better not go wrong! Once I've had a regular-use car serviced in every aspect, I expect another 60k miles before needing to look at it again, bar the basics.

My TT Supra? Hardly used, weekend toy, when fully modified, a few k between work is fine for me, im a realist.

Darren, I agree with you to some extent, but I think the Supra was tested more than any Lotus, at and above the necassary power levels. It hasn't got some of the strongest internals of a production car by accident you know.

It's only weakpoint is the FMS, which can be fixed.

Its a trade-off - you increase the power of your car and it will not last as long...

I've always heard that the Supra is 'over-engineered', and will comfortably take reasonable power increases as long as everything is in good working order. People say its a de-tuned 500bhp engine, an early model released to a mag in america did a 4.6 0-60.

What are the tolerances likely to be on a modern car? When it is tested, is this at stock bhp? stock bhp + 10% etc. Are the internals be tested above any likely output of the stock engine?

More obvious parts like the fuel pumps and turbos are engineered to handle more power, surely the rest of the engine would follow suit?

Originally posted by Syed Shah

Darren, I agree with you to some extent, but I think the Supra was tested more than any Lotus, at and above the necassary power levels. It hasn't got some of the strongest internals of a production car by accident you know.

Afraid I have to disagree with you, Syed.

In my original post I was mainly referring to the way major OEMs work. The majority of the stuff we do here is consultancy for other manufacturers. In fact out of the 10+ years I've worked here I've only spent 3 of those working on Lotus products. It's no secret that we used to do a lot of work for GM, and we regularly work with pretty much all the other big OEMs too.

The actualy tests differ, but in my experience the general philosophy is very similar. Unless Totyota planned for a higher output version that never got built, I don't believe that it would have ever actually been durability tested at anything above the UK spec power of 350bhp.

Some of the base engine parts will have had a factor of safety applied to them when the stresses were calculated, like crank and con rod fatigue reserve factors. Also, some of the off the shelf parts may be good for higher output applications. It also has a cast iron block which by default may be stronger than strictly required in certain areas purely because the casting process will limit how thin you can make the part. However I firmly believe that the whole base engine design will only have been stressed out for the 350bhp application.

Plus it's not just as simple as how physically strong things are. if you up the power and / or put more heat into the engine then you have to start to consider things like lubrication performance - expecially oil film thickness, oil pressurees at high temperatures, etc.

Thinking about it, a lot of the work we do is from OEMs wanting to increase the power of an existing engine - pretty much what you want to do when you tune. In practise, this is never as simple as swapping out one or two parts for uprated ones. It's usually one or two major part redesigns, then a whole raft of minor (and sometimes seemingly unrelated) parts getting changed as a knock on effect.

Example: a recent power hike on a mass market engine required that the baffling in the oil pan be changed as a direct result of a change to the way oil circulated in the cylinder head.

After you do that its a full testing and validation programme again.

Originally posted by outatime

I've always heard that the Supra is 'over-engineered', and will comfortably take reasonable power increases as long as everything is in good working order. People say its a de-tuned 500bhp engine, an early model released to a mag in america did a 4.6 0-60.

 

What are the tolerances likely to be on a modern car? When it is tested, is this at stock bhp? stock bhp + 10% etc. Are the internals be tested above any likely output of the stock engine?

 

More obvious parts like the fuel pumps and turbos are engineered to handle more power, surely the rest of the engine would follow suit?

As I said above, in my experience you only over engineer if there is a specific reason to do so. If you want a 350bhp engine, you don't set out to design a 500bhp one and then de-tune it unless you want to make the 500bhp one later using the same bits. It just doesn't make engineering or (more importantly to the OEM) financial sense.

Oh, and its normally hard enough to get the base power requirement out of the engine, so we don't engineer in an extra 10% just for durability testing! We may apply factors of safety on specific parts, but this is only to ensure they are good for the base power, not to safeguard the whole engine for aftermarket tuning.

Like I said above, you may end up with some parts that are more capable either by accident or by default, but it will only be some parts.

Having said all of that, the J2Z does seem happy to take a lot of punishment and most of the BPU'd cars on here seem happy enough, but BPU is still a long way away from being "durable" at 500 or 600 bhp.

Did the 2JZ-GTE ever get used for racing? JGTC or something?

This would, to me, indicate the manufacturer had tested its durability for higher power levels? Hence, maybe, how we aer under the impression its a 500hp engine de-tuned due to emissions and gentlemen's agreements etc.

Originally posted by Darren Blake

Afraid I have to disagree with you, Syed.

 

A very good a deatiled post mate. But I still don't agree with you.

A fail to believe that they did not know the exact strengths of all materials used, and the torque that various bits could withstand. The whole point of overengineering a car is testing it to ensure it is more than capable of what it has to do.

Toyota gave us an incredible car, superb brakes, chassis and aero, and choked down the power, you do the rest...

Originally posted by Syed Shah

 

A very good a deatiled post mate. But I still don't agree with you.

 

A fail to believe that they did not know the exact strengths of all materials used, and the torque that various bits could withstand. The whole point of overengineering a car is testing it to ensure it is more than capable of what it has to do.

 

Toyota gave us an incredible car, superb brakes, chassis and aero, and choked down the power, you do the rest...

Of course they will have known the properties of the materials and their limits. You have to to even start to design a part - that's just part of engineering, but it doesn't automatically imply overengineering.

What I'm disputing is that the whole engine is over engineered to the degree that people state.

UNLESS... Alex is right and there were indeed bigger plans for the powerplant.

I'm afraid that the mass market automotive engineering business is very very boring indeed. You seldom get to use sexy materials or processes, and it is generally a struggle to get the engine to make the performance you originally set out for, let alone exceeding it. Budgets are very tight indeed, so you cut your cloth according to what know you need right now or in the not too distant future, not what might be nice to try at some undefined point.

Over engineering even in the design stage usually means huge costs. In my experience, the cost driver is usually a big enough one to remove alot of over engineering. Of course, the components and the system would have been designed to a specification, and that spec would have had performance tolerances. Unless they deliberately designed the engine for something like 500 bhp, I dont believe that they would have then spent the time and the effort testing everything to a lower performance metric. It doesnt make too much financial sense let alone engineering sense!

Durability for me means that my supe gets me through the 12000 miles it gets driven per year without breaking. I have tolerances for things dying on me as its a 10 year old car. I am surprised by the reliability of the car, and have been impressed by the tuning I have carried out.

As we have all seen, once you start tuning you move away from the engineered equilibrium that Toyota envisaged and designed. You remove the cats and have to alter the fuel cut, manage the boost creep, put up with increase fuel consumption etc...

Its the same with any system... it has been designed for a specific purpose but invariably with built in tolerances. Exceeding these will not mean an instanteaneous failure, more that the chance of failure is raised to an unacceptable level (one which would not pass a quantative testing process)

Maybe the goal was quite simply, the Corvette.

The Supra had to beat it in the US market, yes?

The service interval for the corvette is in the region of 100,000miles for a major service. Which is HUGE compared to the 40,000miles the average Euro car has to survive between major services.

So whilst not making the service intervals that astronomical, they quietly made sure they had an engine able to last as long as the Corvette's??

So then you can push this reliability in different ways - adding 300hp will, for example, half its reliability (figure plucked out of air).

So you have a 100,000mile engine only having a 50,000mile life expectancy - but its still better than the euro crap...

Now am I barking up the wrong tree in the wrong field in the wrong country or does that make sense - if slightly off at a tangent.

Ok, I get the main point, but isn't the Jap market slightly different? Jap cars and the Jap market are notoriously the ones to be 'tuned' for more power, and the main Jap manufacturers expect this, going so far as to having their own tuning arms; STI, Ralliart, Nismo and of course TRD. All these companies provide power upgrades to stock cars, but surely they wouldn't compromise their brand if they thought they would be likely to affect the longevity or reliability of their cars?

What I am saying is due to the Jap tuning culture, surely the manufacturers must dial in some extra tolerances to their cars in order to be able to provide safe tuning options for their customers?

Mitsubishi offer an FQ330 for example, which increases stock power by 50bhp, but on standard internals. There is no indication that a buyer should expect the car to fail or be any less reliable than a standard Evo with 280bhp, so surely the stock car must be over-engineered to cope with this?

Interesting post!

Martin.

Originally posted by Martin S

Mitsubishi offer an FQ330 for example, which increases stock power by 50bhp, but on standard internals. There is no indication that a buyer should expect the car to fail or be any less reliable than a standard Evo with 280bhp, so surely the stock car must be over-engineered to cope with this?

Darren makes this point a couple of times - the engine will be tested with the maximum planned power output, which means if there is a high-performance version planned to be release 2 or 3 years down the line it may be that the engine is engineered for it from the word go and therefore 'detuned' in original form. Then again, it may get some uprated components added when the hotter one is released...

-Ian

Originally posted by Martin S

What I am saying is due to the Jap tuning culture, surely the manufacturers must dial in some extra tolerances to their cars in order to be able to provide safe tuning options for their customers?

 

 

Martin.

Exactly my thoughts, GM and Lotus may not do this, but the Japs sure do. Like I said, the Supra didn't get some of the strongest internals in the world by accident.

Top post and welcome to the forum Martin

Darren makes this point a couple of times - the engine will be tested with the maximum planned power output, which means if there is a high-performance version planned to be release 2 or 3 years down the line it may be that the engine is engineered for it from the word go and therefore 'detuned' in original form.

Hmmm, maybe you're right, but maybe not.

The current Evo uses essentially the same core engine as the Evo 4, which was released in 1997. Hotter versions with upped power were only offered much later, starting with the Evo 7 in the FQ series years later.

I'd say that, rather than over-engineering for planned higher power versions, the Jap manufacturers either over-engineer their cars because of their general conscience towards reliability, or in response to the general tuning culture in Japan which demands cars which can tolerate extra power, without sacrificing engine longevity or reliabilty.

I think this is more likely surely, as it is no coincidence that people always stump for Jap cars first when looking for a tuneable car.

Maybe I'm talking crap!?

Hmmm think we've drifted off target.

What is durability to you?

Not who produces the best engine in the world ever, vol.2. or how over-engineered an engine is.

This is mainly a general engine thread not a 2JZ one. And quit bringing up Lotus/GM just cause Darren works there. He's asked a general question - about durability - not GM or Toyota or anyone else!!

Well, if they do then its an interesting take on the market.

Martin: Hi there and welcome.

I'd be interested to know just how identical the core engines on the Evo really is. Particulalry in terms of oil pump, water pump, cooler sizes, bearing shells, etc. if you have the information.

Also, what we call "carry back" is not uncommon either. If the older cylinder block (for example) isn't up to the job, but a new casting can be made that is stronger but will also do the job of the older part, then they insert the newer design into the production line for the less powerful versions of the engine. This way the production volumes are high and the part often turns out to be cheaper because they are making more of them. This way you can end up with a base engine that is capable of a lot more than is required of it, but assuming that all the other systems are equally capable is a dangerous thing to do.

Syed: I can only talk about Lotus and GM because everyone knows we work with them. We work with a lot of other companies too, but I can't go into details because it is not common knowledge.

I still don't think I've contradicted myeslf, though. If the Japanese companies regard tunability as a requirement and design it in, then its still the same philosophy. Given all the things that you could potentially do in the aftermarket I still think its unlikely.

As for the race car possibility that Alex mentioned: Thinking about it more, the durability requirements would be very different in a race. Maybe one or two before blowing up (I don't know the rules on engines in JGTC). Very harsh and maybe good enough to equate to a significantly stronger engine if carried back to normal road use, but you still wouldn't be able to say its good for x-hundred bhp in regular use IMHO.

I mentioned GM/Lotus simply beacuse it is their practices that Darren is basing testing on.

I belive Jap testing to be different.

Lets put it this way: If the Supra was not at all tuner-friendly, would it have been successful? No. It was aimed at the US, where tuneability matters, and that is why it is a sought after car in the US.