Okay, this is my first post. It's a bit of an epic but it might be of help to some of you out there....
First off, I don't own an Audi. I drive a little 1.0L Kia Picanto. Second, I'm the world's worst mechanic. I haven't wielded a spanner in anger since 1978! I was also once told that I would be 'no fun on a first date'. All personally very tragic and I'm definitely not the usual person who posts on Audi-Sport.net. However I have one redeeming feature; I know a lot about engine oil....
For a while now, I've been reading about the horrendous oil consumption problems on some VAG's TFSI engines. I've read about worn piston rings, potentially unevenly worn rings due to con rod distortion, pistons being too small, badly designed Pressure Control Valves and dodgy Oil Separators. All of this is might well be ccorrect. If what I've read is true, Audi are now putting their hands up and finally fixing the offending, oil guzzling engines free, gratis and for nothing. Everyone's happy...
However I couldn't help but notice that amongst all the mechanical chatter, one thing I haven't seen is anyone properly questioning whether the oil going into these engines is actually fit for purpose. Yes I know the Audi oil is API this and ACEA approved that and meets the requirements of VW (insert latest incarnation of their bonkers specification here). However one might possibly argue that any oil that exits the crankcase and throws itself at such a rapid rate of knots down the exhaust pipe, might not be the one you should be using! Maybe, just maybe, the officially recommended oil for these engines is actually the wrong oil and another oil (one that actually stays in the crankcase) might be more useful to owners of these problems cars.
Now I have absolutely no inside track on this. I've not formulated any of these oils and I'm not privy to any of the studies that Audi, VAG, Shell, Castrol, Lubrizol and many others will have undoubtedly done to investigate this problem. But I do have a theory as to what's happening and how you might be able to solve the problem. However you need to understand a bit about oil formulation first....
All things being equal, a thin oil will be more volatile than a thick oil. As you progressively move from 20W40 to 15W40 to 10W40 to 5W40, oil gets more volatile because you need more light base oil (and less heavy base oil) to get the improved cold flow performance. There's a second order impact in that the oil progressively needs more polymeric Viscosity Index Improver (VII) to achieve the viscometric balance. The rubber in VII has poorer cold flow characteristics than base oil so you need to further shift the light-to-heavy base oil mix to accommodate the higher level of VII. This latter effect means that, all things being equal, as you move say from a 5W20 to 5W30 to 5W40 to 5W50, the oil will become more volatile (yes, you did read that right, sometimes a thicker oil can make things worse).
Now in the oil formulation game, not everything is always equal because you employ different base oils (the stuff that constitutes 90%+ of your engine oil) to deal with the inherent problem of oil volatility. There are four main classes of base oil. Group I (good old fashioned solvent extracted mineral oils), Group II (a sort of mild hydro-treated version of Group I), Group III (heavily hydro-cracked mineral oil) and Group IV (typically 'proper' synthetic Poly Alpha Olefins or PAOs). Typically Group II is better than Group I, Group III better than Group II and Group IV is best of all. As one might expect, Group Is are dirt cheap whereas PAOs are inordinately expensive. From an oil volatility point of view, a Group IV 5W30 will be less volatile that a Group III 5W30 which will be less volatile than a 5W30 made from Group II. (Group I 5W30's would be very bad on volatility but are almost impossible to formulate).
OEMs tend to like 5W30's. They are good for fuel economy and emissions. Most 5W30s are made from Group III base oil. This allows the oil to meet the volatility specification (typically a Noack of < 13%), have good oxidation stability at reasonably low cost. BUT...they are still volatile and in the right circumstances on a problem engine can worsen oil loss.
So TFSI engine oil loss...what's happening? Well to state the obvious, if your engine's losing lots of oil, and there's no oil puddle of your drive, the oil is exiting via the exhaust pipe. Oh and I say it here only because I haven't read it elsewhere, if you've had his problem, your catalytic converter is probably shot to Hell; poisoned by phosphorus and sulphur from burned ZDDP. The big question is HOW is it getting into the exhaust? As I see it there are three routes...
First is loss via worn valve seals. This puts oil directly into the combustion chamber. I sort of discount this because positive valve seal design has come on a long way and if this was the primary cause of the problem, Audi would have twigged it and simply offered to replace them. Also I'd expect blue smoke to be more in evidence than it is.
Second is direct transport of accumulated oil on the first piston land into the combustion chamber. Think about it. The piston goes up to TDC and drags an oil film with it. The piston must descend but the inertia in the oil wants it to carry on moving upwards ie into the cylinder into the burning combustion gases. There's a horribly complex MIT SAE paper that says at low engine load and low speed, this can be the dominant route for oil loss. This could be happening with the TFSIs but I sort of discount it too. You don't buy an Audi to pootle, you buy a Kia Picanto!
The third, and I think the most likely route for oil loss, involves several interacting factors but it must start with worn piston rings. This might be caused by several things. It could be bad ring metallurgy but I think it's significant that the TFSI is a direct injection engine. DI engines burn atomised, rather than fully vaporised fuel and they tend to produce a tiny amount of soot. Hot soot can be abrasive so once it starts getting past the rings, it could feed the ring wear problem. Another feature of DI engines is that the inlet valves don't get 'washed' by fuel like they do in an normal, MPI engine. Ring wear could conceivably be initiated by a lump of coke dislodging from a dirty inlet valve and dropping in to the cylinder.
Now all engines generate a degree of blow-by. Hot, high pressure combustion gas rushes through the ring gaps on the compression stroke. This gas can shear, atomise & vaporise the oil in the ring pack. However if you start to wear the rings, the rate of blow-by increases and blow-by is a major source of oil loss. The oil laden blow-by gas mixes with oil in the sump and flushing air. The now super laden gas goes through the Oil Separator & PCV, into the intake system. Once in the intake system, it gets routed to the cylinders to be burnt and flushed out via the exhaust. Now the oil separator, like any cyclone, will only separate liquid from gas. It will not separate fully vaporised oil from the blow-by gas. Also all cyclones get less efficient as gas flow increases so if ring wear causes high rates of blow-by, not only vaporised oil with pass through the Oil Separator, so will some oil in liquid droplet form. And this is where things get tricky. Oil, especially oil in droplet form, doesn't burn very cleanly. Burnt oil will generate soot particles. Soot particles can be abrasive and so the whole process starts to spiral downwards.
But hang on, Diesel engines are all DI so why is the TFSI a problem? I think it's because it uses petrol, and volatile petrol is more damaging to oil loss than relatively heavy diesel. Think about it. You inject liquid petrol directly into a cylinder. Let's say the fuel is cold and the cylinder is cold. Yes the cylinder fires but you inevitably push some petrol (or partially burned petrol) into the sump with the blow-by or if the fuel hits the scraped bore. Let's say the engine oil is also cold, chances are the petrol will condense out in the oil. This is normal. My old car would typically have about 8% fuel in oil. Drive for a while and everything starts to heat up, the relatively light petrol starts to vaporise into the blow-by gas/flushing air mix, and get routed through the PCV into the intake system and burned. Petrol burns cleanly so what's the problem? The problem is something called vapour pressure and the way it causes hydrocarbons to interact. I used to work on an oil refinery. Heavy oil residues can be difficult to distil. As often as not, you fix the problem by distilling heavy residues under vacuum. However, in theory, you could mix the heavy residue with something very light like naphtha (the raw material from which petrol is made) and re-distil it. Although the residue and naphtha are poles apart, on re-distilling, he two don't 'break' cleanly. The laws of vapour-liquid equilibrium cause the light naphtha to 'pull' some of the heavy residue with it. Why is this relevant? Well although I can't be sure, I suspect the TFSI engines are dumping petrol into the sump which in turn is evaporating off and 'pulling' with it some of the lighter base oil in engine oil. As the oil would be in vapour form, it would pass right through the Oil Separator, into the intake and onto be burned. Just like before, the oil would burn badly, create soot which would create yet more ring wear.
So what can you do to prevent any of this? Well to me, the obvious thing to do first would be to see if a less volatile engine oil could help. I know these days it's heresy to say it, but no-one actually needs a 5W or 0W oil from a cold start point of view. You could easily make the case that even 10W oils are an unnecessary luxury. Maybe if we ever enter another ice age, I might retract what I just said but not until! Just remember back in the 1960's everyone was tazzing around in their Mini's all of which had 20W50 in the sump. The roads didn't empty every time he weather turned cold. Cars started and life went on!
Ever wondered why full synthetic oils are either 5W or 0W? The reason is because it's actually quite difficult to make a full synthetic 15W or 20W because Group III & Group IV base oils tend to be only available in low viscosity flavours (typically 4 cst and 6 cst varieties). Mineral oils on he other hand are readily available in much heavier weights (like 500N and Bright Stock). If you want a low volatility oil, you need a lot of heavy base oil in the blend. An oil with a lot of 'heavy' in it doesn't necessarily have to be a bad, old fashioned 20W50 all mineral. I would have thought it quite possible to make an ultra low Noack 10W30 from PAO 10 and/or Group II 500N. All it needs is someone with the will to do it. The fact is, given the way things work with these 'bad' TFSI engines, all you would need is one 600 mile oil consumption test to see if the oil made a difference or not. My gut feel it it would cure the problem in a heart beat. I haven't looked but I suspect you can't go out and buy such an oil. Someone would need to formulate it and blend it. That person isn't me sadly (now retired, no access to stuff any more). You could ask one of the big players but I doubt it they would be much help. In my experience, they are slow and to be honest, they can be a bit dim. You could try ringing someone at one of the smaller oil blenders (Millers or Comma maybe). They would probably understand what the problem is and knock up a batch of oil for you to try out.
Oh and a couple more things...
If you move to a heavier oil grade, expect a bit less fuel economy but I suspect what you lose of fuel, you will save many times over in he cost of engine oil changes! Also, I really like heavy base oils. In my view they do a much better job of sealing the ring pack. They also are great for preventing wear.
That's all folks!
