[K-Man S]

Cool tear down pics.

So if your theory is correct, and what you are observing is valve guide wear in low-mileage engines, I wonder why the problem seems to get better over time, and not worse? Also, the engines I've seen smoke the worst are Gen 1 engines, not Gen 2 engines... so does this mean Porsche may have a general, broad-based valve guide wear problem? It is difficult to believe.

I wonder how a new engine would respond to your screwdriver test? I wonder if on new engines all of the valves would require equal pressure for displacement? Without the ability to measure it is difficult to know exactly what the true situation is.

I'm still not convinced. If what you say is true, why did our 9A1 base engine smoke less over time?

Yes you hit the nail on the head. So a screwdriver was able to bend a valve.

1) How much force was required to bend each valve?
(This was never measured or provided nor was there any standard way of mounting the screwdriver as a lever arm, even a slightly different position could have dramatically different force readings if the force was ever measured)
2) How much force "should" be required to move a perfectly good valve with a perfectly good guide?
3) Assuming that some sort of rig could be built to accurately measure the force required to deflect a valve, what would the results be for a 3.2L engine? any other Porsche engine? New vs. Used?


The problem I see here is that a so-called "test" was done but it wasn't done in any scientifically meaningful way but rather based on some backyard mechanics way of "feeling" the play in a valve. Well is that weak metal, a bad guide, or simply the placement of the lever arm? Since nothing was measured and nothing was standardized there's no way to know. Conversely I didn't see any evidence of oil deposits building up on the exhaust valves as if it had been running by the guides.

While I think it is always interesting to see the tear down pictures and video, unfortunately I don't think you can make valid conclusions about the valve guide conditions from what was done. I'm very surprised that Porsche hasn't published the valve guide clearances/allowed tolerances because Porsche dealership mechanics have to rebuild these engines and in order to do so properly they'd need these specs. The next time I'm at my dealer I will ask if they have these specs, and if they don't, there is a dealer service advisory council in the US that unanswered questions can be taken to and I'll ask that they provide the valve guide clearance specs. Once you have those it should be an easy job of measuring with the proper tools (read not a screwdriver) whether or not the clearances are within specs.

 

[86atc250r]

The secret behind startup blue smoke is simple.

These are "flat" engines - otherwise known as horizontally opposed or boxer.

Think about how the cylinders sit compared to a "normal" V or inline engine. They lay on their sides, in a horizontal manner.

Now bear in mind that piston rings are a metal to metal contact seal that does not seal absolutely perfectly - and bear in mind that there is also a ring gap where there is a direct, obstruction free path between the top and bottom sides of the piston.

With that all in mind, where will oil drain in a "normal" V or inline engine who's pistons & cylinders sit mostly vertically? It will drain down back into the crankcase - it will not siphon up on top the piston.

On the horizontal engines, it is entirely feasible that some left over oil will seep past the rings as it pools from draining off the back sides of the pistons and the upper portions of the cylinder walls to the lower portion of the cylinder. most will drain back toward the crankcase, but some will also seep past the rings.

This will manifest itself in a puff of blue smoke upon startup.

My brand new CS blew a fairly nasty puff of blue when the salesman fired it up after sitting several days without being started - but does not do it noticeably upon startup with day to day driving and should be less likely to do it as it breaks in.

There's no mystery - the reasoning is very simple - oil can and does seep past the rings with time. Same as if you poured oil on top of the piston of a "V" or inline engine - some would seep past the rings and drain down into the crankcase, it would not pool on top of the piston & sit there forever.

Ever seen a radial aircraft engine startup? It's an extreme version of the same "problem". The blue smoke they blow out on start is due to the same effect - in their case some of the pistons are upside down and a small amount of oil drains past their rings too & gets burned off upon initial startup.

And -- to me, it looked like the valves were bending/flexing against the screwdriver pressure in those videos.

 

[CaymanPower]

So if your theory is correct, and what you are observing is valve guide wear in low-mileage engines, I wonder why the problem seems to get better over time, and not worse?

I'm still not convinced. If what you say is true, why did our 9A1 base engine smoke less over time?

I cannot confirm that information. What i observed so far was that when the weather temperatures drop the smoke is less frequent. I could bet with you that when the spring and summer arrive here in Portugal (it's winter now) the smoke frequency will increase. I'm just waiting for that.

It's a temperature related problem... more of that later.

 

[CaymanPower]

1) How much force was required to bend each valve?

The question is that there shouldn't be any force differences. And, even then when greater force was applied to cyl. #2 right side valve it barely moved. That's what is supposed to happen.

Conversely I didn't see any evidence of oil deposits building up on the exhaust valves as if it had been running by the guides.

Wait until the weather temperature gets hot enough and the smoke events start to be more frequent.

I'm very surprised that Porsche hasn't published the valve guide clearances/allowed tolerances because Porsche dealership mechanics have to rebuild these engines and in order to do so properly they'd need these specs. The next time I'm at my dealer I will ask if they have these specs, and if they don't, there is a dealer service advisory council in the US that unanswered questions can be taken to and I'll ask that they provide the valve guide clearance specs. Once you have those it should be an easy job of measuring with the proper tools (read not a screwdriver) whether or not the clearances are within specs.

When a problem exists with the guides Porsche orders the head replacement (just like Ford did with the TSB).

Please, do that. I will be most grateful!

 

[CaymanPower]

Before I get into why do the gen2 base Cayman engines have premature worn exhaust valve guides, let's see first how oil is pulled through the exhaust guides. In fact, oil is pulled by venturi effect of the exhaust gas flow past the exhaust guide and that's the reason why, unless the guides are really worn, there's little oil consumption but enough to cause a cloud of smoke (only a few drops are needed) when this oil heats up (not burns) in the hot exhaust. You'll see in the next picture that in a flat engine it's more likely that the oil which is being pulled through the guides enters the exhaust system before it can reach the combustion chamber which is under pressure from the combustion exhaust gases:

Also, see link about Smoke generators. And I quote:

Very large or sustained smoke screens are produced by a smoke generator. This machine heats a volatile material (typically oil or an oil based mixture) to evaporate it, then mixes the vapor with cool external air at a controlled rate so it condenses to a mist with a controlled droplet size. Cruder designs simply boiled waste oil over a heater, while more sophisticated ones sprayed a specially formulated oily composition ("fog oil") through nozzles onto a heated plate. Choice of a suitable oil, and careful control of cooling rate, can produce droplet sizes close to the ideal size for Mie scattering of visible light. This produces a very effective obscuration per weight of material used. This screen can then be sustained as long as the generator is supplied with oil, and—especially if a number of generators are used—the screen can build up to a considerable size. One 50 gallon drum of fog oil can obscure 60 miles (97 km) of land in 15 minutes.

 

[blueone]

oil is pulled by venturi effect of the exhaust gas flow past the exhaust guide and that's the reason why, unless the guides are really worn, there's little oil consumption but enough to cause a cloud of smoke (only a few drops are needed) when this oil heats up (not burns) in the hot exhaust.

On initial start-up, when most of us experience smoke, the exhaust is cold. The only parts of the car that will be hot are the combustion chamber and the exhaust valves, so this explanation doesn't seem plausible.

 

[CaymanPower]

On initial start-up, when most of us experience smoke, the exhaust is cold. The only parts of the car that will be hot are the combustion chamber and the exhaust valves, so this explanation doesn't seem plausible.

No. The exhaust gases are quite hot, hot enough to heat up the oil (a few droplets) and evaporate it but not to burn it. See how a smoke generator works.

 

[CaymanPower]

Think about how the cylinders sit compared to a "normal" V or inline engine. They lay on their sides, in a horizontal manner.

Now bear in mind that piston rings are a metal to metal contact seal that does not seal absolutely perfectly - and bear in mind that there is also a ring gap where there is a direct, obstruction free path between the top and bottom sides of the piston.

On the horizontal engines, it is entirely feasible that some left over oil will seep past the rings as it pools from draining off the back sides of the pistons and the upper portions of the cylinder walls to the lower portion of the cylinder. most will drain back toward the crankcase, but some will also seep past the rings.

Man, look how the engine is moved around from side to side, back to forth, up and down. How can the piston rings seal then?! Alot of oil must be burned here :

 

[K-Man S]

The venturi model doesn't make any sense for initial smoke start up. It only works when the engine is running, meaning gases are going out the exhaust creating the venturi effect. If that were the case then once the engine were shut off, no more oil would leak past the guides, in fact your picture clearly shows any oil between the valve and the guide would drain back down into the head and ultimately back down into the pan. Only after oil pressure comes up and oil gets to the head could oil possibly be sucked up into the valve guide and then through venturi properties into the exhaust gas flow. This should mean that no oil puff would occur on initial startup, but that after the car ran for a few seconds a slow trickle of oil would get past the valve guides and get sucked into the exhaust gases and this would occur the entire time the engine was running. Which should mean continual smoke and continuous oil consumption the entire time the car was driven. (much like my car uses oil all the time). Most people reporting smoke are reporting only a puff at initial start up and then no smoke after that and no oil consumption.

Which goes back to a small amount of oil creeping past the cylinder rings in a horizontally opposed engine and is a 'normal' operating condition of said engines.

And in that video oil is splashing around inside the motor, but it isn't "creeping" by the rings while the engine is sitting still. When the engine is running there is an oil seal in place around the oil ring that prevents oil from simply running by the oil ring and into the cylinder combustion chamber in mass quantities. Two different scenarios. BTW the newer engines have the oil return pumps to get oil out of the heads and back down into the block, which means less oil to ever pool in the heads and get by those valve guides you want to keep blaming.

 

[blueone]

No. The exhaust gases are quite hot, hot enough to heat up a the oil (a few droplets) and evaporate it but not to burn it. See how a smoke generator works.

I see, so you're saying the oil is evaporated in the hot exhaust *gas*. I read the article about the smoke generator.

Let's for a moment assume you're correct about almost everything you've said. Let's assume some of the valve guides have more clearance than they should, and that a few droplets of oil are pulled into the mixture, but such a small amount that even over 5000 miles you can't measure the oil loss. (Like on our cars.) That might also explain some soot on the tail pipe I find annoying. Still, I can't help but think "So what?" Some large fraction of new engines use a lot more oil than this on a regular basis.

The Ford V8s cited in some of the links you referenced often burn a quart every couple of thousand miles or so, yet they are highly sought-after by the US limousine industry because they frequently go 300-500K miles before needing a rebuild. Similar stories can be told about so many modern engines, that are known to use oil more than either of our Porsches, yet have legendary lifetimes. (Another that comes to mind is the Cadillac Northstar V8. Some oil use, but seem to last hundreds of thousands of miles.) No one likes smoke on start-up, but I still haven't seen any evidence from you that this is something to get concerned about.

Also, none of what you have described explains why both of our cars, with two completely different Porsche engines, smoked on start-up when brand new, yet later, after accumulating several thousand miles, nearly stopped smoking altogether.

For someone that would rather drive than post, you sure are putting a lot of energy into this investigation.

 

[CaymanPower]

Ken,

When the cats are hot enough, which happens shortly after initial start-up, the oil droplets are mostly 'burned' there, that's the reason why you don't see smoke anymore. Also, oil pressure rises almost instantaneously upon start-up to assure proper lubrication of the metal moving parts otherwise they start to wear out (which they do at some extent). In fact, you need oil in the heads for the Variocam Plus and the hydraulic valve lifters to operate correctly from the get-go. And, since the gen2 Caymans have an electronic demand-controlled oil pump which sets the required oil pressure and defined volume for each engine operating state i believe there is an high oil supply demand to this pump at start-up. I still believe that's what the DME reflash (per Porsche's TSB) is for - to reduce the (initially programmed) pump demand at start-up in in an attempt to cope with the smoke upon start-up issue. Too much oil pressure for worn guides.

Please, give us the Porsche specs. for the valve stem to guide clearances.

BTW the newer engines have the oil return pumps to get oil out of the heads and back down into the block, which means less oil to ever pool in the heads and get by those valve guides you want to keep blaming.

Now we are talking serious business... that means less oil is available to the guides. It is the oil flow to the guides that creates a second heat transfer path from the valve and guide and thus augments the cooling of these components. With sufficient oil flow across the hot valve stem and upper portion of the guide, both of which are exposed in the head, heat is readily transferred from these components to the oil at the same time that it is passing through the cylinder head to be transferred to the coolant flowing through the galleries and, ultimately, to the air across the radiators. It is this additional cooling provided by the oil flow that is of paramount importance. That second heat transfer path mentioned above is successful in providing augmented valve and guide cooling that extends the longevity of those components which receive it.

 

[CaymanPower]

For someone that would rather drive than post, you sure are putting a lot of energy into this investigation.

Yeah! I like to know the truth above everything else.

 

[K-Man S]

No I said less oil to pool (and cause oil smoking problems particularly on the track), I didn't say oil starvation to the heads. You just jumped from too much oil to too little oil without any facts.

 

[CaymanPower]

Just give us the Porsche specs. for the valve stem to guide clearances, please!

 

[K-Man S]

Just give us the Porsche specs. for the valve stem to guide clearances, please!

When I get to the dealer next I will...

 

[86atc250r]

Further....

In a "normal" V or Inline engine, valve seals are typically associated with smoke on startup.

Why? Because of our friend gravity. Where are the valves located in relationship to the piston of a V or Inline engine? If seals/guides are leaking, where does that oil go? Down on top of the piston unless the valve happens to be completely closed.

Now, with this in mind, note the orientation of the cylinder head on the Cayman engine in it's normal configuration. Exhaust exits out the bottom. With this configuration, the head of the valve is actually higher in the engine than the end of the stem (where the oil is normally kept) - meaning if the exhaust guide was leaky in the traditional sense, oil would have to "leak" uphill in order to reach the exhaust port. Further, it would then have to follow the valve stem uphill again to enter the chamber instead of simply draining out the exhaust port.

Further yet, if not being burn directly in the combustion chamber and instead being burned off by hot exhaust gases - the blue smoke would be delayed as it takes time for hot exhaust gasses to heat everything up enough to burn off wayward oil in the exhaust path - so the smoke would be slow to start and slow to stop.

Seems a tad unlikely, no? If the guide was so leaky that oil was being drawn out during running conditions, the engine would not stop smoking shortly after startup.

So - all that said, if you're chasing valve guides as your problem, you probably need to be looking more toward the intake valves...

 

[86atc250r]

Sarcastic emoticons aside...

While you're hunting youtube videos, search for "Radial engine startup" and feel free to watch any video available. Should show you that if given the conditions & opportunity, oil will, in fact, seep past the rings of an engine allowed to sit, not running for a period of time. Surprisingly enough, these engines also seal up when running and the oil smoke quickly dissipates.

Man, look how the engine is moved around from side to side, back to forth, up and down. How can the piston rings seal then?! Alot of oil must be burned here :

 

[CaymanPower]

Sarcastic emoticons aside...

Please, just look at the spark plugs. They are clean as a 'whistle', just normal decoloration. Top of the pistons only has the normal combustion residue also. Compression tested good across all cylinders. I don't have excessive oil consumption. I'm not burning oil in the combustion chamber otherwise that would show up on the spark plugs after ~10k miles and a lot of smoke 'clouds' events upon cold startup. Forget it man!

As for the 'smoke delayed' argument how delayed is this so called (in Porsche foruns) 'Porsche classical startup exhaust puff' for you?! Not a youtube vid mind you:
'Delayed startup exhaust puff'

 

[CaymanPower]

BTW the newer engines have the oil return pumps to get oil out of the heads and back down into the block, which means less oil to ever pool in the heads and get by those valve guides you want to keep blaming.

Now we are talking serious business... that means less oil is available to the guides. It is the oil flow to the guides that creates a second heat transfer path from the valve and guide and thus augments the cooling of these components. With sufficient oil flow across the hot valve stem and upper portion of the guide, both of which are exposed in the head, heat is readily transferred from these components to the oil at the same time that it is passing through the cylinder head to be transferred to the coolant flowing through the galleries and, ultimately, to the air across the radiators. It is this additional cooling provided by the oil flow that is of paramount importance. That second heat transfer path mentioned above is successful in providing augmented valve and guide cooling that extends the longevity of those components which receive it.

The potential cooling surface area of a cylinder is many times greater than the same potential cooling surface area of the heads. They are more difficult to cool. Most of the area in contact with the burning fuel is valve heads and the spark plug and nearby are the inlet and exhaust ports. There is relatively little room in the heads to flow coolant near aluminium that is directly being heated up by combustion. That's the reason why the valves and the bronze guides so greatly rely on the oil supply not only to lubricate but ALSO to cool them. To understand the extent of the problem look at the following picture where you can see that Porsche engineers had the preoccupation of providing a tiny coolant passage (encircled in red - see picture) ONLY to cyl. #1 exhaust valves side (same is valid for cyl. #3), where there is none for the intake valves side because temperature is, for obvious reasons, much higher in the exhaust side area:

Now, gen2 base Caymans reportedly have had issues, at least that has been my experience with two base Caymans, with the cooling system not being totally free of, what seems to be, air pockets from the factory fill up. So much so, that the coolant will drop below MIN level during the first few thousand miles if care isn't taken. Low coolant level is obviously dangerous but so are air pockets in the engine cooling path because air pockets tend to build up precisely in the tiny coolant passages where the flow rate tend to be less and the temperatures tend to be higher. The same thing to the low coolant level. Localized steam pockets will form where the coolant flow rates are less and the temperatures are higher: the tiny passages in the heads that cool the exhaust valves and guides. See this link still about the above mentioned Ford engines, and I quote (post #1877):
Lincoln Aviator - Car Forums - Edmunds

"From the factory, the 2003-2004 “Terminator” Mustang SVT Cobra was an absolute work of art by Ford with the exception of one issue. The 4.6L DOHC heads used in the Terminators (and other vehicles as well) have a cooling defect that was later recognized and corrected by Ford several years after its production with a revised head. The coolant does not have an escape route out of the rear of the head which aids the development of steam pockets forming in the rear cylinders of the head. This specifically takes place in the #7 and #8 cylinders. Under normal operating temps on an unmolested cobra, the heads will suffice however when modifications are added it becomes more and more likely that these “hot spots” will develop due to inadequate cooling and as a result, the motor will develop a “TICKING” noise. The revised head from Ford is extremely expensive and was not released until most warranties were already expired.

Then, you can see that when the weather temperatures are higher or/and the driving style is aggressive the engine cooling challenge becomes more of an issue. Also, with higher outside temps the oil tend to be thinner, to have lower viscosity, therefore it can be more easily pulled by the exhaust gas flow past the guides as this creates the so called venturi effect. Oil is NOT constantly being pulled by this mechanism. See link, and I quote:
Valve Guide Repairs

WORN VALVE GUIDES
When the guides are worn or there is too much clearance between the guide and valve stem, the engine will use oil. This applies to both intake and exhaust guides. Though oil consumption can be more of a problem on the intake side because of constant exposure to engine vacuum, oil can also be pulled down the exhaust guides by suction in the exhaust port. The flow of exhaust past the exhaust guide creates a venturi effect that can pull oil down the guide.

Finally, high weather temperatures also mean that after engine shutdown the exhaust system doesn't cool so much before the next startup after a few hours of resting. So, the good old 'smoke generator' is ready to heat up those oil droplets (if present) upon startup and give you a magnificent 'smoke screen'.

 

[blueone]

Low coolant level is obviously dangerous but so are air pockets in the engine cooling path because air pockets tend to build up precisely in the tiny coolant passages where the flow rate tend to be less and the temperatures tend to be higher. The same thing to the low coolant level. Localized steam pockets will form where the coolant flow rates are less and the temperatures are higher: the tiny passages in the heads that cool the exhaust valves and guides.

How do you know this is true? Air pockets build up at high spots; that's why, for example, the bleed valves for the brake calipers are at the top of the calipers. Also, the MIN coolant level doesn't necessarily have anything to do with this, since you can be well below the MIN indicator, and still have a "full" cooling system.

CaymanPower, I still get the impression that you are doing nothing but speculating in your posts, but now you're speculating with some very nice photographs and the implied authority of an engine dis-assembly process. But in the end, everything you're saying is still idle speculation, especially in your comparisons to Ford V8s, which only had a problem because of a casting issue, that was fixed.