simsgw

But Tim, has anyone admitted that who knows whereof they speak?

Gary

gota911

Gary, no, they have not. I was very overt in stating "in my opinion" I believe it is possible for a PDK car to have over revs. In another thread I have offered to pay for a DME read out at a Porsche dealer for anyone who has bounced off the rev limiter a few times in a PDK car.

My conjecture is based on the logic that if a manual equipped car can and does have "acceleration over revs" so can a PDK equipped car. That conjecture could be wrong, but until it is proved wrong (or right), it will remain my conjecture.

utkinpol

only matters to warranty implications.
but i would say, many overrevs in range 4+ are a good sign that car was mistreated. why to pay for such car if there are many more that were cared for properly?

sameerg

Yes, Seriously, I have never even been near 6k. 5 K tops. This is the first car i have ever driven with so much power and am scared to go very quickly. I am still learning the car and its behavioral characteristics. I know you guys are gonna make fun of me on this, but, it is what it is.

Having said that, I bought my car used - CPO, and never asked for a DME report (Ignorance is bliss), So, possibly there may be over revs from the previous owners or the techs.

So hypothetically speaking, my car would not have any over revs. I like to keep that in my mind that way. LOL.

steph280

Has there been any actual cases of PCNA denying warranty work because of DME over-rev reports?

Macster

50% strength safety factor? Maybe. Maybe not. There appears to be considerable margin in all engine components or engines would be blowing up all over the place.

There are exceptions in which a rod or rod bolt or something in that general area fails but more often than not the weak link is in the valve train.

Now in the case of our Porsche engines even the valve train appears to have considerable margin for while many drivers hit the red line very few engines blow up or suffer less catastrophic (yet still expensive) problems from overreving.

A large number of overrevs even if they are in the lowest range and even if they occurred long enough in the past that one could breathe a sigh of relief, still suggests the engine and car received the kind of usage that I would prefer my candidate used cars did not receive.

Sincerely,

Macster.

Macster

While I know of none, I was told that when a car is brought in for engine problems the overrevs counters (among other things and the other things can be numerous) are pulled and analyzed by Porsche. Thus the potential is there to have a claim denied if the readings indicate the engine experienced a "money shift" rather than the always blamed IMSB failed.

(As an aside: Porsche techs tells me then under some circumstances when a car is brought in -- cars under new or CPO warranty -- a considerable amount of data is read from the car's DME and forwarded to Porsche. When I asked this be done for my out of warranty Boxster and a printout given to me I was told the report can run pages, 20 was the number mentioned. I received a portion of a couple of pages. I could have had the whole shebang but it would have cost me some diagnostics computer time and the labor of two techs, one to work the throttle (so readouts of the engine's oil pressure/temp could be made at various RPM levels). I declined to spend the money being satisfied with the fragment of the report I did get. Oh, almost forgot: In some cases -- if the engine has been replaced under warranty IIRC -- an oil sample is obtained from the vehicle and sent to specific lab for analysis.)

Anyhow, it is Porsche (PCNA or the factory) that decides if a warranty claim for say an engine is denied based on overrevs, the number, their severity. The dealer may be able to argue in favor of the customer under some circumstances.)

Sincerely,

Macster.

AYHSMB

Just found this one with a few seconds of googling :

http://www.6speedonline.com/forums/9...-warranty.html

I know I've seen others in the past, mainly in the 996 forum.

One common case is for cars just barely out of warranty that have their engine blow up ; in this case the question is whether you get the subsidized replacement cost ($5k-10k) vs the full cost ($20k or so) ; DME over-rev report is used as part of PCNA's decision of whether to give you the "good will".

jakes dad

[QUOTE=SharpMan;8930240]It won't let you.



Can it be over revved during downshifting? I lost my concentration and downshifted instead of upshifting. Don't know how high the tach went but it sure didn't sound good.

simsgw

The trouble is the first part of your syllogism where you say "if a manual-equipped car can and does have 'acceleration over-revs'". If by 'car' we mean a 997, then it cannot have acceleration over-revs. Period. There has been debate in the past about one issue that I'll explain.

We used to have tachometers marked with a yellow zone just below the redline. The idea was a caution zone that wasn't healthy territory for an engine, but didn't pose a serious risk of damage to a healthy engine. It realistically dealt with the fact that damage does not begin like flicking a light switch. I don't suppose I need to discuss that at length. In ye olden days when engine rebuilds were a routine part of the life of a beloved car, we would personally adjust the redline downward as the engine got closer to its next overhaul. The "yellow line" was a sort of half-assed way of the factory suggesting that gentler treatment of aging engines.

The limiting rpm in a modern computer-controlled engine is decided by the designers. They take into account what revs are useful as well as what the design target was that all the engineers worked toward. For example, if 7500 rpm is the agreed limit of useful revs, then your crankshaft material and heat treatment, your main and rod bearings, the connecting rods themselves, and the entire valve train are selected based on that rev limit. We don't choose parts to fail right above a limit, we have margins that are industry standards in safety applications or set by internal engineering standards in the case of a non-life-critical engine application. Porsche will have lower margins than does a heavy truck manufacturer because light weight and high performance are more important to Porsche than million-mile lifetimes in a great boat anchor engine. But in both cases, and everybody in between, the limit definitely matters. The reason we add margins at all is to deal with manufacturing variances and lifetime deterioration of parts. Porsche also uses tighter margins because they have a very expensive production line to make it possible in line with the whole light/fast/efficient motif.

Okay. Now we're coding the engine computer. Without our explicit decision, that computer is not going to accelerate the engine beyond whatever cut-off we choose. The computer controls the fuel flow and the ignition pulses. No fuel, no spark >> no power. And you don't accelerate anything without power. So what the computer decides will happen. Where we decide to put the cut-off is where it shall be, absent the sort of component failures that mean the engine usually won't start at all. But we have choices of how to implement that cut-off. To illustrate, here are a couple of strategies, not necessarily contradictory:

• Approaching the damage threshold, introduce skipped ignitions, and with a DFI car, change the fuel injections accordingly. If more power is commanded as we pass upward through this yellow zone with throttle still open, stop all ignitions. (This is called a soft limit because you can feel it coming.)
• In sporting situations, allow full engine acceleration (no skipped ignitions) all the way to the true limit, at which point reduce fuel feed and ignition strategy to whatever maintains current engine speed without further acceleration. This is usually what's meant by a 'hard' cut-off, though the usage of the term varies.
• Approaching the damage threshold, cause the engine to stutter, cutting power completely for noticeable periods like half a second, hopefully inducing caution in a driver not necessarily aware of what is happening.
• Apply a reduced performance mapping for situations where the engine is not fully warmed. This may be as simple as lowering the soft zone or it may extend to reduced power in the complete zone from the normal torque peak to the rpm limit.

Now the issue that occasionally arises is this: What do the ranges mean in that DME report? We know the numbers vary between models, and we have had people publish the technical guidance that Porsche use for interpreting the reports, but to me at least it still is not clear which models use a soft cut-off and which use a hard cut-off. I'm pretty confident the GT3 uses a hard cut-off, at least when the engine is warmed and ready to foxtrot. I would bet the base Cayenne in touring mode uses a soft cut-off strategy of some sort. In between, I can't say. If I had to guess, I'd suppose a 997S with sport chrono (a true sport mode that affects the engine, that is) provides a mildly soft cut-off in touring mode but shifts to a hard cut-off in sport mode.

Now the point of all that is perhaps range one or even ranges one and two amount to a 'yellow' zone, a region where the engine is using soft cut-off techniques to prevent the engine accelerating above the true limit, which begins at the base of range three. Perhaps. And perhaps not. The DME report of this information is in the spirit of the telltales we used at the track before telemetry came into use, so the definition of those ranges really amounts to "whatever will help us evaluate engine warranty situations." If Porsche want to know you've been spending a lot of time reving the snot out of the engine, then they'll define ranges one and two as "soft cut-off" zone. If they only care that you've blown shifts, then those ranges will be on the far side of the true limit. Still within their design margins, so "almost certainly no damage", but definitely over the limit. (Remember that revs that won't damage a relatively new engine still under warranty might well damage a more mature engine. One that is still within their design life, but starting to show wear. The limit is one that should protect the engine even when it's quite old.)

We still get back to the fact that the engine is not going to over-rev under acceleration when the computer is controlling it. Neither a manual nor a PDK will do that. What they will do is permit acceleration into the old-fashioned yellow zone. Whether they record such approaches to the redline is a decision for the field staff really. Running a design team, that's one of the questions I'd put to the field staff in a design conference because it's a question of their field support. The limiting rev's are a design question, but the treatment of approaches to that limit is a question for field engineers to answer. "Do you want that information recorded?"

Even at that, notice that the engine is not over-reving under acceleration in the sense you mean. It is going no faster than we decide to permit. Whether we use a soft cut-off or a hard cut-off, the engine will stop going faster according to whatever rules we establish. And it will not accelerate to a speed faster than we consider acceptable. Period. Laws of physics. That's the beauty of a computer-controlled engine.

Either way, the serious conditions are the ones caused by blown shifts and those are what the higher ranges of a DME report tell us about.

Gary

simsgw

If you mean the PDK, the answer is no. Those downshifts are computer controlled events and the computer won't initiate one unless the resulting rpm is under whatever downshift rev-limit the designers chose. That may or may not be what we think of as a redline. Intentional downshifts are routinely followed by the application of power, so you want some headroom after completing a downshift.

So if you're talking about a PDK, just relax. It can be surprisingly noisy if you expected an upshift, but it won't damage anything.

Gary

gota911

Gary,

Yes, when I wrote "car" I did mean a 997.

You and I will have to agree to disagree on this topic for a little longer.

Hopefully, within the next few weeks, I will be able to either verify or disprove my base conjecture that a "car" (more specifically, a Meteor Gray 2010 997.2 with manual transmission ) can achieve "acceleration over-revs."

Once I have my findings, I'll start a new thread to reveal the results.

EDIT: Entered at 9:16 A.M.

For the record, I bought my car on 9/3/11 with 2,945 miles on it and an estimated 84 engine operating hours on it (2,945 / 35 MPH average = 84) and, until today, I have never hit the rev limiter.

Just got back from a spirited drive with the intent of hitting the rev limiter. Red line on my car is 7,400 RPM. I watched the tac needle go past 7,600 but it stayed short of 7,800 RPM and, ironically, there was no noticeable "rev limiting" or fuel cut off initiated by the software, which REALLY surprised me! My 996 would have cut off the fuel supply. I am not certain how much higher it would have revved if I hadn't let off the gas. I am certain I added a few ignitions to Range 1 and possibly Range 2. However, I can not prove that because I do not own a Durametric reader

I now have 8,857 miles on it, which would equate to approximately 253 engine operating hours (8,857 / 35 MPH average = 253).

I will see if I can have the dealer do a DME readout in the next few weeks, or buy a Durametric tool to be able to do it myself. The saga continues......

Fred R. C4S

Gary,

in your analysis, have you considered engine rev "overshoots"? When all the rotating masses, particularly the flywheel, are madly accelerating towards the redline (or any given target speed) there is nothing that "puts the brakes on" and magically stops it on a dime at the target speed. We may remove the impetus for further accelerating the masses (fuel or ignition cutoff), but these masses may continue to reach higher speeds due to their momentum and inertia. I suspect the overshoot is small, but it may account for a few "counts" in the low and mid ranges in the DME counter.

jhbrennan

If I've understood Gary's previous posts, he doesn't believe that can happen. He posits that the rev limiter stops the engine at exactly the set rpm. The "overshoots" as you describe is what a PCA Tech Advisor referred to in an earlier quote I posted.

simsgw

Yep, considered 'em, but it can't happen. Our personal experience of rotating masses forms an intuition that is not correct. In gross physical events, when we accelerate something, we normally use a device of some sort, not our muscles. When we attempt to hit a particular target speed, we necessarily operate through the mechanism of our nervous system which induces delays between decision and muscle response. Then our muscles move the controls of the device and on most devices further delay is built into the control system.

All that delay gives our intuition the false lesson that things continue to accelerate briefly after we decide to stop the acceleration. They do not. Laws of motion and all that. It takes force to accelerate anything, including rotating masses. Once you stop applying the force, the 'anything' will cease to accelerate.

Actually I had my first argument about this with an uncle who insisted (in those pre-radar days) that a pitcher's fastball continued to accelerate on the way to the plate "until the air resistance catches up with all that hard toss the pitcher provided." So it doesn't take machines to confuse our intuition, just... well, lack of lab time I suppose.

Furthermore any object we cease accelerating will remain at that same speed absent other forces. Objects do not spontaneously slow either. A baseball starts slowing as soon as it leaves the pitcher's hand because drag is the force. In an engine those other forces are friction, pumping losses and so forth, but we're not analyzing all that right now. The important point is that the engine only accelerates when force is applied, when energy is pumped into the rotating masses to increase their kinetic energy.

Now remember, this next bit only applies to recent computer-controlled Porsche engines. I'll come back to old-fashioned engines in a moment. Every impulse of power is under control of the "digital motor electronics", aka the computer. No distributor, and in the dot two models not even a manifold with previously injected fuel that might spontaneously combust without a spark. (Not that that is likely. I'm just eliminating all possibilities here.) So the cylinder rises, compressing the intake air which lacks fuel at that point. Normally, the fuel is injected, the plug is fired, and then a little more fuel might be injected, the plug is fired again, and sometimes a third repetition. The expansion of the fuel-air mix under combustion converts the chemical energy into kinetic energy. It accelerates the piston downward, which pushes against the crankshaft and the knee bone connects to the shin bone and so forth. Now to know when it should fire the plugs for each cylinder, and to know how much fuel to inject and when, the computer has sensors on the crankshaft and heaven knows where else. It knows the position and the speed of the crankshaft, as well as the current load, state of warm-up and more. I have the spec around somewhere, but I don't feel like digging it up. Something like nine sensors continually reporting the state of the engine to the computer.

To over-simplify the coding technique, it comes to a point in the control flow when it says to itself: "How fast is the crankshaft moving?" It must know this for several reasons. First, because the fuel injection strategy changes with the revolution speed and also because the timing of spark pulses with fuel injection changes. When it asks that question, "how fast", it compares the answer to the limit values. Just discussing a 'hard' limit for simplicity, if the answer is "fast enough" then it applies the "just enough" strategy that accommodates motor sports by letting the engine keep delivering just enough power to remain at the limit rpm. (Remember, I make no assertions about which models incorporate this strategy. I'm just describing how such a technique works.) The idea is to protect the engine with whatever limit the engineers chose, but also to continue delivering power so you don't affect the laptime more than you must.

The bottom line: with such an engine, further acceleration can not happen. With no power impulses, or with power constrained to be equal to the load, that means that only mystical forces remain. And being an engineer, I take no heed of those. Notice the significant point here. In a fully computer-controlled engine, the computer does not try to hold the engine back when it reaches the limit. Every single power pulse was created by an intentional decision of the computer, like a bicyclist moving so slowly he has time to decide at each rotation whether to put his foot on the pedal and thrust. I don't personally know that Porsche 997.2 engines work this way. I've never pulled one apart and analyzed it, but I designed my first such engine back in 1973 and I know that the things Porsche says in their press material are plausible and consistent with such a design. I believe them.

Now let's go back and talk about ordinary old-fashioned engines which are among the devices that misinform the intuition in us older guys. They do not stop gaining speed instantly when you move the throttle. In fact, back up one step and consider the word 'throttle'. It means just what it does when you throttle a person: cut off their air. A conventional analog throttle does exactly that to control the engine's speed. If you build an engine without a throttle or if you disconnect an existing mechanism, the engine will accelerate until something breaks. (Well, most will. Some really old engines were so tightly manifolded that the airflow was self-limiting and such engines could only go so fast even without being throttled. Since they also were built hell for stout, those speeds were below the damage limit. In other words, they had a skinny intake or one with conflicting flow pulses and they self-throttled.)

I'm not referring to the pedal itself you understand, but the throttle it controls. The mechanism is adjusted by springs to close down to some minimal air flow that produces the idle rpm. When you press on the pedal of such an older car, you open the throttle mechanisms against the pressure of those springs. Those springs could break (not to mention floor mats pressing against the pedal), and that's why in driver's training we used to practice dealing with runaway engines. But back to the point: with such an analog system we have factors like fuel previously sucked into the airstream at the carburetor and momentum of the air stream that tends to slow the closing of the diaphragm that constitutes the final step in the throttling mechanism. All that means that even if a computer were sitting in your seat moving the controls, it could not hit specific revolution speed all that precisely with an older engine. We don't own that sort of engine unless we assume Porsche is lying to us in their documents.

So no. The energy stored in the flywheel and the other rotating masses cannot serve to increase the speed still further once the computer decides enough is enough.