Over-rev hit 16,000rpm "money shift" ...possible?? (1999 C4)

Hi All, I recently had an engine die. Some of you may have commented on the related thread that is a general discussion about engine failure and my present options. However, the 16K RPM header may get a few people to chime in on this specific subject. I was told by the first shop that the computer on the car said I hit 16,000rpm. ...thus the engine death was therefore the dreaded "money shift." (mis-shift) Based on the situation, this implies that I was shifting from 2nd to 3rd - and mis-shifted into 1st at high RPM (6800). I do not think I made that shift - or any shift. I believe that I was in third gear before that track section. Could such a mis-shift cause a 16,000RPM condition? ...even at red-line limiter engaged at about 7200 RPM I'm sure there must be some money shift stories out there... What gear were you in/planning to go to? What happened to the car at that instant? What damage was externally visible? Did you get a report of how high the RPMs got? ...the rest of my story: I am hearing a lot of conflicting information from shops regarding the potential cause of the blown engine. I called a couple other shops about a second/third opinions and they have told me that 16K RPM would not be possible and even if it "could happen," the heat would have caused molten metal, most likely a rod would have been thrown or other visible external damage and there should have been much more oil loss. Note: when smoke came billowing out of the tailpipe, there was no "lurching" of the car, violent, loud noises, etc. There was also no oil dropped on the track - only out the tailpipe and minor seepage under the engine. Have a great day!

I'm not sure that 996 DMEs record anything more than Type1 and Type2 for overrev, followed by a number which, if I remember correctly, represents the number of overrevs (counted per stroke per cylinder I think?) If I were a betting man I'd say it was counting it off the crank position sensor for the amount of time the tach registered over a pre-determined redline I don't believe it has a time variable stored at all, so I don't know how they could say anything about rpm

Quotes source post: 6747083

The (type 1) rev limiter works by cutting ignition/fuel to limit revs, thus in a (type 2) mechanically driven over rev, it would be worthless, since the engine is being forced to rev mechanically (thru the wheels/transmission/gears). Given the circumference of the tires, the speed and the gear ratio (owner's manual), it should be straightforward to calculate engine RPM under these conditions. There are also online calculators that do this function as well.

No way in Hell that engine hit 16,000 rpm!! The guys at the first shop are idiots! Colorado? What track? What dealer/shop did you call first? PM me with more info! I'm a Tech at the dealer in Littleton.

did the engine die at the track? or later on... money shifts are usually instantaneous.

Instant death! Smoke screen that could make James Bond proud. (more than I have ever seen from a car not on fire). Oil and coolant in tailpipe. So, to confirm the actual RPM is NOT recorded in the DME? ...just the Type 1 or Type 2 condition? BTW - High Plains Raceway at the Prairie Corkscrew (Turns #13,14,15)

Nobody else has commented, but there's no way that you jammed it into 1st gear at high rpms.....

why does it really matter what the final rpm was (BTW 16k sounds awful rounded or guestimated) - it was maybe a miss quote from a tech who "remembered" a number wrong and blurted it out...for all you know he could have been watching some motorcycle show and a brain cell connected after that last toke and the 16k just came out of the mouth unobstructed by a counter point imput from someone who knows better but didnt want to make them all look stupid so he just kept his mouth shut.... OH anyways, sorry about the engine, what are your plans... options... GL

Quotes source post: 6747360

What do you mean? Is there some type of lockout for shifting from 2nd to 1st at high RPM's?

Sounds like the ECU had limited 16,000 sparks during it's life leading up to the big over rev.. This is very possible, I see cars all the time with 10-12K over revs in their counters.. My test car has 47,00 occurrences logged! If the shop doesn't know how to read the DME correctly they damn sure don't need to be working on your car!

I think you could shift into 1st from a high rate of speed, but I don't see that as a likely mis-shift, finding 2nd instead of 4th is the common mis-shift, not 1st instead of 3rd... Anyway, it's just about impossible that your engine actually saw 16000 rpm, but it may have seen say 10-12000rpm. I've seen and even rebuilt engines after an over rev of that nature. Some run OK for a time before showing symptoms, others don't, but I don't imagine your rods would've stayed intact at 16k rpm.

Its basic math. Take the % difference between 2nd and 1st (gear ratios) and then multiply that by redline. That is your max RPM. No way you hit 16,000RPM though.

The reason I am digging deeper here is that the shop is claiming that I blew the motor with a missed shift. I believe that I was in third the whole time - specifically not shifting through that track section to practice my braking point and car balance. ...in fact focusing on braking for the entire lap. They had just done work on the car: 1) RMS replacement and 2) clutch replacement. This work was done within about one tank of gas of the engine smoking. Thus, the shop has $$$ motivation for putting the blame on me. From what I am reading here, the 16K is not possible, nor is it a "reading" that can be taken from the engine. They said multiple times that they pulled the info from the DME and it read 16k RPM. This was not an offhand comment. I did not believe that I did the mis-shift, so I kept pressing him in our conversation, he kept saying the 16K RPM computer data proved I was at fault. Since the shop has at the very least "exaggerated" the RPM data - or worse lied about it, the next question is whether they shop could have made a mistake with the RMS or clutch replacement that could have caused the engine to fail??? The second part of that question - is there some type of diagnosis that could prove I missed the shift or the shop made an error? Obviously, I have read about random engines "grenades," however, the actual numbers are very low for the amount of cars produced (clearly a topic heavily discussed already...). So, there is a third option here, the coincidence that car just happened to die 10 days after work was performed and the shop is trying to protect itself. Car condition prior to ending failure: This car was in great shape, 80K miles, well maintained (recent 75K service + all records), I had it for the past 10K miles. It was strong and consistent as 1/2-time daily driver (in CO good to have an SUV as well), with frequent mountain trips and 6 track days this summer. The RMS leak was repaired within a couple weeks of noticing the problem. It always passed pre-tech for DE events without a hitch. Thanks again for all of the input!

Quotes source post: 6747554

x= 1st gear = 3.82 y= 2nd gear = 2.20 r= Redline = 7200 RPM (x/y) * r = 12,501 max possible rev for "redline" mis-shift from 2nd to 1st That seems like enough to kill the motor, but certainly not 16K!

If you downshifted from 3rd into 1st you would have been off track backwards with a lot of tire smoke added to the engine smoke! Sounds like you need an independent look see inside that engine (post-mortem!) as it *sounds* highly suspicious that the engine blew just after work major work being done. I'd have it towed to a dealer and have a full report in writing concerning what they have found. It might pay off if they find something *missing*, installed *wrong*, *wrong part* etc. You might have some leeway on cost recovery of the repairs (or reman unit) or they may find nothing and your only out an inspection fee.

I wonder where the RPM is measured? Perhaps after something snaps some part is free to hit higher RPMs? 16K does sound really high . . .

Quotes source post: 6747083

This is a 2 byte (16-bit) number. "16,000" sounds too close to the maximum of 16383 rpms max possible with a 16-bit resolution. (16-bits max. value is 65535 and divided by 4 gives 16383 in whole numbers.) I have not seen inexplicable RPM spikes but I have seen inexplicable coolant temperature spikes from my 02 Boxster. Not very often, but the coolant temperature will come across with a translated value of 315F temperature. This of course impossible. The cause? I have no idea. It is not therefore out of the realm of believability that the rpm sensor value could upon occasion be wrong and a wrong or invalid value recorded. But to convince Porsche this occured and just so happened to occur on a track... Well, good luck. Oh, read some time ago about a mag. writer blowing up a Carrera GT engine. Missed a shift and later analysis found engine hit 14K revs before letting go. Sincerely, Macster.

Quotes source post: 6747787

Holy crap that is high but that is right.

There's just a little mechanical drag in the valvetrain though, never hit that number if you ask me!

Well those are "max possible" as he says, there's likely slip in the clutch and the tires before the engine is accelerated to it's final value, not to mention this presumes an instantaneous shift from exactly redline. Anyway, I've seen too high an average speed reading too. My last M3 trip computer showed average speed of 198mph after a session on track, obviously not very likely. I would like to add though, that an RMS/Clutch job can't really cause the type of failure you experienced, almost regardless of what they may have done wrong, and having worked at independent shops, I can also advise you that the last thing you want to do to this shop (if you like working with them), is to go there telling them how you read on the internet that their reading is wrong, and that they may be at fault. The reading is likely wrong, but due most likely to a bug in your engine computer -though I would also say that any competent engine builder should know that your engine didn't make it to 16k. However, a shop doesn't need to know this to be an effective and efficient independent garage to meet 99% of your needs.

I was pretty sure from when I had PPIs done on cars I looked at that there were ranges of over revs, the numbers which I am sure can be found online, but something like 7300 to 7900 is Stage 1 overrev, 7900 to 8500 Stage 2, etc, with Stage 3 being Over 8500??, and the DME simply counts the number of sparks (as Jake Raby mentions above) in each Stage. I also thought it only logged engine run time at the last over rev spark, so if you looked at the current run time, it would tell you if the last over rev was the final death knell, or if it occured 120 hours of engine run time previous. The DME definitely does not record the overrev RPM.

Quotes source post: 6748127

I wish my clutch and tires would of slipped but the motor is too small (Porsche motors in general). A V-8 would slip the clutch or tires - done it many times before without issue.

out of curiosity I asked a friend, says 996 DME has only the two ranges: Type 1: redline (hit the limiter) Type 2: over redline (missed shift or raised rev limiter) rev limiter redline and overrev counter redline are set independently records total number in ignitions (6 per revolution) does not record duration, only runtime of last occurrence durametric will read it Tell them to give you the actual data for the Type 2 count, and ask them how they determined you hit that rpm. I do have to agree with Ahmet, I don't think anything at the clutch-job end would have caused death by valve float, but I would make sure it wasn't say IMS failure (I still think you'd have a hard time proving their fault if it was) If it was valve float that killed your motor, I think you have a decent shot at rebuilding the top end and still having a good motor, but look at that block.

you very well may not have mis shifted. Present them with challenging info about their lack of ability to pull a 16K rpm # out of the 996 DME, that info just isn't there, where did they get that? 2nd point. them doing a clutch job on the car and replacing the RMS (seal) would in no way ever have anything to do with your engine letting go however it did let go. I would abandon that line of thinking and focus on getting to the bottom of how your engine may be damaged and how best to fix it. Sounds to me from everything you said that this shop in particular is not one that should underake diagnosing much less repairing your engine.. Period.

16k overrev is very possible! i know a guy who has a 20k aftermarket tach. http://www.samsmotor.se/koso/tacho.htm and his money shift sent it to 15k

Get them to send you the 16k number in writing. Maybe the best way to get them to do this would be in an email. Keep every scrap of paperwork and document all interactions with them. Show up unexpectedly and demand to see the engine, including the RMS, all seals, etc. Take pics. If after doing all this you feel that they screwed something up, then take it to another shop for a second opinion. In order to get them to take any responsibility for the engine failure, you will surely have to sue them. Get ready for a fight.

Quotes source post: 6748826

It might be the inertia of the needle that makes it hit that high of RPM. I cant see any motor that is reciprocating other than very, tiny displacement motors or F1's hitting that kind of RPM. But, who knows?

Quotes source post: 6748826

not saying that it is impossible... but to get an actual RPM value that killed a 996 motor is.

+1, we will never know did anyone see https://rennlist.com/forums/996-foru...40k-miles.html ?

Quotes source post: 6749158

I wonder if that guy has priced the GT3 engine and tranny yet. That will be a shocker. If my old gear charts are right you would need to be spinning your rear wheels at 89 mph in 1st gear to hit 16k. I'd be pretty surprised if you could get the car into 1st at that speed and if you did you'd know it. It's highly unlikely you'd keep the car on the track. At a minimum you would feel the rears lock up and hear the engine screaming for mercy.

Quotes source post: 6748826

Look, the M96 powerplant cannot reach 16k rpm!!! No way in hell!!!! Your buddy may have a cool tach that reads up to 20k rpm, but unless he drives a F1 car, it is just another gadget to impress the girls with! Anyone who has disassembeld this engine will tell you the same. Now as for the idea that the RMS/clutch repair is linked to the engine failure is a no go too. As Jack stated, most likley failure is "D" chunk or cracked liner thus the coolant/oil intermix. But without proper engine tear down, all we can do is speculate.

Quotes source post: 6747941

Macster, you may have the number backwords. 1 Bit=8 Bytes, right?

Quotes source post: 6747554

Incorrect. Your logic would apply for a Type 1 (engine accelerating under fuel/spark control). It would not be correct for a mechanically induced engine crankshaft rotation. Example: Turn ignition off. Connect tow strap to vehicle Put vehicle in gear. Tow vehicle now begins to pull the vehicle and continues. The tire circumference, speed and gear ratio will determine the engine (crankshaft) RPM, independent of any 'redline' on the tach or 'rev limiter' (engine is off in this case). It is the same as when you used to have one of those little spring driven mechanical cars as a kid...you would push it against the ground for as long and as fast as you could in order to get the RPM's up as high as possible....mechanically induced over rev.

^ Sorry, but the fact of the matter is, @ 2nd gear, 7200rpm, there is a given wheel RPM. If you shift instantaneously into first and the wheel RPM does not drop (in real life, there will be a slight drop during a shift), then the engine RPM can be calculated by the ratio between 1st and 2nd gear, which is what the previous poster suggested. The wheel RPM which takes into consideration its diameter, angular velocity, yada yada, can be considered the constant in this situation.

1. Would you even be able to put it in first at that speed (not that I would try) 2. How can you not be sure if you actually did or not a money shift?

Quotes source post: 6749986

No, 1 byte = 8 bits. Not including parity, of course. Also, Durametric (they claim to read overrevs like PIWIS) only displays range 1 and range 2 for the 996, not actual values. Third, range two overrevs are not the number of times the car has been overrev'd but how many sparks occurred. i.e. if a car was overreved one time but held there for a while, there could be a huge range 2 overrev number from a single missed shift.

Quotes source post: 6750039

My reply was to demonstrate that this engine (like any other engine connected to a drivetrain) can in fact be spun to 16000RPM via external mechanical forces, not internal combustion forced. That is all. It is still unclear what happened in this particular case, since the original post seemed to indicate possible 3-1 shift, but it still not clear what may have happened in this case.

Quotes source post: 6747386

Not that there's a lockout per se, but downshifting into 1st gear at anything over a crawl is very difficult. That's true with nearly all manuals that I've ever driven. When I drive hard, I'm only concerned about the 4 -> 5 shift...that I might hit 3rd instead. So....you're not going crazy -- you didn't jam it into 1st. Also, if you went from 3rd to 2nd -- let's think about that for a minute. If you would have been at redline in 3rd doing almost ~100mph, is that possible you were going that fast? Then you went down to 2nd? But the other poster is right too, the motor would have collapsed right at that moment.... I don't know about their previous repairs, but what the shop is telling you now does not compute in any way.....

Quotes source posts: 6750020, 6750039

I use redline as the RPM because that is all the rev limiter (Type1) will let you do. Its an assumption. Read below what relaxed90 said - he says the same as me. Exactly.

It is possible to mechanically over rev an engine to infinity via a downshift.. My experience with this has proven that its near impossible to get a transaxle into first gear at such high speeds.. It can be done, but it takes effort. I'll be forcing an engine failure in a few weeks on the dyno just like this and I'll let you know how it comes out.

Quotes source post: 6750920

Why would you force such a failure? What do you expect to learn? Sincerely, Macster.

Quotes source post: 6749986

No. A byte consists of 8 bits. Each bit is either a 0 or 1. A 2 byte number consists of 2 8-bit values thus 16 bits. And 16 bits can express a number bewtween 0 and 65535. Sincerely, Macster.

Quotes source post: 6751048

We do this with our jet engines. We purposely put cracks or parts that are over their time limit in an engine and run unitl failure. Then, you use destructive testing and a lot of data can be gathered. Im sure Raby is doing the same but only see what will let go first and determine the weak link of the motor I assume.

Quotes source post: 6750305

To hit 16,000 in 2nd I think you would need to be spinning your rear wheels at just under 155 mph. Again, not likely.

Quotes source post: 6751064

My computer science degree from 1982 says you are correct although in the early days bytes tended to vary in size depending on what chip set your computer was running on.

Quotes source post: 6751048

Find the weakest link. I force failures all the time, there is no better way to understand the engine.

Quotes source post: 6750920

This just screams out for a youtube video to be posted here!

Quotes source post: 6751277

I have a video of IMS retrofit bearing testing... I held the engine at 7,000 RPM for 4 hours.. Only stopping to allow the dyno brake to cool. Coolant temps were sustained at 250F the whole time.. The headers glowed cherry red and white and made for good video.. I jsut have no time to edit it.

Jake, please do post that video, I'd love to see it, or any other destructive testing video or even some pictures! To all; I don't picture an M96 revving to 16k and staying intact. Most street motors let go long before that (10-12k being a good range before rods start breaking). The shop claiming 16k rpm is indeed an interesting point, but almost completely irrelevant to the problem at hand in my honest opinion. Shifting to first is possible from higher rates of speed with the transaxles in the 996, though it would take more effort. People shift into 2nd from redline in 3rd by accident and don't notice it until releasing the clutch at nearly 100mph in all sorts of cars... In a 911 SC or some other 80s or earlier car, mis-shifting would take more work, but not with these better, modern transmissions. That said, I really don't think the issue here was a mis-shift. I think one would remember the sensation of releasing the clutch in first gear from 4th, particularly on an all wheel drive car(!). Regardless of the above points/opinions(?), it's almost inconceivable for the failure to have been caused by work on the clutch/RMS.