How much damage from a 5-2 shift?

['78 SC]

How extensive is the damage from an over-rev?

I had an unfortunate over-rev at a recent event, aka the "money shift". Based on the wheel speed the engine probably exceeded 9000 RPM. I ran a leak down test to confirm damage, then tore the engine down. Here are the piston tops - looks like the exhaust valves made contact with 1 and 3 (first photo), as well as 4, 5 and 6.



Clearly the heads have to be done. My question - what else needs to be checked? I read here that the rod bolts and bearings can be damaged. When I built the motor last year I used ARP rod bolts (torqued by stretch measurement) and coated Clevite bearings. Did the upgraded parts buy me adequate protection, or do I need to pull the rod caps and check the bearings? What about the pistons? This looks like they took a pretty good shot.

Thanks,
Steve

1983 D stock (3.0 L)

[Tippy]

Pull the rods and have the big ends checked for ovality and resized (cheap). Doubt the rod bearings will show anything. Mine didnt, and I held it there for awhile at the same RPM.

You can have the rods inspected for cracks too.

['78 SC]

Originally Posted by Tippy

... I held it there for awhile at the same RPM.

Ouch. I checked my data and the peak RPM was for less than a second.

RE: checking the rods for cracks, is that just visual or something more specialized (like say, magnaflux)?

[Catorce]

You've already got the motor out of the car and torn down to that point already.

Do you feel like guessing whether there will be further internal damage, or do you want to do it the right way and tear the rest of it down and inspect each piece?

Sometimes we come here and ask the questions we already know the answers to.

[Eagledriver]

This happens frequently among racers. I've known of about 5 instances in the last 10 years. None of them had damage beyond bent valves. I'd inspect the pistons and call it good.

-Andy

[smokintr6]

If you installed the ARP bolts following the procedure completely, you can check them and re use if they're in spec. You need the original free length for each bolt, then compare to the current free length after removing all torque. The document tells you how much permanent elongation is allowed... off the top of my head its something like .001" but you need to double check. If you didn't record the free length I would probably replace them...$250 in rod bolts is still a lot cheaper than new cases and a new crank. I would probably have the rods checked completely at a machine shop too.

[milnejk]

I had a similar money shift on my D Class Race car. Got to the point you are and had the same questions. My car had been a race/track car for several years, so I decided to tear the engine down and check everything. Found rod bearings very worn and was glad I checked. Ended up doing a full rebuild and converting to 9.8 euro pistons.

Don't know the history of your car, but if you have any concerns about the condition of your bottom end, now would be a good time to check.

John

[Flat6pac]

There is minimal damage on the pistons. Buff the point off so they don't become hot spots, do the guides, bent valves. You were in the bottom last year so should be good. It's not like you have forged rockers.
Bruce

['78 SC]

Guys, thanks for the wisdom. In the spirit of "trust, but verify" I'm going to "hope, but inspect" and pull the rods. I have the relaxed rod bolt lengths so I can check them, the rods and bearings. I'll sleep easier that way. I hope to prove Andy and Bruce are right and that the damage didn't go beyond the heads. I'll post my results here.

Bruce, thanks for the suggestions on the pistons. I'll check around the pins for damage and take down any burrs on the top side. The CIS pistons seem pretty beefy.

Thanks, all. Stay tuned.

[DRACO A5OG]

+1, check those rods and valves too they may be bent after those smacks. Geez, 9K RPMs, that is amazing

[jpnovak]

It is also good practice to take the weight of your rods and pistons and calculate the reciprocating force at 9K rpm. Make sure this does not exceed the tensile strength of the ARP rod bolts. I think you have plenty of margin here but I would run the math to make sure. I think ARP can help in this calculation.

['78 SC]

Good point, Jamie. I was thinking of this as an impact event (piston smacking the valve) but the RPM loading is also an issue.

ARP has some good info here. The loading increases as the square of the RPM, so 9000 RPM causes 2.1 times the design load (if you use 6200 as the design RPM) (or 1.6 x if you use 7000). According to the site, standard practice is a 100% safety factor. So, 9000 RPM is probably within limits (barely).

As mentioned earlier, the deformation limit for ARP bolts (relaxed) is 0.001 inch per installation instructions. I'll check to ensure that has not been exceeded.

We all like pictures. Here's how I measured rod bolt stretch and big end ovality (is that a word?) last year.

[E Sully]

I bought a used '86 engine with around 70,000 miles. When I pulled the heads, it looked just like that. Re-built the top end with new valves and guides, polished the piston dings and new rings. The bottom end and cams were still good. Engine came out running like new.

[bpu699]

What's redline on an sc? I thought it was 6700 or so?

Did you car even run after the missed shift? Or was it pretty obvious damage was done while driving?

['78 SC]

The over rev happened in the braking zone of turn 2 at Texas World Speedway; the car made it to the garage (about 3 miles) under its own power at reduced speed.

The driver knew immediately there was a problem (and got the clutch in within 1 second). The car ran, but did not sound healthy. No ugly sounds, just not smooth. The 911SC motor has a reputation for a strong bottom end (see comments above). We'll see.

The redline on the tachometer is a band from 6300 to 7000. There is a cutoff that kills the fuel pump at 7000. i normally shift just after 6K. I will continue to do this until the prize money for winning the DE is increased substantially.

[Walt Fricke]

Redline is a somewhat ambiguous term. The 2.7s, and the 78-9 SCs, and the Euro 80-83 SCs, have an RPM not to exceed of 7,000 RPM when you add in the +/- factor Porsche gives. Same crank, same rod bolts, more torque and inertia with the 95s than the 90s. As a practical matter, if you kept accelerating occasionally beyond 7,000 RPM I don't think you would damage anything.

The tach red line is more of an indication of where you should upshift in order not to waste acceleration potential. You want to upshift where the torque delivered to the wheels is the same in the old and the new gear, which for a US 80-83 motor is around 6,250. A bit higher in the lower gears (you can't reach it in 1st if you have a rev limiter), but 6,250 is right about where the tach says shift. And that is well below where any mechanical issues might arise.

I think the problems with over revs come mainly, or maybe only, from the extremely rapid accelerations from a missed shift - the motor spools up way faster than normal, and the valve springs can't keep up. Stock springs can keep up with normal acceleration up to at least 8,000 RPM. Which is why bent valve stems, which fairly quickly can mess up the valve guides, are the most usual result, along with the small marks on the piston crown.

Which is not to say that a missed shift can't also overstress rod bolts. If they are taken beyond their elastic limit, the clamping force will be reduced, and the mechanical limit of the bolt reduced, leaving less margin for the next time, or no margin at all, or even a gap at the mating faces, which is promptly fatal. Ovality is another issue - I was amazed to learn that just torqueing/stretching the rod bolts changes ovality, hence the need to hone with fully stretched caps/bolts.

My take is that if you didn't break any rockers, damage most likely is limited to the valve/guide area. Those exhaust valves are pretty hot, and it doesn't take much to tilt the head a little, which showed up in your leakdown.

First time this happened to me I ran for over a year, getting faster on the track, until someone who knew what the car should feel like accelerating suspected bent valves - all six exhausts had S shaped stems, and by sawing into the guides had sort of straightened themselves. Risk was that one of the heads might have broken off. Alas, after replacement and a shop guide and valve job, one of the new ones did break off, blowing up the motor - you never know.

I think I'd have done the heads, smoothed the dings, and called it good as Bruce suggests.

[ChrisBennet]

When my valves kissed the pistons they left a "C" mark on the carbon. I told people "I monogrammed my pistons" (My name begins with "C")

['78 SC]

I got the heads broken down and ready to send out for machining. I was surprised that there was no conspicuous damage to the valves - apparently it doesn't take a lot of damage to give a 55-75% leakage. Probably the impact with the piston just tweeked the head of each valve enough to prevent it from seating.

Just for grins, I propped several of the heads up with the valve face horizontal and poured in enough alcohol to cover the valve. I was surprised at first there was no leakage, but after a while the liquid made through and ran out the exhaust port. Only those with the worst leakdown results leaked.

['78 SC]

Heads have been shipped out to G2 (cgarr). Rockers, too since I don't plan to be in the motor again soon. Didn't I say that last year?

Good news on the rod bolts - so far I've checked 4 rods and found no change in (relaxed) bolt length. Glad I went with the ARP rod bolts last year. The extra margin of strength may have been a factor and the fact that they are reusable allows me to check instead of replace.

Rod bearings look good, too.

[Tippy]

Yes, I'd bet just a thousandth of an inch out of concentricity of the valve face to seat would negate sealing I'd imagine. Point is, you can't see that with the naked eye.