Rod bearing failure detection

[txhokie4life]

So if I wanted to use a Metal chip detector mechanism to detect an impending rod bearing failure,
what size particles would I be looking for.

I have a source for these MCD's that will spin into an oil drain plug.

I've lost two 944 NA engines to #2 RB failure this past season
and while we are doing several things to address it,
one more to detect it can't hurt.

I just need to know what size particulate detection I need to order.

thanks,

Mike

[MBruns]

In most all rod bearing failures it happens so quickly that you wont have hours or minutes to detect it before failure, just keep an eye on the filters/screens, and the telltale on the tach. and never ignore an over rev, thats the killer that strikes sometimes hours later when "for no reason at all" it went boom just running down the back straight. Oil temp is another item to watch as well

Mike Bruns

[matt demaria]

What I have been using for the last 20 years is the 944 s2 dial indicator mount for the cam timing. It consists of a 6 inch tube that bolts into the spark plug hole and a 6 inch extension that fits onto the dial indicator. By installing it into each cylinder at TDC (and preload the indicator) you can wiggle the crank back and forth about 10 or 15 degrees. Note the movement of the indicator. A good rod bearing will result in a very smooth swing from a low value to its peak travel value at tdc and then back to a low value. (Basically it is a sinusoidal travel per degree crank movement).
If there is ANY rod bearing wear, it will show up as a dead band of movement at TDC. This will show wear before any metal has been shed off of the bearing. It is an extremely sensitive and accurate test. I have been using it on the '44's and 911's and 356's for years.

[txhokie4life]

Hey Matt,

I'm lucky if they call me a shade tree mechanic.... So let me see if I understand things.

take a dial indicator (similar to the s2 or for a 964 valve timer)
instead of resting on the valve -- you make an extension and rest it on the top of the piston
at TDC through the sparkplug.

observe the travel as you move +/- 15 degrees on the crank, it if moves linearly your good,
if it doesn't move then spikes -- the rod bearing is out of tolerance and is suspect.
Or better said -- if I were to map the movement -- if I have a plateau -- then I have an issue.

thanks,

Mike

[matt demaria]

For the second question---yes.
For the first question, it is the tool that you use in conjunction with a dial indicator to do valve timing on an S2. It consists of a hollow tube about 8 inches long. One end threaded with 14X1.25. The other end is drilled with an 8mm hole with a thumscrew 90 degrees to the hole. There is also a 7 inch extension for the diaL indicator which of course threads onto the indicator. Unfortunately I do not have a P-number stamping on the tool.

[Walt Fricke]

Sounds like something a guy could easily enough make. Bust out the center of an old spark plug (or drill a hole down the center of a 14x1.25 short bolt.

Weld on a tube to extend things out far enough.

Fabricate a fitting on the end of this tube with a hole and a thumbscrew to hold a dial indicator.

Though 911 plug holes are at angle to the piston surface. And those surfaces are usually rounded in some ways. Wouldn't this affect things? Put some sidewise pressure on the tip of the indicator?

Get a suitily long dial indicator extension.

[aws]

I agree with Mike that the time between a rod bearing shredding metal and the rod parting company with the crank is measured in minutes---less than the time for a complete lap. (or maybe two) What I have done for years on 911's has been take a large towel and cover the megaphones (or single open exhaust) to dampen the noise and then rev the eng slightly to vary rpms and listen for any tell-tale rod knock. Over the years this has saved me from one blow up.

I will admit the dial indicator check intrigues me----has this method resulted in some tear downs that resulted in rod bearings being way out of spec??

aws

[txhokie4life]

What I had envisioned was a detection system which caught
debris in the oil -- then some sort of signal that
either put a big stinking spot light on the drivers dash
or maybe even cut the DME signal :-)
Some concern about immediate engine shut off -- but would that be any
different than exploding engine case?

I'm talking a 944 here -- but you could do any engine.
However, I do like the quick method of pulling a sparkplug and
rotating the crank. For the 944 -- I need only check #2.

The last two times it went -- 1,3&4 were pristine.

Mike

mike

[MBruns]

When we raced IMSA firehawk with the 944's both 2 valve and 4 valve that was a common problem with #2 we could not do anything to the engines and remain legal. what we found by adding some washers to the oil press spring and at the time ran Mobil 1 15/50, and added almost 1 quart to the oil capacity (on the round dipsticks) fill above the full line to the beginning of the round section. if it gets over filled by a bit it will smoke on fast LH turns like turn 1 at Sebring. after loosing several bottom ends before that change we ran 5000 race miles and never had that issue (not 5k on the same build) Mobil 1 is not the same as it used to be, but a good synthetic like J.Gibbs and be sure your clearances are not to tight you should be fine.

Mike Bruns