Engineers: Question about engine temp/valve adjustment:

[speeder]

This has probably been asked before, maybe even by me, but if a valve adjustment spec is for a cold engine, can you do it when engine has been turned off for 3 hours and is still slightly warm? Or does it need to sit overnight and be stone cold?

How about head torque? This is for a BMW airhead MC and I know that the perfect scenario is stone cold, but I guess my question is more about the speed of expansion and contraction @ different temps. I'm imagining that things start expanding for real only when they get real hot, (operating temp), and shrink back quickly when that temp is lost, (cool down below metal-stretching temp).

Are these dumb questions? Sorry in advance if so, but this has always bugged me. TIA.

[island911]

Yeah, the math has been done here before. (search)
The thermal expansion differences here - not rocket surgery. But, it is easier to tell people to wait overnight . . .just so they'll wait a hour after flogging the thing.

I'll say again, the gap tolerance is huge. .004" with +/- .002"

Or, you can make sure you find an air conditioned garage (68F) at sea level with 50% humidity, and then fret over getting them to .004 +/-.000 01 . ...after letting it sit over night, of course.

[speeder]

Thank you. That's sort of what I thought.

[cockerpunk]

coefficient of expansion for steel is 6 millions of an inch per inch per degree F.

aluminium is about 12.

[onewhippedpuppy]

We engineers have to make everything complicated, it's called job security.

[island911]

oh, right....

disregard the above post of mine and read this Temperature Estimation in the Combustion Chamber of an Internal Combustion Engine

And then we'll build an Finite element model to run the computational aerothermodynamics to quantify the transient behavior of a cooling 911 head assembly.

and by "we" I mean cockerpunk.

[cockerpunk]

actually im such an american "eh, this back of the envelope calculation means it will probably work" type engineer.

which is really funny because i work in hyper precision engineering, where numbers like the above coefficients of expansion matter A LOT. haha

[Jeff Higgins]

[island911]

LOL

with the tag-line: Because that's how we roll... with a hex. Close enough to round.

[wdfifteen]

Originally Posted by onewhippedpuppy

We engineers have to make everything complicated, it's called job security.

Are you the guy who wrote the specs for my Dodge flathead? Valve clearance is given for a hot engine, but the adjusters are behind the exhaust manifold. You don't have to take it off to get to the valves, but you'll burn a lot of skin trying to work around it while hot.

[island911]

Denis, if you want to be a quick expert on the subject, you could put a dial-indicator on a rocker of a hot engine, (Don't burn yourself.) measure the gap, then let it cool, re-measure the gap. --Most dial-indicators have resolution to half a thou (.000 5) inch. ...so maybe if the hot CHT is ~280F you can measure a thou difference when it cools to 100F. ...See if the Close Enough /Probably Okay/ Engineering numbers from cp deliver as expected.

[on2wheels52]

After you get the temperature correct, let me know when you get on the subject of the correct amount of drag on the feeler gage.
Jim

[speeder]

Ah yes, the drag of the feeler gauge must be quantified.

[LakeCleElum]

Friend of mine is a well respected wrench at a Honda (car) dealership......He pulls the valve cover and lays a big fan on the top of the engine for an hour.........He sez: Don't overthink it.........

[onewhippedpuppy]

Originally Posted by speeder

Ah yes, the drag of the feeler gauge must be quantified.

Better hook your fish scale to your feeler gauge to measure removal force. I'm sure you already have one to precisely tighten belts.

[Tobra]

Cool enough that you don't burn yourself taking off the valve cover

[Pazuzu]

The old Toyota trucks were specs for adjusting the valves when the engine was hot, but that sucked. It was shown over and over again (with measurements) that the difference between the hot and cold measurements was 0.001 inches (10% for the 0.01" specs)

If you either use the 0.001" or the 10% variation spec, either way, the 911 valve spec of .0039" can be .0029"-.0049" (0.074-0.124mm) or 0.0035"-0.0043" (0.089-0.109mm).

The second set of values is the same range as the backside method, which is pretty well accepted as "perfect". both sets of values are well within the expected 0.004+-0.002" range.

Yes, i know the original question is for a BMW MC, but real world facts are that an average engine can have the valves specs anywhere between stone cold and "long enough after running the ending to find your tools, open the valve cover, turn on the work light and get to work" without really affecting anything.