crank bolt right or left hand thread?

Right hand thread.

The crankshaft pulley bolt is typically extremely tight.

See discussion at

"swhaley" wrote

We never did do a test to determine whether the bolt is actually tightening over time or whether the bond is just strengthening, did we? All everybody agrees on is that it is a devil to get loose, at least for the first timing belt change.

Mike

"Michael Pardee" wrote

You wouldn't want this group to be boring, would you? ;-)

I am not sure a test exists that would persuade everyone or persuade effectively.

I wasn't willing to perform tests on my 91 Civic's.

I am convinced that the fact that it's not merely a fine thread but a super fine thread means the bond becomes stronger over time, especially when subject to heat cycling. Bolt metal melds into the metal of the crankshaft. I certainly think it's possible that this is all that's causing it to become so difficult to break loose. I won't hold my breath, but we might all agree on this point.

Having conquered (well, with a lot of group support) eight frozen, fine thread, suspension bolts, and noticing that at least one appeared to be welded to the inner sleeve (and had to be completely cut out with an air die grinder), I can believe that the pulley bolt sees similar forces and so similarly becomes "more tightly bonded" as time goes on.

The contravening evidence is J. Beam's claim that the pulley bolt becomes extremely tight again after just a very short time driving (like less than a day, IIRC).

After two timing belt changes, mine still required over about 300 ft-lbs, by my calculations.

I absolve you of the burden ;-)

I just recalled that you suggested a spot of nail polish to match-mark the bolt head and pulley, and I'm fresh out of nail polish.

All that aside, I agree that it is probably a cold-weld process that makes the break-away torque so high. People have also reported that working both in the loosen and tighten directions with an impact gun helps, which supports that theory.

Mike

"Michael Pardee" wrote

Oh right you are. I do a timing belt change next summer and might try this then.

I figure that bolt is good for only so many cycles of tightening and loosening by hand.

I'll think about that. Seems reasonable enough. :-)

"Michael Pardee" wrote in news:oa2dneT-Eoy11C3ZnZ2dnUVZ snipped-for-privacy@sedona.net:

I had an email conversation I had with an engineer at

.

He said the bolt is not rotating. Period. There are several other well- established mechanical reasons for the apparent increase in torque over time. Rotation after the final installation setting is **NOT** one of them.

with respect, the pics i showed definitely /do/ evidence rotation. there's angular galling on the washer/bolt interface. that simply cannot happen if it's static. besides, bolt precession is well known. much more attention is given to loosening since that tends to be catastrophic, but tightening happens as well.

you can do this experiment at home: loosen the pedals on a bike so they're only finger tight. now, pedal around the block. you'll find you need a wrench to get them undone again. that's precession that tightens. and it's why left hand pedals have a left hand thread.

"TeGGeR®" wrote

Be fair. You shared the message with me. He said it was unlikely, or words to that effect. That's his opinion.

He also did not offer any particularly compelling alternative explanation.

That's a very different physical arrangement, though. The friction of the bearings, small as it may be, works to screw the pedals in - as you point out, that's why the left hand thread on left side pedals. There is no equivalent force on the crank bolt.

Mike

Only the pully does I believe.

yes and no. yes, it's a different physical arrangement, but it's not the bearing friction that works the thread in [a remarkably widespread misunderstanding!] - the rotation direction is incorrect. check it out yourself. it's precession - just like the rotation of gears in a planetary gearbox.

correct, it's not directly equivalent, but there /is/ a tightening impulse from the momentum difference between an ignition stroke and the inertia of the pulley. saying that bolts don't tighten is incorrect and i used the above example as an easily tested home illustration. once we can agree on examples of where tightening /can/ occur, hopefully we can move on to examine the facts of exactly how it happens. just saying "it doesn't happen" when there's clear evidence to the contrary, makes no sense.

interestingly, the later versions of the honda pulley/crank have both splines [in addition to the woodruff] /and/ loctite as oem, unlike the older series motors that end up being discussed here. they still take torque to remove because of the loctite, but they do /not/ evidence the galling and the amount of torque to move is substantially less. i posted pics of all this back in the day. i can repost if required. it appears that the splines help reduce the amount of lash, and if there's no lash [in conjunction with the loctite], the bolt can't move hence it's now easier to unscrew.

"Michael Pardee" wrote

ISTM a number of bicycling enthusiasts throw the term "precession" around (IMO, so loosely as to be reckless) to explain the ordinary, expected motion between female and male (bolt, stud, nut etc.) threads against each other when it comes to bike pedal rotation. It seems that some of the more sophisticated ones point out that it is not the tap-on-a-gyroscopic-and-watch-it-spin-around-now-a-second-axis notion of precession meant here. Some guy named Andy tries to make the distinction as follows:

I am not sure he's being entirely accurate with his wording, either, but at least he recognizes it's not gyroscopic precession that's meant here.

As one pedals to move a bike forward, the left pedal rotates so as to loosen a right-hand thread (put bearings in between, if anyone wants). It's still simply torque being applied directly to the pedal stud so as to loosen (if it's a right hand thread) and tighten (if it's a left-hand thread, which it is on the left pedal).

I still wouldn't bet money on all the causes I propose at my web site being behind the pulley bolt becoming so tight. I would bet money that the very fine thread and heat and high dynamic load cycling does have something to do with it.

Getting material specifications would help, but using Google I can't even nail down exactly what steel is used in Snap-On socket extensions, never mind the "special bolt" used for the crankshaft pulley. (Any fool can guess of course, and plenty do.)

Well, I'll be! You're right!

Mike

good point about "all causes". i don't know all causes either, but i am however trained to observe carefully, and from that the following facts emerge:

1. there is angular galling under the bolt head. that's hard evidence of some rotation.
2. the rotation direction on the crank is such that the bolt would tighten against a "stationary" pulley wheel.
3. the apparent pulley bolt torque increases from ~120 ft.lbs to ~300ft.lbs in ~30 miles.

i also know from other research that bolts can tighten. now, the dots on this may not all be joined, but an outline appears to be there.

"jim beam" wrote E wrote

That's properly qualified and so reasonable, AFAIC. It's an outline, but nothing certain as yet.

I am interested in point 3 above. I remember your mentioning some months ago that you had generally evaluated the tightness after torquing to spec and then driving briefly. Did you redo this experiment a few times, estimating as best you could the torque necessary to free the bolt each time?

I am still not willing to remove my Civic's pulley bolt more than is necessary--too lazy and I don't like putting wear and tear on such an expensive bolt with super fine threads, and so more susceptible to stripping in my estimation, at that. Admittedly that might be overworry on my part.

I will say that in 2004 when I first got some experience with my 91 Civic's pulley bolt that the first time I broke it free (some three years after it was last removed) demanded, from memory, notably more force than the next few times I freed it. (I spent a few weeks researching and preparing to replace the front crankshaft seal blah blah, and so ended up freeing the bolt I think maybe four times algother during this period.) I did not try to estimate the torque to free it after the first removal, since I was kinda hurried.

Why is it again that you feel the abrasion beneath the bolt head could not occur while torquing the bolt to spec with the pulley fixed?

yes, best estimate #'s. i've done it twice on the 91 crx and twice on the 89 civic. pretty much the same tightening experience on both. i will say though, second release was not /quite/ as high as first. full body weight at 18" = 300ft.lbs for second release, near enough. first release requires a little "bounce" of that weight, so what's that? 330? not 400 though.

it's not delicate. pitch is 1.25mm, so not that fine.

sure, but it sure is tighter than the torque-wrenched tightening that precedes it!

there will be some abrasion on simple tightening, but that's usually really superficial. comparison between two identical bolts, one from a splined/loctited pulley wheel and one from a single woodruff/no loctite show that the latter is abrading substantially and therefore lashing, the former is not. this is consistent between junkyard hondas i've inspected of the splined/unsplined eras. the type of galling is also inconsistent with that seen from large angle rotation - it definitely appears to be lashing within a limited range.

I have experienced the tightness with age in other cars with clockwise rotating engines, also. Our Volvo took much more than the spec'd 190 ft-lbs the first time I changed the timing belt. I had a floor jack under the 9 inch socket handle and the tires had started to come up off the ground before the bolt moved. When the harmonic balancer failed a few months later it took much less. With the second timing belt change it was back to its wicked ways.

Mike

I think it's hard to estimate the effect of bounces. A person jumps say six inches, s/he accelerates to a certain velocity. Whatever s/he hits decelerates the person from that velocity, producing the force that is higher than mere body weight, as I trust you and others here are aware. Surely your bouncing is less than around 550 ft-lbs., but I base this number only on general reports of how much torque is needed to free the bolt, not any rough physics calculations involving body deceleration, and so force applied, by the breaker bar.

I am not troubled by the second release being a bit easier. Goes towards arguing that years of heat and load cycling do contribute to the tightness.

It's delicate to me, though maybe not because of the fine pitch. Maybe it's the fatigue it sees.

I have a vague recollection that the bolt is supposed to be replaced every so often.

That might be something to expect. IIRC, one important point (of many) Tegger has brought up on this subject is that the torque to free can vary quite a lot from the torque to tighten, even if it was just a few moments before that the bolt was tightened. I believe plenty of sources back this up.

It's a very inexact science, though, like many sciences, with high reliability. Torque does not directly, formulaically correlate to clamping strength. Or, rather, formulae used to determine clamping strength from torque are crude estimates. There is just so much at play: Dry surface age and so condition, lubricants, torque wrench inaccuracies, material differences from one bolt to the next, temperature...

Your opinion is noted.

"Michael Pardee" wrote

Your Volvo has a rock stiff engine mount. We have an 82 Volvo and it still runs but drives like a tank.