to change a Honda Accord 1988 model timing belt

WHAT A BOLT IS, AND HOW IT WORKS
--------------------------------

1) A bolt's sole function on earth is to serve as a clamp. Its job is to
keep two or more things from moving relative to each other. That's it.
Nothing else.

2) A bolt works /only/ because one thing happens: Its shaft rides up an
incline formed by helical threads, and the shaft thereby gets "axially
loaded" (stretched) in the process.

3) The one-and-only purpose of helical threads is to impart that "axial
loading" (stretching). Axial loading IS what gives the torque (clamping
pressure) necessary to hold the bolt's clamped parts in place, and the only
thing that keeps it from eventually backing off again. And the only way the
bolt can ride up the threads is if it rotates.

To sum up: Rotation + stretch = torque.
If no incline, then no rotation, no stretching, and therefore no torque.

The sentences above are as fundamental to the concept of a bolt (or screw)
as air is to life on earth. It is why the idea of a bolt can exist in the
first place. If the sentences above are /not/ true, then the very concept
of a bolt cannot exist.

If, as some have contended, the bolt stretches/distorts with heat but does
NOT rotate, then it has not ridden up the incline, cannot impart additional
stretch to itself, and thus cannot apply additional torque. In order to
permanently stretch, it would have to skip threads, and jump up to the next
rotation of the helix. I think we all would agree this does not happen.

If the bolt DOES rotate, then this should be readily apparent by placing a
paint mark that crosses the bolt head and the pulley, or the pulley and the
crankshaft. I GUARANTEE to you that bolt will not have rotated, and neither
will the pulley.

If the bolt has somehow rotated only on its shank, and not at its head,
then you have torsional loading (twisting), which is something that is
death to fasteners and is never allowed to happen, because a bolt that is
allowed to twist will eventually snap.
(In some rare, low stress cases, such as a bicycle brake, the bolt can be
constructed so as to allow a small degree of bending. Also, some bolts can
be specially constructed to allow shear forces, such as in a movable clevis
joint. A clevis joint is loose by design though, so not applicable to this
discussion.)

A bolt, especially one used in a high-stress automotive application, cannot
be allowed to deal with anything more than stretch. It cannot undergo
torsion, shear, or bending. The parts it is clamping are supposed to deal
with that load. It is critical, imperative, fundamental to the function of
any bolt, that it clamp with enough force to prevent its clamped parts from
moving relative to each other. If the clamped parts should start moving
relative to each other or to the bolt, the joint has failed, and soon the
bolt will also. Or it will come out.

In the case of a Honda crankshaft pulley, the pulley itself is located by a
Woodruff key. This key resists most of the torsional forces imparted by the
crankshaft and the engine's accessories. The rest of the resistance is
donated by the crankshaft pulley bolt. If the crank pulley bolt is not
tight to the point where no relative movement is possible, the Woodruff key
will get hammered flat from constant shock loading, and/or the bolt will
eventually come out. The bolt will not work as a clamp if it is
insufficiently tightened, even if it "looks" like it's tight enough.

So why are crankshaft pulley bolts so hard to remove when they've been
tightened properly?

1) Corrosion around the perimeter of the bolt head and washer. When a
clutch disc seizes to its flywheel, it's not held by very much pressure,
but it is held enough that you won't be able to free it without some
effort. Same thing happens with that ring of rust. It's not holding by
much, but getting the seal to break takes effort. The more rust, the more
effort needed.

2) Breakdown of the friction stabilizer coatings on the bolt. High-stress
bolts are coated with materials ranging from cadmium to Teflon. These
coatings make actual friction more predictable, so the engineers have
better control over actual bolt stretch in the real world. When these
coatings break down, as they can with time and heat, friction will tend to
increase.

3) "Embedment", which is when the thread mating surfaces deform with time
and vibration, and mesh more closely together at a microscopic level.
Embedment, believe it or not, actually results in BOTH a slight /decrease/
in actual bolt tension as well as an /increase/ in removal torque.

Much of the information above comes from the Web site
www.boltscience.com , which is run by Bolt Science Limited, a consulting
firm in Great Britain.
The rest comes from a series of emails between me and an individual at Bolt
Science Ltd., and some conversations with my mechanic, who has owned his
own shop since the early '80s, and has some thirty years of experience as a
licensed mechanic specializing in Japanese cars.

Finally, there is no such thing as a "one way" thread in a bolt that is
intended to be removed and reinstalled.

SoCalMike > wrote in
:

> Burt S. wrote:
>> If the bolt doesn't tighten itself on a part that has a potential for
>> play then their is a potential for the bolt to unwind. Either stake
>> it, castlenut/cotter pin it or in this case use a one-way threaded
>> bolt.
>
> interesting...
>
> now if THIS doesnt "muddy the water" i dunno what will
>
> a one way threaded bolt that relies on the pulley moving to tighten
> it... whoda thunk it?
>



There is no such thing in a bolt intended to be removed and reinstalled.

The jpeg provided by Burt shows absolutely nothing except a fuzzy bolt.

--
TeGGeR®

The Unofficial Honda/Acura FAQ
www.tegger.com/hondafaq/

"SoCalMike" > wrote
> Burt S. wrote:
> > If the bolt doesn't tighten itself on a part that has a potential for
play then their
> > is a potential for the bolt to unwind. Either stake it, castlenut/cotter
pin it or in
> > this case use a one-way threaded bolt.
>
> interesting...
>
> now if THIS doesnt "muddy the water" i dunno what will
>
> a one way threaded bolt that relies on the pulley moving to tighten
> it... whoda thunk it?

Dunno.

Burt, I am not convinced that the pulley bolt's threads are cut such that
the vibrations of the back and forth motion of the pulley tighten it. I am
not finding anything of this nature described on the web, and it doesn't
exactly pass the common sense test. I don't have a good bolt textbook,
either, so my engineering texts treat this only generally. (Usual
disclaimer: No engineer knows anything special, anyway.)

However, Burt, your discussion of how much force the pulley applies to the
bolt head during operation does provoke thought. If that bolt weren't there,
that pulley would go flying off, right? So of course the pulley exerts a
force, and surely a sizable one, on the underside of the bolt head. (I think
Tegger and Jim touched upon this reality, too. I was a little focused on
stretching by thermal effects and should have considered stretching by
mechanical effects.)

The question to me is whether then the force is enough to stretch the bolt.
If so, then of course since stretching a bolt is a known means of reducing
its diameter, then the crankshaft will tend to screw up on the bolt. (Note
for total newbies: Industry uses hydraulic devices to literally stretch
certain bolts, screw them into place, then release the hydraulic pressure,
all to achieve a certain force.)

Once the crankshaft stops rotating, the bolt stretching ceases, the bolt
length collapses as much as the thread engagement allows, and it will have a
higher axial load in it, translating to a higher torque to free the bolt.

The higher torque won't necessarily translate to axial loads that are
standard, since the threads of the pulley bolt are non-standard in at least
one way: Super fine threads.

Burt, I don't want to buy a new pulley bolt, and I want to minimize taking
my Civic's bolt off and putting it on. I may free it in the next month or
so, when I have a tire rotation to do, torque it to spec, paint a line
across pulley and bolt, then monitor the line.

"TeGGeR®" > wrote in
:

> WHAT A BOLT IS, AND HOW IT WORKS
> --------------------------------
>

<snip!>


>
> Finally, there is no such thing as a "one way" thread in a bolt that
> is intended to be removed and reinstalled.



I'm in the middle of my front brakes as I type (by candlelight, so to
speak; it's so romantic...)

Each wheel has, of course, 4 nuts. Each lug nut was tightened by me to 75
ft lbs back in the summer. Well, I just measured the torque necessary to
break them loose just now, in two pound increments on the click-wrench.

Guess what? Each one required about 90 lbs to crack free.

No corrosion, no friction coatings to break down, little time, and still a
20% increase in torque needed to release.

Now what happens to that crank bolt? It's undisturbed for years, suffers
rust, and is done up to almost twice the torque figure. No wonder thay're
so hard to undo.


--
TeGGeR®

The Unofficial Honda/Acura FAQ
www.tegger.com/hondafaq/

"TeGGeR®" > wrote in
:

> "TeGGeR®" > wrote in
> :
>
>> WHAT A BOLT IS, AND HOW IT WORKS
>> --------------------------------
>>
>
> <snip!>
>
>
>>
>> Finally, there is no such thing as a "one way" thread in a bolt that
>> is intended to be removed and reinstalled.
>
>
>
> I'm in the middle of my front brakes as I type (by candlelight, so to
> speak; it's so romantic...)
>
> Each wheel has, of course, 4 nuts. Each lug nut was tightened by me to
> 75 ft lbs back in the summer. Well, I just measured the torque
> necessary to break them loose just now, in two pound increments on the
> click-wrench.
>
> Guess what? Each one required about 90 lbs to crack free.
>
> No corrosion, no friction coatings to break down, little time, and
> still a 20% increase in torque needed to release.
>
> Now what happens to that crank bolt? It's undisturbed for years,
> suffers rust, and is done up to almost twice the torque figure. No
> wonder thay're so hard to undo.
>
>

You also need to consider the difference between static and dynamic
coefficients of friction, which could account for at least some of your
difference in torques. If you have a sensitive beam style torque wrench you
can see it if you watch carefully. Torque a bolt, stop moving the wrench ,
and tighten it further. It will take more torque to start it moving, but
once it is moving, torque decreased as long as it is moving. WARNING do not
loosen torqued fasteners with a click style torque wrench, when you see
this difference in a loosening direction, the sudden change in torque as it
starts to move can impart enough of a jolt to the torque wrench that it may
be damaged. Remember that loud click as the bolt came loose? That was
it.(Experienced it, Paid for it. $40 to recalibrate wrench.)
Keep up the good work, T!
Scott

scott > wrote in
:

> Remember
> that loud click as the bolt came loose? That was it.(Experienced it,
> Paid for it. $40 to recalibrate wrench.)


Yeah, but all my bolts came loose slowly. At 88 lbs, the wrench clicked. At
90 lbs, they turned, slowly, easing their tension gradually. No snaps,
crackles or pops, no sudden-ness anywhere.

Kinda odd, actually. You get bolts that are rusted in place, and when they
finally (sharply) let go, they can send a shock wave through your fingers
that you'll still feel days later. *My* wheel nuts are never like that.


--
TeGGeR®

The Unofficial Honda/Acura FAQ
www.tegger.com/hondafaq/

"TeGGeR®" > wrote
> Each wheel has, of course, 4 nuts. Each lug nut was tightened by me to 75
> ft lbs back in the summer. Well, I just measured the torque necessary to
> break them loose just now, in two pound increments on the click-wrench.
>
> Guess what? Each one required about 90 lbs to crack free.

For the archives, I think it's important to note that the torque wrenches
accessible to the ordinary consumer are not supposed to be used to measure
loosening torque.

I'm not saying your measurements are necessarily wrong. They sound
reasonable and not inconsistent with the reality that loosening torque tends
to be a little higher (according to boltscience, among others) than
tightening torque, due to the differences in dynamic friction (while
tightening) vs. static friction (while loosening).

I am saying from my reading, this is inappropriate use of a tool, leading I
suspect either to incorrect measurements, abnormal wear and tear on the
torque wrench, or both.

"Elle" > wrote in
.net:

> "TeGGeR®" > wrote
>> Each wheel has, of course, 4 nuts. Each lug nut was tightened by me
>> to 75 ft lbs back in the summer. Well, I just measured the torque
>> necessary to break them loose just now, in two pound increments on
>> the click-wrench.
>>
>> Guess what? Each one required about 90 lbs to crack free.
>
> For the archives, I think it's important to note that the torque
> wrenches accessible to the ordinary consumer are not supposed to be
> used to measure loosening torque.
>
> I'm not saying your measurements are necessarily wrong. They sound
> reasonable and not inconsistent with the reality that loosening torque
> tends to be a little higher (according to boltscience, among others)
> than tightening torque, due to the differences in dynamic friction
> (while tightening) vs. static friction (while loosening).
>
> I am saying from my reading, this is inappropriate use of a tool,
> leading I suspect either to incorrect measurements, abnormal wear and
> tear on the torque wrench, or both.
>
>
> Hi,
go to this link and click on the "safety" icon, where they say not to break
fasteners loose with a torque wrench
It probably was in the manual, and I try to read it all, but I sure didn't
remember it when it counted. But on the bright side, it only cost me $40
and I didn't damage the customer's machine.

http://buy1.snapon.com/catalog/item....=55265&group_I
D=954&store=snapon-store&dir=catalog

"Elle" > wrote in
.net:

> "TeGGeR®" > wrote
>> Each wheel has, of course, 4 nuts. Each lug nut was tightened by me
>> to 75 ft lbs back in the summer. Well, I just measured the torque
>> necessary to break them loose just now, in two pound increments on
>> the click-wrench.
>>
>> Guess what? Each one required about 90 lbs to crack free.
>
> For the archives, I think it's important to note that the torque
> wrenches accessible to the ordinary consumer are not supposed to be
> used to measure loosening torque.



It's not supposed to be used to measure loosening torque because of the
possibility of overloading the mechanism. 90 ft lbs is well within my
wrench's maximum of 150 lbs.

I started at 75, and worked my way up in 2 lb increments.

There is no harm done to a torque wrench used in such a manner.


---------------------

An update to yesterday: I picked up a nail in one tire two weeks ago. I
brought it to a tire place to get the puncture repaired, and watched while
the tire guy used a torque wrench to tighten the lug nuts. I noticed /all/
the tire guys were using torque wrenches as a matter of course.

By the pressure he appeared to be applying, I'd /guess/ the force used was
on the order of 100 lbs before the wrench clicked. I did not ask to see
what the setting was.

When I tried to undo the nust on that particular wheel yesterday, I reached
100 lbs with the torque wrench, with no apparent movement visible at the
nuts. I then laid aside the torque wrench, for fear of the very damage you
mention.

I enmded up having to STAND on an extended wrench, and bounce up and down
on it! I weigh 180 lbs, so I must have been applying removal torque of over
200 lbs before nuts came loose.

I'm wondering if removal torque increases exponentially compared to
tightening torque once you get closer to maximum tension.



--
TeGGeR®

The Unofficial Honda/Acura FAQ
www.tegger.com/hondafaq/

"TeGGeR®" > wrote
> "Elle" > wrote
> > "TeGGeR®" > wrote
> >> Each wheel has, of course, 4 nuts. Each lug nut was tightened by me
> >> to 75 ft lbs back in the summer. Well, I just measured the torque
> >> necessary to break them loose just now, in two pound increments on
> >> the click-wrench.
> >>
> >> Guess what? Each one required about 90 lbs to crack free.
> >
> > For the archives, I think it's important to note that the torque
> > wrenches accessible to the ordinary consumer are not supposed to be
> > used to measure loosening torque.
>
>
>
> It's not supposed to be used to measure loosening torque because of the
> possibility of overloading the mechanism.

> 90 ft lbs is well within my
> wrench's maximum of 150 lbs.
>
> I started at 75, and worked my way up in 2 lb increments.
>
> There is no harm done to a torque wrench used in such a manner.

I don't agree. Using a torque wrench as you describe puts a lot more wear
and tear on it for each assessment of a bolt's loosening torque (as opposed
to the one-time application of tightening torque). The wrench "allows" only
so many clicks before calibration becomes a concern.

Without good calibration, the torque wrench is useless.

> An update to yesterday: I picked up a nail in one tire two weeks ago. I
> brought it to a tire place to get the puncture repaired, and watched while
> the tire guy used a torque wrench to tighten the lug nuts. I noticed /all/
> the tire guys were using torque wrenches as a matter of course.
>
> By the pressure he appeared to be applying, I'd /guess/ the force used was
> on the order of 100 lbs before the wrench clicked. I did not ask to see
> what the setting was.
>
> When I tried to undo the nust on that particular wheel yesterday, I
reached
> 100 lbs with the torque wrench, with no apparent movement visible at the
> nuts. I then laid aside the torque wrench, for fear of the very damage you
> mention.
>
> I enmded up having to STAND on an extended wrench, and bounce up and down
> on it! I weigh 180 lbs, so I must have been applying removal torque of
over
> 200 lbs before nuts came loose.
>
> I'm wondering if removal torque increases exponentially compared to
> tightening torque once you get closer to maximum tension.

Off the top of my head, I would say the removal torque probably varies more
at higher axial loads than lower axial loads. That is, if one takes several
measurements of loosening torque at a given lower axial load, this
collection of "low load removal torques" will have a lower standard
deviation than that of a corresponding collection of "high load removal
torques."

The thread mating surfaces are closer together at higher axial loads. So
static (as opposed to dynamic) friction effects should prevail increasingly
more the closer one gets to the maximum allowable axial load in the bolt. In
addition, surface conditions can vary a lot due to wear, how lubricant was
applied, temperature, even the manner in which the bolt was tightened. The
variability of surface conditions from one tightening to the next should
increase the unpredictability of loosening torque at higher axial loads.

By contrast, at lower loads, the thread mating surfaces are farther apart,
so the mating surfaces' condition should have less of an effect on the
friction between them. Below a certain axial load, it's more likely to be,
say, oil molecules sliding over oil molecules. Whereas at higher loads, one
can't say with as much certainty whether it's oil sliding over oil or steel
sliding over steel, with imperfections in the steel and wear in the threads
at work as well.

Otherwise, if your car were just for lab experiments, I would be tempted to
repeat the experiment with a torque wrench of reasonably-well known
calibration. This sounds fishy, like those guys are abusing their highly
sensitive torque wrenches, to me.

Regardless, yours is a good example of why folks should probably do their
own torquing of their lug nuts, even after they've been torqued at the shop.