Torque settings, what can affect them?

[Sphinx]

Hi guys, I'm sure I remember seeing a similar topic on this very forum but can't find it for the life of me and it got me wondering;

What can affect torque settings? Gaskets? loctite? super clean threads (oil) etc etc

The reason I ask is becuase I've recently replaced the thermostat housing on my e36 BMW, anyone that knows these cars will know the housings are made of plastic which over time becomes brittle. Also, the bolts were looking a bit cruddy so i sprayed some WD40 on them and ran them in and out again to clean all the threads. Anyhoo, after I'd replaced the thermostat I went to re-fit the housing and was nipping up the bolts bit-by-bit in a circular pattern until I got the torque wrench on them set to 10Nm (sod all) and surprise surprise the casing snapped before it could stand the minimal pressure of compressing the O-ring.
After some faffing (read, breaking a new housing in the same way) I bought an alloy version.
Now a few things with this: it's an upgraded part not specced by BMW so I was concerned that the torque may find the next weakest link, the bolts (tiny little 6mm bolts into an alloy head made of chocolate)
Also, being alloy on alloy I thought it best to use some form of gasket between the two surfaces, paper was ruled out becuase it would get in the way of the sealing of the o-ring so I smeared a tiny bit of sealer around the mating surface.
As I was nipping the bolts up I could see the sealer starting to squeeze out and it felt tight so I didn't dare go up to 10Nm.

So finally, my question again; what kinds of things normally affect the torque setting that you should be aware of before things start snapping.

Thanks

 

[rtbcomp]

It's the relation between torque (twisting) and the stress in the bolt (tension) that's important.

In specifying a torque they're really specifying how much tension there is in the stud, so for a given torque thread pitch (fine pitch = more tension) and friction between the nut & stud (more friction = less tension) are the important factors.

A manufacturer will assume clean threads when specifying torque, if they are to be oiled they usually say so. If a thread is dirty the nut may stick on the stud which could then break due to twisting rather than over-stretching.

A gasket or seailng compond between the mating surfaces won't affect the relation between torque & tesion.

For your application with an O-ring I'd just make sure there is even torque in all the nuts so the housing sits square and doesn't distort. Spanner lengths are usually designed so you can't overtighten a nut, but this depends a lot on the strength of the operator!

 

[gordon stephenson]

As rtbcomp says it's the relationship between the torque reading and how much tension this gives the bolt/stud,
Quite a lot of critical machines have bolting figures expressed in "thou's etc"
i;e measure each individual bolt length with micrometer etc and record, then tighten said bolts to achieve the required elongation,which will be well within it's range of elasticity

 

[rtbcomp]

It's worth noting that as a nut tightens the friction between the bolt & stud will increase so you're mor likely to shear a stud than snap it, lubricating the thread will reduce this tendancy.

 

[tigler]

Also, torque figures are specified with DRY threads unless specified. If a fastener is lubricated the force applied with a dry specified torque figure will be much higher.
IME, torque figures on small bolts like the thermo housings are best ignored and tightened by feel, After all, it's non-critical as you can easily go back to it to tighten a bit more if needed.
Where torque becomes critical is on stuff like big-end bolts etc, get that wrong and it will end in tears !

 

[Sphinx]

Thanks guys.

What about fatigue? ie, the bolts I'm talking about are very small and will have been on the car for 16 years going through continuous heat cycle. Becuase of the 2 previous housings (and perhaps some before my ownership) they would have been fitted and torqued up at least 4 times. Would this make the specified 10Nm setting a touch too high?

I suppose what I'm getting at is it's pretty common for some of these small nuts and bolts to break on these cars or to strip threads, even on a careful user (cam covers, water pump studs, thermostat housings etc etc) so I'm wondering if some of these settings should be adjusted and for what reasons.

 

[Chipper]

Depending how cruddy, and how old, I would replace cruddy bolts, just for ease of in/out as much as anything.
Smidge of copperslip, especially if steel into ally.
That would affect torque but 10nm is naff all so wouldn't worry about 10-20% on that.

If the O-ring is good (pref new) and the mating surfaces are clean, there's no need for gloop. The torque is just enough to make the O-ring do it's thing, and stop the cover falling out.

 

[mike os]

I would very rarely try to torque any small fixing (less than 10mm) unless in a critical area....as above.

I find small spanners or 1/4" sockets the most useful as there is less temptation to apply that extra little bit of force.... usually ending in tears.

there is a reason say a 6mm spanner is that size....... becasue it is more than you need to fully tighten that size fastening..... same with a 1/2" socket tool.... too big for small fittings & too easy to overtighten without thinking or effort.

for 10Nm.. pinched up is about it.....

was there any fluid in the bottom of the hole? or is it a through hole?

 

[Sphinx]

What's been said about not torqueing up non-critical small nuts and bolts like this makes a LOT of sense; I was thinking that doing everything to the correct torque would be the best way but thinking about it and seeing some of these replies, it's obvious not to. The bolts weren't as cruddy as I'm probably making out, just a little dirty and I didn't want any bit of grit getting in and ruining the thread.
The O-ring was new along with the stat, but I only smeared a veerrrrry thin layer of sealant (almost translucent it was soo thin) to create some kind of barrier between the alloy head and the alloy housing.
What's been said about the torque being set on a dry thread is interesting, so 10Nm on a lubed up thread will end up higher and possibly damaging something. Does the same apply to the likes of using loctite?
The other problem is that although my torque wrench goes down to 8Nm, it's too large to give any sort of feel on these low settings, which is why I stopped and called it a day when I felt it was tight enough.
I'm not sure what the last reply is asking about the hole, it's not a through hole and it's horizontal so any oil that was used to free it before I started (and again to run it through) will have settled to the bottom.

Thanks again, some very interesting technical advice here

 

[tigler]

Any liquid on a thread will reduce friction and so increase the effective force for a given tightening effort. All you need to remember is that unless otherwise specified manufacturer torque specs are for dry assembly.
As a rule of thumb I do everything up by feel except for main and big ends, cam bearing caps etc, also suspension, drive shaft, hubs, brake calipers etc.
Also, most good quality wrenches, spanners etc, are designed to give an adequate tightening torque for most general purpose fasteners with hand force applied by the average person without exerting extreme effort. So you can reasonably rely on bolts being tight enough without using a torque wrench.
You need a very good quality torque wrench for it to be accurate at very low torque figures. Also, torque wrenches need regular calibration if well used.

 

[the snooper]

im with tigler ignor the torque setting for non critical components such as thermostat housings just nip them up

 

[mike os]

liquids in the bottom of holes has a nasty tendency to split parts or strip threads........basically turns it into a hydraulic screw jack.....

 

[tigler]

I've actually used that technique to remove blind bushes on old British bikes ! Thread the bush with the closest size then fill the cavity with grease, screw in a bolt and Bob's your Auntie !

 

[rtbcomp]

I've actually used that technique to remove blind bushes on old British bikes ! Thread the bush with the closest size then fill the cavity with grease, screw in a bolt and Bob's your Auntie !

I've seen that recommended for removing pilot bushes in flywheels - fill with grease and push a close fitting piece of round bar &c into the bore, the pressure forces the bush out.

 

[series2]

I've seen that recommended for removing pilot bushes in flywheels - fill with grease and push a close fitting piece of round bar &c into the bore, the pressure forces the bush out.

works a treat, use that method on most occasions. The thicker the grease the better!

 

[Justme]

Check the torque wrench for rough accuracy by trying to tighten a big bolt & see how much pressure it takes on the wrench to trip the 10nm setting. Could be its a bit on the stiff side.

 

[Sphinx]

I started to wonder about the torque wrench do I did a similar test a it felt fine to me, however by then I'd decided against using it and just nipped it up. I was just fretting that it wouldn't be as good a repair as using the correct settings, but the replies here have confirmed that's not the case.