Engineering question – bolt torque

Is the torque required to undo a bolt the same as that required to do it up?

If I do not have the torque spec for a bolt, could I set the torque wrench to increasing values until it turns without clicking, and therefore gain an approximate value for the tightening torque?

Interesting question.

I imagine it depends a lot on how long the bolt has been in place. 🙂

Depends on the size of bolt but wouldn’t have thought so, breakout torque is usually higher than tightening torque

In my world working with 10s of thousands of foot pounds the break out torque is higher than the make up torque.

It can take up to 2.5 times as much torque to break out as it did to tighten it ….. And thats even if i just make it up and break it out straight away. – dependant on thread.

your method would be a poor way of working out torque required on a bolt as the torque setting for a bolt is based on many thing but. In mtb world the main thing is “clamp force” and thats a function of design and surface area. , the bolts deformation/stretch limitation could theoretically be much higher , by which time your componant is crushed.

Speaking from a position of total ignorance I think your method would be fine if you can eliminate all friction, static or otherwise. I wonder if bolts would unscrew themselves and pop out in my new frictionless world?

It’s an M5 bolt I think, and won’t be corroded or anything. It’s not MTB, it’s in the transmission in the car. The bolt holding one of the hydraulic valves in place (don’t ask me what I’m doing, please don’t…)

No idea of the torque other than feel, otherwise.

I know the pressures the valve has to handle, so theoretically I might be able to calculate the force required to clamp the valve body to the back plate to seal.. but really the main worry is not leakage, it’s the bolt coming undone.

Fundametally no, you can’t use that method to obtain an accurate measurement. It is likely that you would come up with a value that is too high and this is not a good thing as overtightening a bolt can mean that it will just come loose (I’m not sure of the exact mechanism but I think it’s the difference between elastic and plastic deformation of the metal.)

Whether it’s important to have the correct torque however depends on the answer to the question “what would happen if the bolt comes loose?”.

If the bolt comes loose.. I guess I’d get symptoms and warning lights and the tranny would stop working before it fell out completely. If it did fall out I don’t think it would lunch the tranny.

I think I have torque specs for other bolts of the same size in the same environment, so I guess if that’s enough to stop those coming loose I can just use that same spec, given lack of any other info.

Can we get s big hand for the henkel corporation

A whole host of products for the paranoid 🙂

(don’t ask me what I’m doing, please don’t…)

So… Mols…

Always seems much harder to undo some bolts than to do them up to torque spec. Many factors involved in them seizing or locking, plus if there’s loctite etc involved (though that itself makes tightening to spec more complex unless the spec includes loctite).

Anyway, more importantly is most (if not all?) torque wrenches should never be used to undo bolts, so they say. More in that the wrench can be damaged I believe due to wrenches not working well in reverse than the torque will be wrong, but perhaps indicates the torque for undoing is not going to be right.

can you make a mark on the bolt and the corresponding metal its in then undo it a bit and retighten to that mark with the torque wrench set low, and then gradually increase the torque wrench settings until it reaches the mark?

I’m not an engineer so have no idea if that method would work, but in my non engineering head it seems like a good idea 😀

I reckon he’s removing a chandelier from above.

The torque wrenches we use at work stipulate that they shouldn’t be used to undo bolts. I’ve always assumed this is because of the much higher force required to undo a bolt.

That’s for bolts at 1500N.

it’s in the transmission in the car.

Here we go again…

Just look at the materials you are bolting together, the function that they are performing, and tighten as necessary. Exercise some common sense.

cbmotorsport – Member

Just look at the materials you are bolting together, the function that they are performing, and tighten as necessary. Exercise some common sense.

Alternatively find the correct torque spec, as developed during testing, carried out by the particular OEM in question.

I’ve an A8. Wouldn’t dream of friggin with the transmission, that’s what independent VAG specialists are for. A nice one might even tell you what torque to use. This would give me more time for changing my own car tyres, for example…
Then there are workshop manuals available online, also.
Other thoughts drift towards the application of the correct thread lock adhesive to prevent the bolt in question, loosening in service. However, if thread lock is applicable, there will be some indication of it’s use on the thread, from the initial installation.

Edit:
can you make a mark on the bolt and the corresponding metal its in then undo it a bit and retighten to that mark with the torque wrench set low, and then gradually increase the torque wrench settings until it reaches the mark?
Do not do this.

I would not suggest you use your torque wrench for loosening said bolt

Is this the same gear box you repaired before or have you gone into business fixing them 😉

Same one yes.. it developed a juddering when pulling away, that turned out to be a worn valve seat in the clutch control valve. Rare fault, no-one on the internet had it, so I’m going to assume it was down to the incorrect operation caused by that dodgy ECU business.

So, rather than £1500 for a new mechatronic unit I found a bloke in Denmark who sells the valves for £60, and changing it is just about the simplest fix imaginable. However just as I had the unit on the kitchen table I realised I should probably have changed BOTH valves since it was a faff to remove and the valves were cheap. Sure enough, there’s still a problem with the second valve that was masked by the first one which was worse.

VW are being rather strange with this thing. It’s a great piece of design, putting all the electronics and valves on one single unit at the front of the transmission, where it’s very easily serviceable; however they won’t service it, they will only sell whole new units at huge cost. That’s damaged the brand imo – if people knew you could repair many if not most mechatronics faults for £60 and an hour’s labour, it’d make the thing a much more attractive purchase proposition.

Alternatively find the correct torque spec, as developed during testing, carried out by the particular OEM in question.

It’s not published, cos it’s not a recommended repair procedure.

Wouldn’t dream of friggin with the transmission, that’s what independent VAG specialists are for.

Hmm.. so let me see. £1500 for a new unit, £450 + lots of postage for a refurb by an indy, or £60. Ooh that’s a tough one.. hmmm…

It really pisses me off how people say ‘take it to a garage’.. ok .. you gonna pay for that? Do you think I’d be farting about with it on the driveway if I had the cash to pay someone else? Let me remind you that all this shit started when I took it to a bloody garage.

Tighten it until it breaks or strips and then back off a bit.

I certainly have a lot of sympathy and tend to agree with your attitude to repair. Recently had a problem with the CH valves – the bloke who replaced the fan on the boiler (which I wasn’t doing myself) suggested that I try to source identical replacement valves, so that just the top electromechanical bit could be replaced without touching the wet bit. Couldn’t find any as they’d been discontinued, so decided to take it apart to see what was wrong. Traced the fault to a microswitch – found an identical replacement online, and was about to order, but thought I might as well pull apart the microswitch first. It came apart easily, I filed off the spark damage on the contacts, it went back together just as easily and CH now works fine. Would have been £60 for a new valve (if I could have got one), £2.50 for a microswitch, free to use a file.

Is this the same gear box you repaired before or have you gone into business fixing them

Is this the same car that everyone told you to sell after you finally got it going again? 😉

Buy some Nordlock washers if you’re worried about it backing off. Incredibly good bits of kit.

Yes.

Don’t start.

What gearbox?
(I never seen the original thread, and can’t be arsed searching for it)

If it’s a ZF automatically shifted manual in disguise, then there is a chance I may still have a complete valve/actuator mechanism lying at work that you could have for parts. It was changed on the advise of another workshop who’d done several, but it made **** all difference to the problem vehicle I fitted it to, and the old one was still lying on the bench at last sighting.

However, a juddering/kangaroo clutch on these boxes is more likely to be a dirty/worn clutch position sensor. A leaky valve should still give a consistent engagement, unless the valve was sticking.

Alternatively find the correct torque spec, as developed during testing, carried out by the particular OEM in question.

OK. Provided it was ever tested, catalogued and published, otherwise my suggestion still stands. Torque ratings are mainly to stop things coming undone, and to prevent damage from over tightening, and damage when using different materials with different properties. A bit of common sense and basic engineering understanding should allow you to get it right.

What gearbox?

VAG DSG.

I eventually caved in and took it to the main dealer – they found a reference to the fault in a VAG technical service bulletin – worn valve seat. The car can’t control the flow of fluid properly and hence clutch pressure – so has no choice but to oscillates either side of the desired value.

The nice folk recognise me at the dealer, so they didn’t charge me for that one 🙂

Once we’d figured out it was the valve, google’s first hit was a great page by the Danish guy detailing valve problems and replacements. Replaced the clutch 1 valve and problem went away. Only later noticed clutch 2 doing it as well, it’s much less obvious.

I think it’s to do with compression/extension of the bolt when you’re tightening/loosening it, so that there is greater friction on different portions of the bolt depending whether it is going in or out. I’ve tried thinking about it in a more exact manner, but I fear my head might explode.

no, because STATIC friction is always greater than dynamic friction.

If you want to estimate the torque of a bolt, the best method is the “return to mark” method (mark head, undo bolt, retighten, measuring torque until marks re-align) but this can be difficult without a “dial” type torque wrench

(assuming you can’t get to both ends of the bolt to measure stretch in-situ)

http://electronicfilters.tpub.com/TM-10-4330-237-13P/css/TM-10-4330-237-13P_105.htm

in my new frictionless world?

I don’t think that’s a good idea, getting around could be interesting as could stopping. Yes I know a rocket doesn’t use friction but it would be a nightmare trying to get out of your chair with no friction.

It’s not published, cos it’s not a recommended repair procedure.
Of course, you are always correct… However, once upon a time a man in a factory built the gearbox you are messing with and the tool set to drive in the bolt in question, did so to a specific torque, so a torque value does exist, fact!

Hmm.. so let me see. £1500 for a new unit, £450 + lots of postage for a refurb by an indy, or £60. Ooh that’s a tough one.. hmmm…
It may be tough for you to understand, certainly seems that way to me.
Not to mention that you’re focusing solely on cost in GBP and not including the cost of hassle. Again, as always, you won’t be told, as you are correct as usual.

Do you think I’d be farting about with it on the driveway if I had the cash to pay someone else? Let me remind you that all this shit started when I took it to a bloody garage.

Ah!, it’s Molgrips, in Hissy-fit mode… Again.
So, you’re throwing your spanners out of your shed now, cos you own a car you can’t afford to run and repair. That sounds like all kinds of smart, you know, the kind of smart that replaces only the one valve.

OK. Provided it was ever tested, catalogued and published
You think the power train dept for that OEM hasn’t tested a fleet of vehicles to establish just exactly this type of thing. That the bolt didn’t come loose within 150K miles? I work in the Auto Design industry and I can assure you someone knows the torque, exactly.
Personally I’d of approached the G/B supplier for the torque spec.
If I wasn’t too busy ranting on a cycling forum…
😉

Well clearly you have connections, so why don’t you find out for molly instead of being all rude and sarcastic (and wrong – I could pick your post apart and point out where but CBA)? Because if somebody does know, they’re not telling him – are you suggesting he hasn’t tried?

p.s. when you design stuff, do you do so in order to make life as easy as possible for the customer, or in order to make the most money for the company?

honeybadgerx – Member

I think it’s to do with compression/extension of the bolt when you’re tightening/loosening it, so that there is greater friction on different portions of the bolt depending whether it is going in or out. I’ve tried thinking about it in a more exact manner, but I fear my head might explode.
It can be a very complex calculation, the Analysis guys at a place I worked at, about this time last year, just groaned when I mentioned thread torques and years ago a friend of mine working in a specific area of the offshore industry said his company employed a few guys who could do that shit, but they didn’t socialize too well, had poor personal hygiene and strange habits/twitches. But they could calc the thread torque loads and for fun they use to include the effect of paint on the thread, including the shear values for the paint itself.

So it’ll be a walk in the park for Molgrips.

and wrong – I could pick your post apart and point out where but CBA
Hhmmm, so you’ve also visited a power train assembly supplier to see their line and see people assembling equipment at a station using tools pre-set to certain torque values. Nice one. And there was me thinking you were thinking such a supplier left greg the gear box builder to decide for himself what today’s gearbox value assembly, fixing bolt tightening torque should be.

Well clearly you have connections
Don’t you have micro switch contacts to de-carbon with your file. In them olden days we’d of used t-Cut or brasso to polish off the carbon and minimize the surface scratching on the switch contacts so as to offset the time taken for the carbon to build up and once again prevent the switch from operating properly. But, hey, well, as it’s you.
🙂

M5? Just give it a nip FFS.

Hhmmm, so you’ve also visited a power train assembly supplier to see their line and see people assembling equipment at a station using tools pre-set to certain torque values.

Pretty much any assembly line uses tools pre-set to certain torque values, you’d struggle to find one that didn’t.

so a torque value does exist, fact!

Yes, of course it does – but it’s not published, which is what I said.

Not to mention that you’re focusing solely on cost in GBP and not including the cost of hassle.

Er.. well yes, but if there’s no money in bank what shall I use to pay the mechanic? I have free time, so why not use that when there’s no money? Is this really a difficult concept?

Why do you have to turn every thread into a shit slinging fest? Doing your own repairs to save money is hardly an unusual thing, so just let me get on with it, without having a go at me please.

thestabiliser – Member
M5? Just give it a nip FFS.

I don’t think you’ll find that particular measure of torque quoted for anything on say… a Toyota.
But yes, perfect, common sense eventually prevails.
It’s an M5 which should tell one, something.

Standard torque settings we used on swithchgear (a hi high vibration application) in standard 8.8 steel

M10 47Nm
M8 31Nm
M6 13Nm
we didn’t bother with M5 but I bet 13Nm on a crush/spring washer would be plenty.

Hhmmm, so you’ve also visited a power train assembly supplier…[/quote]

<point>
.
.
.
.
.
.
Solo

Just to help I’ll provide a sample:

It’s not published, cos it’s not a recommended repair procedure.

Of course, you are always correct… However, once upon a time a man in a factory built the gearbox you are messing with and the tool set to drive in the bolt in question, did so to a specific torque, so a torque value does exist, fact![/quote]

molly never suggested the torque value didn’t exist…

In them olden days we’d of used t-Cut or brasso to polish off the carbon and minimize the surface scratching on the switch contacts…

You know that I didn’t do that? In reality I needed to test the fix would solve the problems – though if it fails again it will take 5 minutes to fix better (it lasted 17 years before failing, I’m not holding my breath).