Whilst sitting bored - still in lockdown in Scotland I thought I would think more about something raised in another thread regarding preference over TORX or Hex keyed cap screws. This is my take on the subject and why I’ve moved to TORX keyed bolts both in stainless and titanium. I know some swear by hex which is fine and I’ve never had a problem with high tensile hex keyed cap screws but moving to stainless mostly for cosmetic reasons and titanium for cosmetic and safety critical fasteners like disc bolts have caused me no end of problem using hex keyed cap screws? I’ve never experienced any failures but I’m sure some have. Never had a problem when fitting and torque but when trying to remove they invariably round and the air turns blue? Moving to TORX keyed Cap screws have never caused problems. My hex keys are snap on so should be a good fit but here’s a pic of what happens with a 0.1mm difference between key and head. In the case of the hex when torsional force is applied its only the apex of each point that presses on the bolt and easily rounds both stainless and titanium bolts in my experience. In the case of TORX a greater area is covered reducing the risk of rounding. Basically the TORX head is more forgiving. Anyway here’s some pics.
I agree with all you say apart from snap on Allen sockets or Allen keys being a good fit. I have mainly snap on tools and most of their stuff is top notch but the Allen sockets in particular wear incredibly fast and are much too soft. Give me a Stahl Wille one of those any day and most of the rounding off issues are a thing of the past. I prefer the bolts in picture over both types you compared. If only they were available in metric.
+1 on Snap-on Hex keys / sockets being completely shite and nothing like the quality of their other tools. Facom, or Wera for hex , and all good What brand are your torx tools out of interest ?
Have sets of several makes of both hex and TORX but not really the point I’m trying to make. Again just my observation from my experience and prefer TORX . Regarding the bi hex type mentioned I’ve purchased fasteners from ARP which supply both imperial I use on my British bikes and metric. I’ll probably purchase this set for fixing callipers to fork legs. They are a special stainless. Bolts fitted are 8.8 Grade 5 with a tensile strength of 115,000 psi. The ARP bolts are rated 170,000 psi. https://autopeers.co.uk/arp-773-1002-bolt-stainless-steel/ Head size outer flange too big but I’ll turn down to fit. I’ve posted under technical help maybe that’s the problem and I’ve posted under the wrong section. Not really looking for help just a simple observation.
A couple of observations, the contact patch between the tool and the bolt head is larger for a torx, also the radius from the centre of the bolt to that contact patch is smaller and therefore the tool is less likely to twist out, although more pressure/stress at the contact patch would be required to undo the bolt for the same torque applied at the end of the undoing tool. If ever I get a hex head that appears stubborn and in danger of rounding off I put some fine valve grinding paste into the head and it takes up any slack. It is a pain to get the paste out afterwards but, if the paste has been used, that bolt is usually scrapped anyway.
If you use quality keys, not the tat sold on ebay, you are not likely to have issues removing either type. I have a variety of types and have found that when a bolt has been trying to round out using one make of key, another will bite and remove it OK. I should really throw all my crap stuff out - but dont have a full set of proper stuff. The cheapos work most of the time so I persever, just try to stop shy of rounding out
Geoff, have you considered the changes the drive point radii make to the tightening torque? For example if the torque setting required on the fastener is specified with a driving radius of 10mm for a hex key the driving radius of the torx equivalent will be say 9mm. Does this mean that there will be a tendency to over-torque or under-torque the screw. The answer is most likely within the safety critical torque tolerance but if you do the maths the answer is not necessarily obvious
I think all the drive points/loads of keys or sockets all exceed the torque that is required for a given thread size. The only issue for overloading the drive capabilities, is when you undo a seized fastener.
Not so. The Torx/Hex/whatever is only the method to transmit the tightening torque. Imagine you are tightening the bolt from the central axis (i.e. with 'pure' torque); the radius of any connections in place between the pure torque and the bolt thread is irrelevant (they cancel each other out if you see what I mean). I haven't explained this very well (it's crystal clear in my mind) - try to think of this 'pure' torque and go from there...
Clear as mud. I'm sure I learned that in physics. Torque is a tricky thing. The clamping force and therefore the tensile load on a bolt changes also significantly with the type of lubrication on a thread. I once did a test to figure out the torque at which a M8 stud of 10.9 quality would plasticly deform. The result was how long is a piece of string? Bone dry it was 42 Nm. Lubricated with light oil it was 26Nm. Sometimes a certain lubricant is specified and should be strictly adhered to otherwise you might over or under-torque.
