Thread: Hammer A3 31 Torx screw problem - won't come out

Hi,
I usually repair old machinery and Bohdan mentioned one of the best ways. You state you have a small Butane torch. These are ideal. Heat the head of the screw red hot,(as in Orange),let cool a bit and when down to Black melt candle wax on it and leave to cool. TC is designed to take heat which is why it has replaced HSS in high speed machining of steels. Use an impact driver the manual kind. Power tools do more damage than anything else other than small hammers. In very rare instances it may be necessary to heat and wax more than once. Working back and forth with a T-driver is good also. If you get any movement slow working is what to do not "It's moving now crank it" as this will probably cause galling.
Someone asked if the head was Aluminium and another mentioned Loctite Copperslip. DO NOT UNDER ANY CIRCUMSTANCES USE COPPERSLIP ON ALUMINIUM! Copper and Aluminium do not like each other, I have seen unrepairable corrosion caused by the reaction.

Originally Posted by FenceFurniture

Yeah I reckon that will be next. Plenty of spares (70+).

I'd give the "heat with the small butane torch" a try before breaking the TC.

I'd put in a call to International Rescue - I'm sure Thunderbirds have come across this sort of problem before.....

Just as I was reading these last posts NCArcher called to say he has a tiny flamethrower, manual impact wrench and a couple of other goodies which he is bringing around tomorrow morning for us to have at it with.

FAB Mr Brush. Thunderbird 43 ¾ is a go.

Good man Virgil, I mean NCArcher.

I'll keep an eye on the TV news for further updates -the "stuck screw" has gone viral

fixed yet?

Just break the damned insert!!!!

It's 1:00 pm and no bloody news, the suspense is killing me.

We have news!

NCA brought his butane burner around so we torched it for...I dunno...2 minutes. The candle wax didn't melt, it just scraped off. "That's a bit weird". Turns out we could still touch it. Back to NCA garage to get the MAP gas which put some serious heat into it.

Bottom line? Bloody thing was still giving a two finger salute, and the Torx hole in the screw head was getting too close to the point of stripping out the splines. I bought a couple of Irwin Torx bits at a Bunnings I was passing on Thursday. The Suttons ones were just shearing at the slightest provocation. The Irwin at least had the dignity to just twist rather than shear.....so it's useless too ( but a little better than the Suttons).

"Rightio, get me a Centre punch!"

Snapped it off and the screw undid with fingers only, so it was clearly just over torqued in the factory. There were large and small chips of TC everywhere so plenty of vacuuming and blow down the machine with CA. I'm sure I'll be finding chips of TC for the next few years.

It looked and felt like there was the slightest deformation of the seat (in the head) so with a 60x loupe and the end of a file we scraped it down a touch.

Put some timber through and there is one tooth that is just a poofteenth too high, but it looks like it is one of the adjacent ones to the problem (removed two each side before heating and snapping). I'll take those back off and see if there is anything on the seat, but it's so minor as to be not a big deal.



Thanks to Tony for his assistance!

Good to hear that you have a decent result, Brett! We were all holding our collectives breath in united mental effort.

When turning bits, I am fanatical about removing the bit and then cleaning the surface thoroughly before replacing it. The margins for error are tiny. All is tightened with the torque driver I posted earlier. That way everything is equal. Sounds more effort than it really is.

Regards from Perth

Derek

So I decided to test what Nm the clutch drill puts on, as close as can be determined.

I was using clutch 9 for tightening, and the same setting would actually undo the screw as well most of the time.

Using Chris' torque wrench set to 5Nm (which is the very lowest setting possible), I tightened up a screw until the wrench popped. Then marked the screw head, using clutch position 9, the head rotated very slightly - perhaps ½-1mm on the circumference of an 8mm diameter screw, which translates to roughly 10° movement. I suspect that clutch setting 8 will be pretty close for tightening, but it should be noted that a screw tightened with 9 can be easily undone with a hand wrench that has a handle length of only 100mm. So in other words it is not super tight at all using a power drill on clutch setting 9. AND clutch setting 9 will undo what it did up, so I really am struggling to see what is wrong with the clutch drill method. I actually see nothing wrong with clutch 9, even though 8 will be closer to 5Nm - in other words I think 5Nm might be a bit light.....

It's all about not over-tightening the screws so that they can become jammed on (like the pain in the screw that I had - done in the factory no less). So if a clutch setting of 9 can be undone easily by hand with no straining, then I fail to see how there is anything wrong with the technique - they are not over-tight at all.

Furthermore, Elan has done hundreds of these with a power impact driver, and the torque setting on that is much higher (but totally unknown and arbitrary).

Originally Posted by derekcohen

Good to hear that you have a decent result, Brett! We were all holding our collectives breath in united mental effort. Indeed! Please, everyone, relax!


That way everything is equal. Yes, but I believe mine are all equal tension too.

Originally Posted by derekcohen

When turning bits, I am fanatical about removing the bit and then cleaning the surface thoroughly before replacing it. The margins for error are tiny.

Yes Derek, the slightest bit of dust on the seat can translate to a noticeable groove in the surface of the board. I blew the seat (breath, not CA), and wiped the side of the cutter that was about to register with the back of the seat (i.e. so it didn't get pushed forward and therefore raised). I was very pleasantly surprised that with the exception of the problem cutter, everything else was like glass again.

The small butane, pinpoint, torch did nothing. The mass of the head just soaked all the heat up. A few seconds of the MAP torch however put a lot of heat into the head which stayed warm to the touch for quite a while