Drill out a ht c/s allen screw

[JimmieSParticipant @jimmies]

[JimmieSParticipant @jimmies]

Advice on drill for drilling out six 5/32" high tensile c/s allen screws in an alloy c/case.. Heating is not an option as the allen key holes have been distorted.

[Michael GilliganParticipant @michaelgilligan61133]

The core is usually quite easy to drill [a good hint being that the sockets are already damaged] … If you drill to the screw shank diameter, the countersunk heads should just pop-off.

Then [depending on the design of the component] hopefully this will leave sufficient shank to get hold of with suitably sized 'Mole Grips'.

The threads may not be seized … it's often corrosion in the countersink that causes the problem.

MichaelG.

[Steve PaveyParticipant @stevepavey65865]

Personally I’d go straight in with a carbide end mill, either on the mill or a pillar drill, with the part firmly bolted down to the table. In the past I’ve had to remove lots of fixings in aluminium (mostly 316, but that can get quite hard if someone else has been playing before I get hold of it), and always found it best to hit it with the best you’ve got right from the start.

Edited By Steve Pavey on 21/11/2017 16:29:40

[David George 1Participant @davidgeorge1]

Have you tried to peen the holes in the screws with a punch round the hex hole to make the hexagon in the screw smaller. I have a punch with a domed end which is very useful for this. Then take a T Allen key and hit it into the original hex with a hammer or mallet and unscrew as usual. The shock of hitting on the end loosens the screw as well as reforming the hex but replace with new screws. it rarely fails to undo such screws especially on engine covers on motorcycles etc.

David George

[colin hawesParticipant @colinhawes85982]

You can drill HT screws with a HSS drill provided you keep the drill sharp, don't let it rub and use a slow speed.Colin

[Anonymous]

Before you drill try hammering a torx bit into the hex, big enough to bite into the head , hammering helps to shock it loose a touch and then use ratchet or impact driver on the torx.

[Don CoxParticipant @doncox80133]

Most Allen keys are quite tapered at the end due to the manufacturing process, so in cases such as this I usually grind the end of the key flat and square to maximise the contact area between the key and the walls of the screw socket.

Don Cox

[john carruthersParticipant @johncarruthers46255]

I had success with an alum solution, it loosened it up enough to get it moving.
Otherwise, as suggested above, carbide mill/drill; or spark it out if you have access to an eroder?

[HopperParticipant @hopper]

Sometimes a dab of coarse grinding paste on the allen key will give it that little bit of grip it needs to turn the screw. Sometimes!

Otherwise, you can buy lefthand drill bits made for exactly this purpose. If you can drill out to a bit less than the root diameter of the thread, the lefthand drill bit, being run counter-clockwise, will often unscrew the remaining carcass of the screw. Just run the drill slow for the hi-tensile material.

[J HancockParticipant @jhancock95746]

What is essential is that the core drill goes exactly down the centre of your broken screw.

You should make every effort to ensure this happens.

[VicParticipant @vic]

Posted by Michael Gilligan on 21/11/2017 16:20:44:

The core is usually quite easy to drill [a good hint being that the sockets are already damaged] … If you drill to the screw shank diameter, the countersunk heads should just pop-off.

Then [depending on the design of the component] hopefully this will leave sufficient shank to get hold of with suitably sized 'Mole Grips'.

This is what I did with some M4 C/S socket screws that had self tightened on a component.

[Martin 100Participant @martin100]

Posted by JimmieS on 21/11/2017 16:06:14:

Heating is not an option as the allen key holes have been distorted.

Ignoring that for now, a sacrificial hex key, tacked into place with a weld (stick or mig) keeps the key in place and thermally shocks the threaded zone. Use anti-splatter spray on the surrounding area to protect the casting. Oversize the hex key (4mm) or use a circa T25 torx bit tapped into the recess if necessary.

[CirclipParticipant @circlip]

Assuming you can drill the heads off first, before ravaging with Mole crushers, application of heat and dowsing with hydraulic/penetrating oil may help to break the Aluminium Oxide corrosion.

Regards Ian.

[Ian PParticipant @ianp]

I recently removed ten 2BA countersunk Allen screws from an aluminium casting that had been in a marine environment. I used a worn down (so small diameter) cutting disk in a Dremel type tool to cut screwdriver slots in the heads. I unscrewed all of these with a well fitting flat blade held in the drilling machine chuck so that I could apply axial pressure.

I ground the driver blade to a similar radius as the bottom of the groove to get the maximum engagement. I also initially used a hammer on the bit to try and break the joint surfaces.

I dont think I would have succeeded with a hand held screwdriver so this method might not suit your screws.

Ian P

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Martin 100 on 22/11/2017 09:57:29:

Ignoring that for now, a sacrificial hex key, tacked into place with a weld (stick or mig) keeps the key in place and thermally shocks the threaded zone. Use anti-splatter spray on the surrounding area to protect the casting. Oversize the hex key (4mm) or use a circa T25 torx bit tapped into the recess if necessary.

.

Martin,

I may be wrong, of course, but … I think you will find that the opening post refers to 5/32" countersunk screws.

These would surely have significantly smaller hex-socket than cap-heads of the same thread size.

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Circlip on 22/11/2017 12:08:07:

Assuming you can drill the heads off first, before ravaging with Mole crushers, application of heat and dowsing with hydraulic/penetrating oil may help to break the Aluminium Oxide corrosion.

Regards Ian.

.

Ian,

I take your point, but … the last time I did this [albeit with a cheese-head screw down a counterbore] the screw remnant was only just above finger-tight … The corrosion under the screw head typically creates a lot of preload; which is what prevents the head turning. … When the head comes off, so does the preload tension.

MichaelG.

[JimmieSParticipant @jimmies]

What a wonderful group of ‘workmates’ I have on this forum, freely providing advice with no thought of reward. Thanks to all concerned. Have tried all the suggested bench work methods without success – except for braking two new drills despite using slow, steady speed with coolant – so will buy some carbide drills. I am now wondering if the screws are indeed h/t.

The engine is from a Bantam racer and had been put together with a lot of care and attention, eg plain bush replaced with needle rollers, homemade short stroke crank, shortened barrel, etc. The screws are securing the c/case packing and need to be removed as I intend using an ordinary crank. The conrod is Yam RD400 which indicates it may have been built in the South West area. Could the engine builder have worked in the aerospace industry with access to high spec fasteners. And adhesive?

Jim

[Gordon WParticipant @gordonw]

Crankcase packing screws were always fitted with some sort of locker, could be very strong. May need heat to shift. I used to work with a chap who was bantam racer, did quite well in the TT years ago.

[Martin 100Participant @martin100]

Posted by Michael Gilligan on 22/11/2017 12:45:20:

Martin,

I may be wrong, of course, but … I think you will find that the opening post refers to 5/32" countersunk screws.

These would surely have significantly smaller hex-socket than cap-heads of the same thread size.

MichaelG.

Possibly, but I wouldn't be so sure the hex recess is by design actually smaller across the flats, It certainly isn't to my knowledge on metric fasteners. The recess depth is however slightly smaller at least at smaller thread sizes.

The first time I encountered this was on iirc M6 countersink screws in some high voltage air blast switchgear in the mid 1980's The 'usable' depth of the recess in the countersunk screws was something like 10% less than a comparable cap screw with the same thread but with only around 3mm recess on the screw anything other than a perfectly fitting hex key either with the formed end ground perpendicular by the end user or a perfectly formed relief at the bottom of the hex recess to match the off the shelf hex key would compromise removal from the start.

Despite the application of considerable end load damage to the hex recess often occurred with cam out and it made removal even with perfectly sized hex keys extremely difficult. Even more so when the orignal factory installation methods were 'dry' into the aluminium and with either uncontrolled torque application or overtorquing with slightly damaged / worn hex bits.

By using the welding method together with a tiny amount of specialist anti-seize* on the threads on refitment it ultimately saved a few million quid over the following years in significantly speeding up overhaul times and contact replacement.

* I think it contained glass beads as everything else such as oil, grease, nickel or copper was off limits.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Martin 100 on 22/11/2017 16:51:56:

…. I wouldn't be so sure the hex recess is by design actually smaller across the flats, It certainly isn't to my knowledge on metric fasteners. The recess depth is however slightly smaller at least at smaller thread sizes.

.

I can't [and wouldn't wish to] argue against your experience, Martin … but these charts are consistent with my own: **LINK**

http://www.roymech.co.uk/Useful_Tables/Screws/cap_screws.htm

MichaelG.

[Martin 100Participant @martin100]

Just checked MG and you are correct about the hex across flats size changing between cap and countersunk sizes 5mm and 4mm respectively for M6 and I quite clearly recalled incorrectly.

That change in across flats size clearly has an impact on the recess depth.

I know at the time the recess on an undamaged screw was measured and checked against a British Standard (BS 4168-9:1983 would cover the period in question) and it was compliant.

[Michael GilliganParticipant @michaelgilligan61133]

Thanks, Martin

MichaelG.

[JonParticipant @jon]

Posted by JimmieS on 22/11/2017 14:46:59:

What a wonderful group of ‘workmates’ I have on this forum, freely providing advice with no thought of reward. Thanks to all concerned. Have tried all the suggested bench work methods without success – except for braking two new drills despite using slow, steady speed with coolant – so will buy some carbide drills. I am now wondering if the screws are indeed h/t.

The engine is from a Bantam racer and had been put together with a lot of care and attention,

Jim

Assume your still having trouble its very easy.
If the hex is still in tack but rounded your easy on, all you need is a few decent drills carbide not necessary even for 12.9. More important is the quality of drills used, not to generate much heat and pressure by feel.
5/32" = 3.95mm. Initially start off with a suitable 'decent' drill that will remain centred by continually opening up the hex to underside countersunk taper. Now you have the depth from first drill used just keep opening up the hole youve just put down the hex to around 3.6mm.
Last drill sub 6mm dia should be able to just peck head off from the 5/32" bolt or screw.

Have even done this blind on turbo to manifolds with a right angled air drill, works every time if keep the hole central, thats the key.

If its not cutting dont force it, drills had it.

[J HancockParticipant @jhancock95746]

If the 'core' drill goes out of centre and breaks into the aluminium you will be in big trouble.

Go the extra 'mile' to make a little jig up to make sure it does not wander.

[Author]

Posts