Wonky tapped threads

[MWParticipant @mw27036]

Alternatively have you tried just using a new tap? they don't last forever, and I've had a few dogs every now and again.

Michael W

[Dave HalfordParticipant @davehalford22513]

I have an ancient plain lathe that had miss-aligned centres, you could watch the drill wobble this was blamed on bed wear.

Then one day I snugged up the tailstock barrel clamp and suddenly everything trued up.

A few thou at the tailstock will be amplified first by the chuck , then the drill.

[FarmboyParticipant @farmboy]

Posted by Phil Boyland on 27/02/2018 08:09:07:How I first noticed the problem was when facing off then drilling/tapping a hole, then fitting the part to a mandrel to turn the outside on a full taper for example. It would not being running true, despite when in the chuck previously to have no noticeable deflection. I accept tapped holes & bolts etc are not precision bits of kit, however I would expect them to fit better than they do.

I wouldn't expect a screw thread to provide accurate location on a mandrel. You really need a plain unthreaded mating section, on component and mandrel to ensure concentricity, and a faced shoulder on the mandrel fitting against the faced area on the part.

[Phil BoylandParticipant @philboyland37326]

Yeah was thinking I could partially tighten the barrel clamp and see if that helps.

I agree RE. the threaded type mandrel, however, a tapped thread should give you a reasonable location I believe. The wobble I am seeing is milimeters out.

I've tried new and old taps by the way and does not make a difference.

Have a few things more to try this eve so hopefully might get an answer soon.

Thanks everyone for throwing out ideas, much appreciated,

[not done it yetParticipant @notdoneityet]

I would second farmboy on threading, for a mandrel. Likely errors on both male and female threads.

Does the threaded mandrel bottom out, or is there a wide faced section to stop the thread pulling off-square?

Seems like poor methodology is at least part to blame.

[FarmboyParticipant @farmboy]

A few thoughts come to mind:

Surely, any misalignment that produces millimetres of wobble would break the tap, or at least the wobble of the tap would be visible as it cuts the thread?

Can we safely assume you have clocked the piece true in the chuck before drilling? Although if the turning, facing, drilling and tapping are done without removing the workpiece everything should be concentric anyway.

Have you tried taking the part off the lathe and tapping it by hand? That might prove whether the lathe is inducing the wobble.

I feel the drilling must be at the root of the problem, a tap can only follow the existing hole. If it were forced to deviate by millimetres it would only be cutting on one side of the hole, and would surely break, as above.

[Dave HalfordParticipant @davehalford22513]

Posted by Farmboy on 27/02/2018 15:12:43:

I feel the drilling must be at the root of the problem, a tap can only follow the existing hole. If it were forced to deviate by millimetres it would only be cutting on one side of the hole, and would surely break, as above.

I can tap a hole on the pi## you just haven't tried hard enough and it doesn't break :O)

[Neil WyattModerator @neilwyatt]

I feel that a wobbly tailstock barrel can't help matters.

Neil

[Phil BoylandParticipant @philboyland37326]

Ok bit of an update. Checked tailstock barrel wobble at various positions and the most I got was 0.002". I think when I did it before I had'nt clamped it to the bed. I drilled a hole last eve and noticed that it is slighlty bigger than the drill, so a 5mm hole gave a tiny bit bigger, so possible the wobble is coming from a shallow thread? Not noticed before but when the drill starts cutting it does wander a slight amount till it starts cutting, worse of course with thinner drills. I always start hole with a centre drill. Tempted to try a spotting drill but not convnced that would stop the wander.

i tapped a hole with first tap in the drill chuck last eve as was suggested a few posts back and that did produce much better results.

So my conclusion is slightly off size holes and my gorilla style hand tapping are the issue. Problem can be almost elimianted by taking care and using Tap N°1 in drill chuck. When possible I may try an undersize drill to get a tighter fit. Getting somewhere anyway I think

[HopperParticipant @hopper]

Whoa, don't drill smaller holes. That would be the Number One cause of broken taps. An oversize hole will not cause thread to go "wonky". The tap will still follow the hole if held correctly. A drill will always drill a few thou oversize and this is taken into account when making up charts of tapping size drills. Best charts to use are Tubal Cain's in his ME Handbook or the one on drilling and tapping (WS Practice series). As he points out, you can go down to as little as 65 per cent thread engagement with hardly any loss of strength. The slightly larger holes make tapping much easier and probably therefore more likely to run true due to less force used on the tap.

Sounds like you have got it sussed with holding the tap in the TS chuck, with TS clamped down.

[Phil BoylandParticipant @philboyland37326]

Ah ok, wont do that then

I will do more testing this eve, got bloody cold last night so came in from the man cave, but confident the collective have nailed down the problem.

I should have said first off I am using the correct tapping drills to, not just a close match. That was drilled(pardon the pun) into me during apprentice training days.

Thanks all for contributing, Much happier now. As daft as this sounds, I don't recall ever fitting anything to a tapped hole that I made during my training, so this phenenom most likely existed then, but just never noticed as I was a spotty teenanger with my head in the clouds.

[Ian S CParticipant @iansc]

Currently there is a series of articles in"Model Engineer" Lathes and more for Beginners by Graham Sadler, I'v been reading through part 6 in ME 4579, I think that would help a lot of people having trouble with starting holes in the lathe.

Ian S C

[Author]

Posts