internal screw cutting

[Martin Botting 2Participant @martinbotting2]

Hi all.

As a daily new bodger I have run into a problem thats vexing, namely screw cutting on my ML7. I was trying to make an ER32 chuck for my lathe and using a mixture of words and music of Mr H Hall and the step by step video guide of John aka doubleboost on Youtube and I have so far messed up two billets.

The problem I have is after boring out and putting the relief groove then the trouble starts, I do the scratch cut fine, cross slide out wind back cross slide back in, cut on. wait for the number to come round on the tread dial,Clunk in with the half nut off we go again… on the 3rd pass I notice I am making a second thread! D*$n n blast ( better than that in reality)…..

Now I found the apron was moving and on closer inspection the caps screws were rocking! threads buggered, Hellicoil got it clamped down tight, I renewed the bronze bushes as well while it was off so no more floppy hand wheel…

Next thing I noticed was the lead screw does have some slop in the horizontal plane could this be the problem? If so would some kind soul tell me roughly how much slack there should be in the lead screw, if none how best to cure it. Or am i best suited to needle work…. PLEASE HELP!

The firm I work for might notice the bright mild bar stock is shrinking, but the steel making industry will be giving me man of the year award.

Many thanks in advance…

[Martin Botting 2Participant @martinbotting2]

[Ady1Participant @ady1]

Are you working in Imperial? TPI? Metric?

What's the lathe and banjo setup?

Sometimes using the tailstock as a backstop helps, wait for your dial to come round then throw in the nut/clutch

edit

sounds like your machine is a bit of a clunker too

everything needs to be ship shape to do decent screwcutting work

Edited By Ady1 on 10/06/2014 00:53:00

[John McNamaraParticipant @johnmcnamara74883]

Hi Martin

Hmm….. "wait for the number to come round on the tread dial"

The ER usually chuck has a metric thread. Is your lathe Metric? The numbered indicator dial on the saddle only works on imperial threads. Not most, or is it all? metric threads.

Maybe this is your problem.

If you are running under power into a run-out groove it can be a problem if you are not quick enough to stop the lathe it will crash into the end I am assuming you are making the collet closing nut for the chuck, you may have to turn the chuck by hand. because you cannot release the half nuts or you will loose the positioning.

There are workarounds using marks on the chuck and the bed to re position, I cant find a link at the moment. The hand method always works, for a small run of thread not that hard, keep your cuts light or you will work up a sweat!

BTW. Commercially made collet closing nuts for ER chucks are readily available, for the work involved in making one worth considering.

Regards
John

Edited By John McNamara on 10/06/2014 01:31:59

[Ady1Participant @ady1]

The ML7s greatest weakness was no leadscrew clutch IMO. Beats me why they never included one

[Martin Botting 2Participant @martinbotting2]

I should have said I was attempting to cut the internal spindle nose thread 1 ⅛th" X 12tpi of the Whitworth form (please forgive me Mr Whitworth for i know not what I do)

The lathe is imperial and the job was being held in the four jaw chuck. I was taking it all very slow and made sure the overhang of the screw cutting tool was at the bare minimum.

[Roger HartParticipant @rogerhart88496]

Try using a mandrel handle. Then everything happens at a sane speed and is much easier to see (and stop) when things are not right. I do most of my smaller screw cutting by handle. I will even admit to keeping the clasp nut closed and winding out – did a 4 start thread that way – fairly easy.

[Martin Botting 2Participant @martinbotting2]

My thoughts were going down the same route with the use of the mandrel handle (nice ring to that) and winding it backwards keeping the half nuts engaged, theoretically the tool, which is set over to the prescribed angle will stay in the same grove, yes?

The slop (understatement) in the lead screw must need taking up, is there a suggested maximum slop for a lead screw?

[RobCParticipant @robc77385]

I don't know if your ML7 has the ability to reverse the motor direction, but I found that leaving the half-nuts engaged and winding the carriage back in reverse (slowly), after moving the cutter away from the job, was slow but consistent.

Alternatively, make sure you move the carriage a reasonable way off the job, so that any backlash is accommodated.

Using your slowest back gear is also helpful, while threading internally – it gives you more time to react.

As Roger mentions, making mandrel handle is a useful thing. I am just finishing up the one designed by GHT from a kit of bits purchased from Hemingway kits (usual disclaimers). It's a neat little exercise in construction too!

[Rick Kirkland 1Participant @rickkirkland1]

Hi Martin, try and adjust the thrust collars on your leadscrew so that you have no more than two or three thou of logitudinal play.

Rick

[Gordon WParticipant @gordonw]

Never had a Myford but all lathes work the same. Make sure everything is adjusted correctly and do the screwcutting by hand. I made my mandrel handle from an old flywheel, find it easier to use than a plain handle. Once you get going , for a short thread, it is just as quick as power cutting.

[John McNamaraParticipant @johnmcnamara74883]

Re leadscrew play:

The big gremlin can be cam action of the collar(s) and Thrust bearings that retain the lead screw moving the whole screw sideways back and forth. ideally it should be under .0003" (Three tenths of a Thou) Max

You can set an indicator on the end of the screw and check it under power while moving the saddle to provide a load. It can be a bit fiddly to get a reading if the end of the screw is not nicely finished and clean,

Any wobble is going to be replicated in your threading.

Regards
John

Edited By John McNamara on 10/06/2014 09:51:10

[HopperParticipant @hopper]

Posted by Ady1 on 10/06/2014 01:37:15:

The ML7s greatest weakness was no leadscrew clutch IMO. Beats me why they never included one

Because then it would be as good as a Drummond, which was a more expensive machine.

Martin, you might try some external screw cutting tests on a piece of scrap bar to get things set up right before you continue chewing up 'billets' to make the internal thread.

There should not be any discernible up and down slop of the lead screw. Maybe a thou or two, but if you can jiggle it up and down, the bushings/bearings at each end need looking at. If it is flopping about the place it may not be meshing right with the half nuts every time. Check also the condition and alignment/adjustment of the half nuts.

[John McNamaraParticipant @johnmcnamara74883]

Oh yes and the half nuts…..

Not always easy to see sometimes impossible to see are they clean? really clean. A bit of swarf trapped in the threads is all it takes to make messy threads. And yes are they set to close properly?

[speelwerkParticipant @speelwerk]

Posted by Ady1 on 10/06/2014 01:37:15:

The ML7s greatest weakness was no leadscrew clutch IMO. Beats me why they never included one

If you have one with gearbox you can retro fit a leadscrew clutch, works fine. Niko

[frank brownParticipant @frankbrown22225]

If you engage the half nuts on your mark a couple of inches out from the start of the cut, any backlash in the lead screw will be taken up as it "drags" the carriage along the bed. i.e. the lead screw will be in the same position at the start of the cut.

Best thing to do is not to disengage the half nuts. However if your lathe has no reverse and you don't fancy turning the spindle backwards for twenty turns, there is another way.

Set up a back stop for the saddle (piece of MS bar against the tail stock?), put a mark on the chuck. So you are ready, wind the saddle back against the stop, when the mark on the chuck comes up to to the top, look for your number and engage the half nuts. The saddle position will be locked to the gear train like it was last time. At the end of the cut, disengage the half nuts, retract the tool, wind the saddle back for the next cut.

I invented this method when I was cutting a 5 1/3 TPI thread, the first two cuts were great, the third was a new thread. i believe its because of the fact that your number can drop into the lead screw at any position so for a 4 TPI lead screw, its every 1/4" in travel, the problem is if you thread is not a sub multiple of 1/4", the angular position of the chuck can be anything.

Frank

[Clive FosterParticipant @clivefoster55965]

Have you made the matching external threads yet? If they turned out OK use exactly the same technique inside out and things should be fine.

If you haven't yet made the external ones, following the usual (good) advice to do the inside ones first so the easier external ones can be cut to fit just so, grab a bit of scrap and try an external thread so you can see whats going on. Hopefully that will make the problem visible. Consider wrapping a sticky paper label or similar around your test piece after turning it true and use a sharp pencil as your lathe tool which should clearly show you exactly how much the mis match is.

Franks back stop idea is good. Maybe stop the lathe at the end of the cut, bring the saddle back to the stop and turn the chuck by hand to align leadscew and half nut. Mark the chuck so you can verify that half nut engagement occurs at the same rotational position and check all the thread dial positions you propose to use. I'll confess to mostly using full thread dial turns, waiting until it comes back to the same line each time. One less thing to worry about.

If this sort of problem isn't quickly solved by checking the usual suspects I find it better to set-up for a comprehensive careful one thing at a time, slow enough so I can see whats going on, verification. Going straight from basic checks to full monty is usually quicker than steady escalation and on an older machine often uncovers various not quite rights before they turn into mid job problems.

Lack of leadscrew clutch on the ML7 is one of those price / performance / specification balance things. If everything is included the beast is liable to be too expensive for the target customer. Sometimes the knock on effects of an apparently simple change can add up quite seriously. A simple leadscrew clutch requires proper attention to leadscrew and drive shaft support for best performance. The Drummond is inadequate in this respect for a post war machine.

Clive

Edited By Clive Foster on 10/06/2014 11:10:52

[Neil WyattModerator @neilwyatt]

12 tpi could be an issue with the 8 tpi leadscrew on an ML7. one screw turn is 1 1/2 threads, so you can only engage the leadscrew with the indicator on the even-numbered positions or you will get a double thread. The safest way 9if you have patience) is always to engage on the same number every time (this works for any thread that is a even number of 64ths of an inch in pitch withan 8-graduation dial and an 8 tpi leadscrew).

Incidentally, Martin Cleeve fitted a threaded rod to one of the half-nuts on his Myford. this ensured repeatability of the amount by which the nuts closed, improving screwcutting accuracy.

If you can find a Chinese mini-lathe to copy, they have this arrangement fitted from new

Neil

[Roger HartParticipant @rogerhart88496]

Just to add re use of handle. I find you cannot just wind back out, it is necessary to move the tool away from the work a bit then wind out then reset the tool position. I clip a terry-clip to the cross slide collar to remind me where I got to and slide it back up to vertical when I have added the extra cut.

[Nigel McBurney 1Participant @nigelmcburney1]

I was taught during apprenticeship,before screwcutting,clean the leadscrew with stiff brush held against the screw while it is rotating,flood it with oil then engage the half nuts and run the saddle up two or three times,this will get most of any debris out of the halfnuts ,then give the leadscrew another brush, if the lathe tool follows the initial tool mark once and then goes haywire on what is the third cut then either the half nuts were engaged in wrong position due to misreading the threading indicator(easily done) or engaging the half nuts on the top of the thread of the leadscrew,it has been known to happen ,the soft material of the half nuts will sometimes just catch thread top and move the saddle and cut the work if a very light cut is being taken,an operator should be aware that the half nut lever is not fully down but accidents do happen. Some endplay on the leadscrew will be taken up as soon as the saddle moves so will not cause the problem. I suggest beginners engage the half nuts when no 1 shows on the thread indicator,takes longer but stops errors. I assume the top slide is adjusted fairly tight or locked.

As an instrument maker ,a lot of threads were produced using chasers on plain instrument lathes with a threading attachment, and if anyone was threading say on a Boxford ,to save grinding a screwcutting tool a chaser would be used instead, of course one had to be very quick with both hands , the chaser has to be retracted from the cut just before the half nuts were disengaged , I once saw another apprentice get it wrong and disengage the half nut first resulting in a series of parallel V shaped rings around the work, !! I still use chasers for screw cutting (I bought a lot for £2 each some years ago ) at least they produce a full form thread and with care 8 tpi and smaller can be cut quickly in mild steel , bit tricky on bores ,getting the right depth and one hand on the half nut lever and the other on the crosslide but you do get nice threads with the correct form.

[Martin Botting 2Participant @martinbotting2]

Thank you all for your help, I have just got home from work with two bits of 2" round BMS that just happened to get caught in my work bag and will go through all your suggestions and blow the dust off the mangle handle I made a long while back and sit down the workshop and give the leadscrew a looking at and a talking to.

Again many thanks to you all, to be able put it into words that I can follow is a great help, for someone that only had a secondary education with engineering as a 1 hour lesson a week trying to recall all the terms and phrases to try and describe something is hard, and your suggestions will be followed thanks again all!

[Neil WyattModerator @neilwyatt]

Nigel,

Your comments ring true, I tried out thread cutting with a chaser for a brief review in the current MEW. I found it was easiest for me to use my e-stop (coast to stop on my inverter) which stopped the work within less than a turn under load, then retract, disengage and wind the saddle back..

Yes, half way through the first cut I thought "what happens next?" and resorted to disengaging the half nuts. Fortunately it was just a 'scraping' cut to check I had the right gear setup.

Neil

[Clive FosterParticipant @clivefoster55965]

Neil

Have to wonder just how repeatable the "stops in less than a turn is". If it really is very repeatable maybe a bed stop incorporating a microswitch for automatic stop at repeatable position is a practical proposition. In principle a system with some sort of spring loaded actuator device able to move past the switch without damage if things don't work could be quite simple. A plausible concept would be a double diameter rod made so that one diameter safely passes the switch when the button is up and the other when it is down with a short cone between the diameters to operate the switch. I have a salvaged device of this style, but with two switches, in one of the it will be useful one day boxes.

Something of this ilk, but incorporating a 6 position turret with individual micrometer stop adjustment, is on the to-do list for my Smart & Brown 1024 VSL. (Unusually for a high end industrial toolroom lathe the the 1024 VSL has no spindle drive clutch relying on simply switching the motor on and off.)

A further thought for those without VFD drive is to use said switch to control an electromagnetic clutch off a car air-conditioning compressor interposed between to the motor shaft and drive pulley. I have a feeling such use of an air-conditioner clutch has already been suggested in MEW some time ago.

Clive

[Neil LickfoldParticipant @neillickfold44316]

Posted by speelwerk on 10/06/2014 10:02:38:

Posted by Ady1 on 10/06/2014 01:37:15:

The ML7s greatest weakness was no leadscrew clutch IMO. Beats me why they never included one

If you have one with gearbox you can retro fit a leadscrew clutch, works fine. Niko

What does this do???

Thanks , Neil

[speelwerkParticipant @speelwerk]

It is a clutch that grips the modified large wheel, with it you can run the slide against a fixed point without cousing damage. With screw cutting it is of no use and you change it for a normal large wheel. You can find a few more pictures in my album. Niko.

[Author]

Posts