How to remove?
|Paul Fallert||03/07/2020 16:43:01|
|86 forum posts|
You might try making an aluminium nut and hacksawing it in half longitudinally. Apply one half to the screw and you will see where the cutter needs any "minor" correction.
Being able to SEE how your thread fits the screw is a big help. I was surprised to see the mismatch on the first one. Yet, a quick test would have me believe that it was "good", but the contact area when cut open was very slight and would have become sloppy very quickly.
I made three before I did brass and the brass nut was a really good fit. Another dodge, assuming the screw is worth keeping, is to make a nut that best fits the worn area of the screw and use it to lap the less worn area of the screw. This avoids the loose/tight problem, but only if we are talking about slight wear of the screw. One "safer" lapping compound brand is Timesaver, which comes in grades and stops lapping after a certain point. Of course, use a very fine grade, frequent testing and meticulous removal of the lapping compound. When you have the threading process working, make a few extra brass nuts. It takes very little time passing before you forget how to set-up and cut a proper left hand Acme thread.
|Keith Long||03/07/2020 17:04:09|
|837 forum posts|
Possibly getting right and left hand mixed up?
|Andy Carlson||03/07/2020 19:00:37|
|258 forum posts|
A pretty fundamental incorrect assumption that I've had for so long that I didn't question it.... Oops!
No great drama since this is a test and the tailstock feed screw allowed me to do at least some checking for fit.
I'm glad now that I'm doing this one step at a time - could have been expensive if I'd gone out and paid for the wrong threaded rod.
Incidentally I also thought I'd try checking it on a more worn feed screw. I took the screw out of the 'spare' tailstock... and found that Faircut had changed to 8TPI in the tailstock by the time that one was made.
Still waiting for silver steel...
|Andy Carlson||03/07/2020 19:08:28|
|258 forum posts|
Thanks Paul. Sounds like a good tip. I might do that with the 'wrong un' - like I said, it's a learning experience.
I'm not sure I can achieve minor corrections though - I really need a better setup for less 'agricultural' grinding jobs. I'm still looking for the 'putting metal back on' wheel too. I'm sure it's around here somewhere.
|old mart||04/07/2020 14:30:50|
|1825 forum posts|
Normal lathe design is for the cross slide leadscrew to have a left hand thread. You turn the wheel clockwise to cut deeper.
|Paul Fallert||04/07/2020 15:43:03|
|86 forum posts|
I only had a (worn-out) lathe and some hand tools and bits. Stuff that any self-respecting model-engineer would have binned. So, I made a L-shaped mini-boring-threading-"bar" to internally thread that 3/8" dia thread. I did not know that you cannot easily do such a small internal thread, so I naively proceeded. Because there was only one Acme shaped tooth, my shaping was by a tiny triangle file and pocket stone. This boring/threading tool was silver steel. When the shape of the thread was wrong, I just ground a little more here and there and made another attempt . My point (no pun), is that filing and grinding this single point was very little time. The difficulty was patience. The advantages were the diameter of the bore was under top slide depth control, not re-grinding a tap. Starting over is quick, just re-heat and re-bend and quench-harden. You fit the tool to a visible section of the screw, not disassembled. To make a tap, you have to make a single-point tool AND then a tap. The angle of the thread is adjustable via the tool-post. Not so easy to adjust a tap. The depth of the bore is within the capability of a small-dia round-boring-threading-tool close to the edge of the top-slide. Very little tool extension/flex. [If I were to do this today, I would know to make a holder for the internal-threading-tool, just a 1/2" x 1/2" x 3" rectangular bar, with a hole in the end to closely fit the thread tool and a set screw to tighten and easily adjust the tool angle. Also, I now have a crank-handle on the back end of the spindle to hand-turn the spindle with absolute total control and no crashes. This is supposed to be relaxing, not a factory]. Harold Hall featured this tool-holder in several articles and his books.
|Andy Carlson||04/07/2020 16:17:39|
|258 forum posts|
That's what I thought until Thursday night.
The Faircut handles work in the expected way - you turn the handle clockwise and the slide moves away from you.
It depends on the design of the cross slide though - on the Faircut the handle moves with the cross slide and runs in a nut which is inserted into the saddle, so effectively it works like a normal nut and bolt so it's a right handed thread. The compound is the same. The tailstock is a LH thread because the nut is in the barrel.
On my Unimat SL the handle is mounted on the saddle and the 'nut' is on the cross slide (if you can even call it a cross slide on the Unimat).. so it's a left handed feed screw thread.
|Andy Carlson||04/07/2020 16:25:47|
|258 forum posts|
Cheers Paul. You are right, using a silver steel cutting tool would allow me to be more precise in forming the tool shape. I've been using HSS because I generally assume that HSS is what one uses for cutting tools but we've seen where assumptions get you. I don't think I have any 3mm silver steel just now though... 1.5mm, 6mm and 17mm and I'm waiting for 10mm to make the tap and a counterbore for another job.
My lathe is in a very tight space in the shed so a handle on the back end of the spindle would be pretty awkward. Having one on the big pulley works well though and once the vee belt is off it's really easy to turn. I did make a nut for the back end of my Cowells spindle and use a socket and ratchet on that. I completely agree that this is the best way to do thread cutting. Not sure 'relaxing' is a word I'd use though.
|Andy Carlson||05/07/2020 19:28:13|
|258 forum posts|
I took Paul's advice and sectioned the left handed nut. I generally find that my compact camera is pretty good at really close up inspection so I took some photos in daylight and artificial light. None of the photos is perfect but I present the better ones here for your personal perusal.
The nut (before sectioning) was very tight on the thread and I could only get it onto the thread by clamping it in the vice. It got stiffer the further I went and I never quite got it all the way through. I got the same result from either end. Had the cutting tool not come loose I was intending to take one more thou (radius) to deepen the thread.
The threads on the tailstock screw measure with calipers at 0.85mm. I think there is wear but it is pretty even along the screw. For comparison the very end thread on the compound screw measures a full millimetre.
The odd 'step' in the thread is because I was unable to correctly 'pick up' the thread for the second session so you have a partly cut thread overlapping the almost fully cut one.
|Andy Carlson||17/07/2020 23:13:36|
|258 forum posts|
A little update on the cross slide nuts...
I've made a short section of male thread from some 3/8 steel bar so that I have an unworn example to test with. This did not turn out perfectly. Practice nut No 2 went most of the way on with a healthy degree of drag but is quite tight on the last half inch. As far as I can tell this is the result of pitch error. I noticed when tidying up the point where I stopped threading that the tool did not pick up the thread correctly at that end even though the screw cutting train had not been disengaged and the backlash was going the correct way. I can only conclude that the tool was pulling itself ahead of the leadscrew and into the backlash a little.
For practice nut number 3 I made another threading tool from the 3mm HSS mills to 1 thou over the nominal groove width. I was expecting the usual struggle and fettling when I finished cutting but the test thread screwed in at the first attempt. I foolishly allowed myself to start thinking that I was getting the hang of it and moved on...
Practice nut 3 has been cleaned up and pressed into recess in the spare saddle. That put an immediate stop to any impression of progress. Even the worn feed screw would not screw in without a major struggle. Clearly the nut had compressed significantly. My plug gauge would not go in. A quick touch with a tapered reamer allowed the plug gauge in and the worn screw went in after cleaning up a slight wear ridge that was making it oversize. This revealed that the screw was not quite parallel to the slides - an error of about 1mm in 90mm.
I've concluded that trying to make a correct concentric nut, press fit it into the saddle and expect it to work is not going to fly. I've now started making a tap which hopefully will be able to finish cut the nut threads in situ and allow the correct alignment for the feed screw. The tap needs over-width threads and underwidth grooves... so another threading tool was made. Including the rejects I think I've made six so far. I'm glad I'm using 3mm HSS - it would take a lot longer to make six of them from 6mm HSS.
|not done it yet||18/07/2020 07:14:52|
|4739 forum posts|
... Clearly the nut had compressed significantly. My plug gauge would not go in. ...
That would indicate that your ‘press-fit’ was too tight? Less interference on the the outer diameter would be the better refinement. Even with a sliding fit, the nut could be secured with, say, a ‘bearing fit’ grade of loctite without any distortion to the threads, or fear of it turning in the housing.
|Andy Carlson||18/07/2020 11:13:39|
|258 forum posts|
The nut was based on measurements of the original one that came out. That was a bit uneven but has a taper from 511 thou at the fat end to 506 at the thin end. The lower measurements at the fat end are 509 thou so I was cautious and made the new nut to 509 thou tapering to 506.
To be honest though the clearance is not the big issue - that could be solved by changing the tolerances when I make the nut. The alignment issue is the big problem and I think that for this I need to do the finishing cuts on the nut thread in situ... which will hopefully also mean that I have control over the final major and minor diameter too.
Whether the alignment issue is coming from the pressing operation or the hole is not clear but I did press it out and back in again after turning through 180 degrees. That made it worse, but still to the same side... so I removed it again and put it back how it was originally. I think this makes it pretty certain that the error is not coming from the nut.
4649 forum posts
You need to measure the hole. It should be parallel. Then you turn the nut so it is half a thou bigger or so than the hole but parallel along its length. And as you say, turn the outside diameter in the same set up as you bore the hole and cut the thread so it is all concentric.
Trying to press in a nut with 5 thou taper is likely to go all over the place.
Edited By Hopper on 18/07/2020 11:30:14
|Nigel McBurney 1||18/07/2020 12:15:47|
717 forum posts
unlike v threads ,square threads have clearance on the o/dia when cutting a square thread during my apprenticeship I was told that these threads have 10 thou clearance on the odia,this applied to all such threads irrespective of diameter. I have made a number of square threads over the years from the brake screw on a 1 1/2 inch Allchin TE to some full size brake shafts on Avelings and Fowlers and a very long screw for a sluice gate,all these had a good clearance,Its the fit of the flank of the threads that matters,looking at the photos of the cut brass nut,top and bottom clearance looks minimal or nil. One of my next jobs is 4 square threads for the jacking screws of a 1920s showmans living van,bit difficult at 4 tpi, one unusual feature of the thread is that the depth of thread is half the normal depth.
|Andy Carlson||18/07/2020 13:12:51|
|258 forum posts|
@Hopper: yes, hole measuring... tricky. The nut goes in for about 3mm easily so the fat end is somewhere around 507 thou. I have resolved to make another plug to measure the thin end but right now there is a nut in the way and the lathe occupied by a rather blank looking tap. 'All over the place'... not so sure. Assuming for a moment a parallel hole at the 'fat diameter' and the nut goes fully to one side of the taper then that's 1.5 thou in 14mm (nice mixed units there) which I work out as 1 in 367. There is of course an unknown amount of distortion and a bit of material being scraped off so the pressing is definitely still on the suspect list.
TBH after trying an approach involving doing all of the accurate work in the lathe chuck and finding that it didn't work out too well I'm becoming more convinced that I need to do things in a similar way to the way that Faircut did them... and given the eccentricity in the original nut it seems that 'in situ' was definitely how it was done.
@Nigel: I've struggled to find much concrete info about square threads (as opposde to Acme or modified square). The stuff I've found is fairly vague and hand wavy and says that the screw should be made to size and the nut should have about a thou of clearance. I've gone to plus two thou on the major and minor diameter because it should make no difference to the backlash but your post is the first time I've seen 10 thou quoted. The major diameter is not a big issue - it's just a matter of leaving enough meat in the surrounding nut. 10 thou on the minor seems like a lot - that's a quarter of the depth of the thread.
|Andy Carlson||19/07/2020 09:20:23|
|258 forum posts|
Well that went fairly badly.
I've made a tap, all bar the square on the back end anyway.
Two issues... Firstly both corners chipped off the thread cutting tool - one after about 10 thou of cut and the other towards the end. I cut a bit deeper which should hopefully put the untidy part of the profile in a place where it will do no harm.
The bigger and more puzzling issue is that my 0.0406 thou tool managed to cut wider grooves somehow. The threads measure a tad under 1mm rather than a tad over. I'm not sure why so that does not bode well for being able to correct the problem.
In spite of the issues I decided to finish off the tap and cut the flutes using a 3mm ball end mill in the Potts Spindle. There was a 40T gear on the lathe spindle so I decided that four flutes seemed like a good number. The square for the tap wrench is still to do because the poor little Unimat motor was getting rather warm.
The tap almost certainly won't work for an unworn feed screw but hopefully I will at least be able to find out if the tap approach can produce a cross slide that actually works, albeit still with a worn feed screw.
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.