Metric Screwcutting

Hi All,
I'm sure I will eventually have some answers for your questions but at
the moment I am in a serious re-learning curve.
My Myford is Imperial so I bought a 127 tooth change gear to get
accurate metric pitches, I've also created a theoretical table of gear
trains which I am updating empirically taking account of the constraints
of the standard Banjo. That's all well and good and I know that once
engaged I should not dis-engage the half-nut . . . SO Why am I
getting multi-threads?
The tool post is set up at 30 degrees,
the single point tool is square to the work,
cut applied via the top slide,
the nut is engaged and the first pass commenced
At the end of the thread I . . . .
disengage the countershaft,
swop the position of the tumbler gearing,
pull back the cross slide,
re-engage the countershaft
let the apron get beyond the start of the thread
disengage the countershaft,
swop the position of the tumbler gearing,
return the cross slide to zero
apply cut via the top slide
re-engage the countershaft
and so on and so forth
What am I doing wrong?
TIA
JG
Reply to
JG
Loading thread data ...
The way I have done it in the past is totally different. The problem I bet you are having is the switching of the tumblers.
What I have done is to disengage the half nuts, wind the saddle to the start, re engage the half nuts (lathe is off) and re-align the tool to the thread, hand wind the chuck backwards to move the saddle clear of the thread, put on the cut and start the lathe.
This worked for me, but was not perfect. It allowed me to cut enough thread so that I could run a die down without going wonky.
Regards, Dave. ~~ Customise your internet experience
formatting link

Reply to
Dave
Usual procedure is not to disengage either counter shaft of half nut. Just stop the lathe at the end of the cut, withdraw the tool, reverse the lathe and run back past the beginning of the thread. Then wind the tool back in, set the lathe in forward gear and make the next cut. Note, due to backlash in gearing etc, run the lathe back well past the start of the cut.
Reply to
Tim Christian
Absolutely, you are loosing the relationship of headstock to gear train by changing the tumblers over. They may both have the same number of teeth but gear back lash means that you cannot be certain of engaging them in exactly the same place tooth for tooth every time.
As already advised stop the lathe leaving everything engaged, reverse the motor and run back past the start. Put motor to forward again and take the next cut.
Alan
Reply to
Alan Marshall
from "Dave" contains these words:
That is *exactly* the issue I had that made me decide to screw-cut rather than die!
Making a Slitting saw arbour I found that the 12mm thread was so far out of true (even though I had supported the die-stock via the tail-stock) that it was useless.
With assistance from Darrel Crook via e-mail and Tim Christian, I have now found out how to reverse the motor - so I'll have another go now (I've already spent too much on eBay today anyway!)
JG
Reply to
JG
Sorry to tell you what you knew.
I invested in an invertor/controller so reverse is just a switch. The very slow speeds available mean I can power screw cut a couple of threads up against a shoulder. I'm spoilt!
Reply to
Tim Christian
Yep, I know that is usual (from my engineering days) but as my lathe doesn't have a reverse (yet), I had to improvise.
There was an article that I saw in engineering in miniature (vol 7 or 8, not sure which one) of another method. Setting end stops on the lathe bed (one of the stops was the tailstock). You worked out the TPI against the TPI of the leedscrew, calculated a distance that the bed must traverse and set the tailstock 1/2 of the leedscrew thread longer than necessary. The idea being that you will have moved a multiplier of the threads needed to re-engage in the same place. This method was also good for multi start threads.
The way I did it though was good enough for me. I didn't need to make any end stops, I had no reverse and I was working on long bar. I also ensured that I was taking out the backlash (after re-engaging the half nuts, wind the saddle to the rear using the main traverse handle to lock the nuts against the thread) then re-aligning the tool against the thread using the compound slide.
There are many ways to set up for screwcutting. Some ingenious, some not so... I prefer the reversing the chuck method as it will always be spot on, but when you don't have the facility you prefer, you have to look at alternatives.
Best regards, Dave. ~~ Customise your internet experience
formatting link

Reply to
Dave
from "Tim Christian" contains these words:
It seems that my lathe came with one already installed but I hadn't got round to finding out what it really did.
I had already set the speed to 35rpm and have now cut my first thread - not the best thread that has ever been cut but it will do for a Tee-Bolt and has been wonderful as an exercise - including grinding the single point tool by hand without the aid of a 50:1 shadow-graph.
What I need now is a screw-thread micrometer . . Oh yes, and some ACME chasers . . . . :))
JG
Reply to
JG
I may have got the wrong idea here, but I always thought that one had to use a compound gear to get from imperial to metric, a 120/127 gear. Drive the 120 and take off the 127.
Is this what you have now after picking up the 127 tooth gear, have I picked up the wrong thread!!! Sorry.
Adrian
Reply to
Adrian Hodgson
With any Imperial lead-screw machine you must introduce a gear in the change-wheel set that has a multiple of 25.4 teeth. The first such gear is 127 (since you must have an interger number of teeth).
Introduce this at any convenient point in the sequence and you are effectively turning the lead-screw (and hence moving the saddle when the half-nut is engaged) an accurate metric distance for each turn of the chuck.
If you put a 40 tooth gear on the tumbler and the 127 tooth gear on the lead-screw then any idler that will comfortably link the two on stud one you will cause the saddle to move 1mm for each revolution of the chuck - thus enabling a 1mm pitch thread to be cut accurately.
Without the 127 tooth gear and using Myford's change-wheel set-up for 1mm pitch - 45 tooth driver, 40/21 on stud one, 50 idler on stud two and 75 on the lead-screw - will actually cause the saddle to travel 0.999603mm per rev, a small error of 1 in 2520 but an error none-the-less. To keep it in perspective it means that if you were to cut a 450mm long thread (close to the maximum on a standard bed-length machine) it would actually be 449.821428mm long - 0.178571mm short.
I can't fathom where the 120/127 combination comes from and it would not be possible to fit both on a standard set-up as the distance between the lead-screw and tumbler shafts is 109.641mm and the distance between a 120 and 127 would be 156.845mm.
JG
Reply to
JG
Some lathes do use this combination, especially the Chinese imports.
Reply to
Tim Christian
I am quite sure my old imperial lathe used this combination it was a Smart and Brown Sabel with a Norton Gearbox, to get metric threads I had to introduce a compound gear into the chain down to the box. The leadscrew was a standard 8 TPI, Unfortunatly I do not have the machine or any documentation to refare to for this lathe.
What does the Myford in question have with regards to leadscrew and change gears?
If I find any links to above i will post
Adrian .
Reply to
Adrian Hodgson
Ok got it wrong it was a 127/100 compound gear drive the 127 gear from the mandrel and take off from the 100, sumat like that.
The gears may only have been needed due to the fact I had the norton box but not sure.
Reply to
Adrian Hodgson
It is a late ML7 with a 3/4" 8tpi Lead-screw (as on the Super 7) and therefore the Super 7 saddle but it has the ML7 oil-pots on the head-stock so does not have the conical spindle bearing. Likewise the change-gearing is the ML7 type. The tail-stock is the Super 7 type.
JG
Reply to
JG
For the last forty odd years I have used a 21 tooth gear wheel on the ML7 when cutting metric threads. No problems or complaints so far.
Donald, Isle of South Uist
Reply to
Donald

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.