Gallery, Projects and General > How to's

Single point threading on the lathe

<< < (5/9) > >>

CrewCab:

--- Quote from: Jonny on August 20, 2009, 05:20:10 PM ---Certain Harrison lathes you can disengage half nuts, wind back carriage and engage half nuts again any where on 1, 1.25, 1.5 & 1.75mm pitches. Cant do it on imperial it must remain engaged as it was built as a metric lathe in 1965. 
--- End quote ---

Interesting, would that apply to my Boxford metric lathe .........  ::) ......... mmmmm probably not I suspect  :scratch:

CC

bogstandard:
Dave,

Did you eventually manage to get a metric screwcutting dial for your lathe?

If you did, you should have a chart to show which marks to drop into, dependent on which gear you are using on the screwcutting dial. But even on some metric pitches, you still have to use the continuously engaged leadscrew to cut the thread.

John

CrewCab:

--- Quote from: bogstandard on August 20, 2009, 09:18:15 PM ---Did you eventually manage to get a metric screwcutting dial for your lathe?
--- End quote ---

Not yet John, but I'm still keeping my eyes open, still looking for a faceplate too.

CC

Jonny:

--- Quote from: CrewCab on August 20, 2009, 06:02:47 PM --- would that apply to my Boxford metric lathe
--- End quote ---

I very much doubt it but by some chance it does that'll be a bonus.

Must say on my 140 its a right pain to power in opposite direction as i have to wait up to a minute for motor to stop (been known to stall it like a car), then select direction and power up hoping it dont blow a 13 amp fuse. Same process repetitively often quicker to select fastest gears and wind back by hand, then select back gear again etc. Buy fuses by the hundred never trips the 20 amp feed that the welder and compressor do.
Might rewire it over Christmas plus wire in the 32 amp sockets.



John Hill:
Hmmmmm..... might be a need for lateral thinking here.... :scratch:

The problem as I see it using a metric feed screw to cut imperial threads and vice versa is that it is difficult to ensure the feed screw is in the same position relative to the spindle at the start of each pass.  

Now if you know the pitch of the lead screw and the pitch of the thread you are cutting, (and how could you not know this) you could divide one into the other and determine how many spindle revolutions until they come back to the relative positions they had when the thread was started. If you are using the 120/127 ratio change wheels (or the other equivalent pair)  you might need many revolutions of the lead screw until things line up again, which requires a bit more... :coffee:

Although I began by thinking it might be necessary to fit some sort of revolution counter (did I tell you how to make a counter from an old calculator?) the answer may in fact be very simple,  maybe, perhaps someone will give this a try?


Begin by putting some obvious mark on the metric/imperial change wheel pair, that is the 120/127 or its equivalent (which I can not recall right now  ::)),  anyway, put something you can see while the lathe is turning, something like a dab of paint and if necessary rig up a mirror so you can see it from where you stand to operate the lathe. Fit some sort of pointer  (magnet and a bit of wire?) so that you can line it up with the rotating mark on the change gear.

Start the first cut by engaging the half nuts as your mark lines up with its pointer and continue as normal to make the cut, disengage the half nuts, wind the carriage back to the start, advance the tool then watch the change gear and engage the half nuts when those points align.


I await reports from anyone who tries this.   Of course I may be totally wrong and all there will be to show will be a dab of paint on your change gear. :coffee:

[There is always the possibility that I dont understand the problem  :doh:]


[Later]
OOPS! No that idea will only work if the various pinion gears have teeth counts that are integer dividers of the 120/127 change gears, which of course is impossible!   :doh:  More thinking to be done!

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version