Mill leadscrew nuts

[old mart]

After not being able to get hold of ir16 trapezoidal laydown inserts in 2.5mm pitch for years from any UK supplier, a Chinese source on ebay listed them. The drill mill uses right and left hand 2.5mm leadscrews of about 22.6mm OD so making the nuts was not much of a problem. By luck, the idler gear on the Smart & Brown model A has 120-127 teeth, so driving the 127 and outputting from the 125 turned a 10tpi thread into a 2.5mm exactly.
Fortunately, on both axes, there was plenty of room to fit a more sophisticated system of antibacklash nuts with over double the engaged length of thread than the originals. I bought sone gunmetal SAE660 which is a leaded bronze to make the nuts, which are fitted in aluminium bodies. The price of using all bronze was prohibitive, hence the aluminium. One nut in each assembly is Loctited in and the other is threaded 40tpi and a clamp stops it from rotating by itself, hopefully. The Y nut is on the left.

[old mart]

Having fitted both new nuts, I have checked the backlash and found it is 0.1mm, 0.004" on both X and Y. This is axial play in the thrust bearings, confirmed by tightening the antibacklash nuts till the nuts are getting rather tight to turn. The drill mill thrust bearing design leaves a lot to be desired and I will be thinking of ways to have decent adjustment.

[Cuttings]

Good job

[old mart]

Both nuts have been installed for some time and I have just finished the thrust bearing mod for the Y axis.
The mod addresses two shortcomings of the originals, non adjustable bearings and the poor design of the locking screw on the scale which obstructs the view of some of the numbers. I forgot to take a photo of the scale improvment, it will have to wait till next week.

[mattthemuppet]

Very slick work! The back lash adjustment in particular is a very clever approach

[old mart]

Unfortunately there must be some wear in the Y axis leadscrew which I had not detected. With the ends being tight, the centre has 0.1mm (0.004") backlash and while I will try adjusting the nuts so each end rubs on the opposite flank of the leadscrew threads, it is unlikely to improve things. At least there is zero axial play.
At the last minute, I found that the parts could be sealed with an O ring to keep muck out. I got extra needle roller bearing plates to go outboard of the bearings as the shoulder of the adaptor and the nut diameter only went halfway up the roller length, and the plates are very thin.

[Mark Rand]

After not being able to get hold of ir16 trapezoidal laydown inserts in 2.5mm pitch for years from any UK supplier

Too late now, but Cutwell do them.
I've used them for 10tpi ACME and 0BA inserts.

When I did the feedscrew nuts on the Beaver mill, I didn't have any laydown threading tools, so I used an HSS blank, ground to the right size, in a boring bar. It seemed to work!

[old mart]

I found the Cutwell listing, I had not noticed it before, it is for an external LH thread. I could have used it in a left hand internal bar cutting towards the chuck. The helix angle is quite small so I could have got away with it. There is only the one in their listing, a 16EL2.5TR VTX, the VTX is a general purpose grade.

[Mark Rand]

Ah!
The 10tpi ACME insert I got was for an external left hand thread as well. I used the right hand external insert with a pair of 6° shims above and below the tool holder to correct the angle and provide for clearance. The thread was 7/16"x10tpi for a BCA jig borer. It worked well.

[old mart]

I have a pair of shims which tilted the external threading tool shank for when I cut a 1/2" X 10 ACME double start thread for a second toolpost for the Smart & Brown model A. The nut is attached to the compound. The helix angle was quite pronounced. I have a supply of 12mm and 1/2" diameter Densimet (a tungsten/cobalt sintered alloy like tungsten carbide but not so hard) which is big enough to make ER/IR16 laydown threading insert internal bars. It is easy to make the insert tilt one way or the other during milling the seating. This alloy is also just in the tapping range of HSS for the screws.

[old mart]

Here's the Y axis with the scale fitted, having the locking screw away from the scale makes for easier reading.
Today, Thursday, Rod and I managed to get the Y backlash down to 0.05mm (0.002") by re adjusting the new nut. This is on a 40tpi thread and can move about 15 degrees before binding. Any tighter and the handwheel gets too hard to turn.

[old mart]

Finished the Xaxis yesterday, and the backlash throughout the 500mm of travel is 0.05mm (0.002") in the centre and zero at the ends.

[RB211]

Nice job!