Making a small lead screw for a Palmgren milling att.

[_Axel_]

Gentlemen,

I have no idea what I'm doing wrong. I am converting an old Palmgren milling attachment to a metric configuration. I need to make a 3" lead screw for this. I have chosen 8x1mm thread. Doing this I first tried a piece of LaSalle cold rolled material. It's called Stresseproof or something similar. I single pointed this. But it warped. So I tried normalizing this material. Still warped, this time using a die.

Since then I run out of this material. And have tried both 12L14 (equv.) and now a piece of C45. All warp even after its been normalized. I use a proper hardening furnace of good quality. Normalizing temp has been 900C for he C45, hold time 15mins.

Intelligent input needed!

Thanks!

 

[_Axel_]

I have done more tests. First I raised temp to 950C and held it there for 15min. Cool in air. Material still warps badly when cut with a die. Second test was heating to 950C, hold time 15m, and cool slowly over night in the furnace (annealing). Still warps badly when cut with a die.

 

[abby]

A quick web search shows that you can buy an M8x1.0 70 mm long leadscrew for under £10 , cutting a leadscrew with a button die is not a good idea as over a 3 inch length it will be difficult to get an even cut by hand .
I am presuming that you haven't got a lathe for screwcutting which would be the obvious way to go ,but I am sure there is a member who would make this for you given your location.
Dan.

 

[clockworkcheval]

Funny enough the recommendations for treatment of C45 seem to differ across the Atlantic. From your attachment I understand a recommended normalizing temperature of 870 - 920 C. Here in The Netherlands the specification is 840 - 880 C. Moreover the attachment states that C45 is weldable, whereas our specs say to avoid welding because of the relatively high C content. Still I'm not sure whether this difference is the cause of you problems. I would do 1) heat treat at 860 C for yes 15 minutes 2) straighten 3) cut 4) straighten. I would avoid die-cutting at this length and do single point cutting with support of a traveling steady.
That said I would choose another material. In horology many shafts are in the same dimensional area as you state, say length 50 - 150 mm and diameter 6 - 10 mm. These shafts are machined all over and often have a integral gear cut upon them. For the clock to run reliable these shafts must be straight. For a very long time we used Silver Steel for these shafts, which is the generic name for a low alloy group of tool steels with carbon content of 1,0 % - 1,2%. They work very well but are not really easy to machine. Some ten years ago we moved to the Swiss made ETG100 steel. ETG = Extra Temperatur Gezogen or drawn at a high temperature. It is essentially a free cutting steel that is strengthened by drawing and during that same strengthening process somewhat stress relieved by applying extra heat. It cuts beautifully and stays straight. It has very good mechanical properties, strength and wear resistance. No additional heat treatment because doing so would destroy the good properties. Preferred applications are shafts, gears, spindles and arbors. I'm not sure of the availability of this or similar steels at your place.

 

[Richard Hed]

I would not use something so small as 8mm. can you get 10 or better, 12? can you show a photo of where the screw will go?

 

[clockworkcheval]

Maybe my first question should have been why normalize at all? You would expect steel to be delivered in a normalized state.

 

[Richard Hed]

I thimpfks that if you use quality metal, you could single point a good 10 or 12-1m . good smooth threads and relatively tite will help with accuracy. However, if yuou use your measuring tools correctly and use a brake to tighten down the ways, then just about any screw would doo.

 

[_Axel_]

Thanks for all replies, I have figured out it is not a heat treat problem. It's a problem with the die. As I said above, the first try was a piece of cold worked medium C steel. It actuall bent when i screw cut it on the lathe. But after reviewing my pile of junk leadscrews. I noticed the bend was even on one side only on the first one (if that makes sense). It did not warp, or spiral, like the others, that were cut with the die (Kamaza tools). The die was the problem all along! The die makes a nice thread, but for some reason the workpiece becomes a spiral. Normally when I get a wonky thread with a die, one side is flat, and the opposite full depth.

I'm not sure material that is colled rolled, or bright, is normalized at all. It has been drawn through an eye to become a certain dimension, and sometimes center less ground after that.