Single point threading speed for aluminum

Couple of thoughts, which may or may not be relevant.

Timing matters. By that I mean one needs to consistently hit the same point on the wheel to get good threads. It's a skill which is best developed at slow speeds. Coarse threads spin the dial faster, and are harder to get consistently. The picture I posted above was by using the half nuts in the ordinary way - using the half nuts and hitting the mark every pass. I practice just using #1 on the dial.

If this is an older lathe, check the condition of the lead screw, is it worn? It is an acme thread, so it should have a nice square top all along the thread. Are your half nuts in good condition? Are they worn?

You seem to be addressing the cutter. That's good. Can you show a well lit picture that is in focus of the tool you were using? So we can see the grinding marks? As if you were holding the bit up close? We don't know if it was ground correctly, or the edges are in good shape.

Another question, how deep are your cuts when you are threading? Please provide a few numbers, just trying to get a sense of what you have been trying. Threading usually takes quite a few passes, so each pass isn't that deep. (That's the way I do it, and it seems to work.) Deepest cuts are ~0.005", tapering off to 0.001" as the thread gets to full depth.
 
At this point, I'm waiting for my new cutters. I'm aware of how critical the half nut timing is, and I just installed brand new half nuts. I'm only going to use a single chosen number on the thread dial to engage the half nuts so I eliminate mistaken engagement as a problem. I have designed my experiments to find out what thing/things I'm doing poorly/wrong and I'll report back afterward... Thanks all for the many comments
 
yes, I used the program on little machine shop to generate the gear choices for cutting the 20 TPI on my lathe given my 2 mm lead screw and my collection of gears. The threads I actually cut are actually a close approximation of 20 TPI and are off by .0001" per every 1" of thread. Here's my chart....

I liked the 48-21-20-54 combination based on my success using that combination last summer when I made my first batch of modified parts. I mapped it on a sheet of paper and it worked fine the first time. My only issue was not very smooth threads, so I thought I would modify my lathe motor switching to make it reversible, make a draw bar style chuck retaining mechanism to secure my screw on chuck, and attempt to get a smoother thread result by turning parts with an upside down cutter at a higher speeds going away from the headstock.

I've done a lot of research. I know the gear selection and assembly is correct. I am awaiting new cutters and plan to compare the results on test pieces of a few different procedures to see if I can find the flaw in my latest procedure. (which I have said above is reverse spin/upside down cutter/moving away from the headstock/ disengaging the half nuts winding back to the start and re-engaging on the same number.

My previous successful batch, I did not have reverse spin capabilities, and I wound the lathe backwards without disengaging the half nuts to eliminate that potential error. (So I'm thinking my error might be related to poor half nut engagement destroying my threads)



20 tpi gear change results.jpg
 
Last edited by a moderator:
Metric lead screws aren't guaranteed to line up correctly for imperial threads. They may only truly line up on the nth revolution of the lead screw, which is impossible for someone to really know. I suggest you leave the half nuts closed for the duration your threading. One less variable for you would be good. There's enough stuff to sort out at the moment.

I have to leave the half nuts engaged whenever cutting metric threads on an imperial lead screw lathe. At some point they sync up, but without a computer knowing the position of both the spindle and the lead screw, it's not possible for a human to easily determine it. The half nut will engage, but it is not guaranteed to be in the correct place - due to the weird conversion factor between imperial and metric.

With imperial threads and an imperial lead screw, it is possible to use the half nuts and thread dial to sync whenever they line up. I don't believe that is easily done (in general) for metric threads and metric lead screw, with a simple dial. There are apparently special thread dials for metric lathes, which differ depending on the required pitch.
 
I didn't know that previously, but I was starting to think that exact thing because of my previous experience having success cutting those threads when I left the half nuts engaged on my first batch back in august. (which did come out correct... but as I said I wanted a nicer finished thread so I did a ton of stuff to my lathe.... (metal gibs to replace the plastic ones, new half nuts, modified the motor switch to turn in reverse, and a chuck lock so the chuck can't unscrew in reverse)... in order to get a better quality of thread.

I did all this improving because I thought speeding up the thread cutting on aluminum would increase the quality of the threads, but to speed up I had to cut upside down moving away from the headstock. I'm still going to try and do it that way. I'll just use the hand wheel on the draw bar to back the cutter up to the relief zone so I keep from crashing while leaving the half nuts engaged. Thanks again for all the help.
 
Speeding up generally improves the finish (more heat in the cut softens the metal) but also increases cutting forces. The latter can ruin the finish when stock, tool and lathe start flexing.
 
Not that I am asking for more help, but just to keep the information correct, I just measured my lead screw and 10 threads measured 20mm, so I assume that means I have a 2mm lead screw. As I said previously I though my lead screw was 1.5mm. Here's the real killer.... I made a first run of these parts in the summer and diagrammed the gear change assembly based on the calculations using a 1.5mm lead screw... And those threads came out correctly (but rougher than I liked)

So now I have mounted those same gears in the same locations and the threads don't match the first run of parts...

OK, so now I tried a new gear combination using the 2mm lead screw number in the gear calculator, and the new gear set up with the correct lead screw number comes out wrong....

Luckily, I have plenty of scrap to keep trying as I try to find my mistake... It's like an episode of the twilight zone.

The only thing I am sure of is 10 threads on the lead screw is 20mm, so the lead screw is 2mm.
 
Not that I am asking for more help, but just to keep the information correct, I just measured my lead screw and 10 threads measured 20mm, so I assume that means I have a 2mm lead screw. As I said previously I though my lead screw was 1.5mm. Here's the real killer.... I made a first run of these parts in the summer and diagrammed the gear change assembly based on the calculations using a 1.5mm lead screw... And those threads came out correctly (but rougher than I liked)

So now I have mounted those same gears in the same locations and the threads don't match the first run of parts...

OK, so now I tried a new gear combination using the 2mm lead screw number in the gear calculator, and the new gear set up with the correct lead screw number comes out wrong....

Luckily, I have plenty of scrap to keep trying as I try to find my mistake... It's like an episode of the twilight zone.

The only thing I am sure of is 10 threads on the lead screw is 20mm, so the lead screw is 2mm.
Don't worry about asking for more help, we all have had twilight zone nights in the shop. It maybe that you are not accounting for the gearbox if you have one. On mine, I have two knobs and they do affect the final ratio.

On this last set of threads, what are your gears? Are your gears correctly marked? Believe it or not, it can happen that the marking is wrong. What are the gearbox knob settings? Can you measure the pitch of the thread you cut? If you have ten threads, we can estimate the pitch. (or how ever many you have, measure and divide by the number of full threads. Can you show a close up of the ragged threads?
 
Back
Top