Why did the lathe take off 6 thousands when the cross feed was indexed at 2?

Welcome to the hell that is the South Bend 9... Hitting your numbers on this lathe can be a serious PITA. It's an extremely capable lathe given its size, but there's a lot of flex in that machine. All I can say is spring cut, spring cut, spring cut!! Of course, also make sure that you're using a tool with a very small nose radius if you're trying to take small finish passes like that. But in the end, having worked with my SB9a practically every day for the last 1.5 years: it's springy. I consistently need to take 3-4 spring cuts before the tool stops cutting.
 
It sounds like you're using both the cross feed and the compound to attempt to hit your target. Personally I would use either one or the other, not both. In most cases I use the micrometer dial on the cross feed only and lock down the compound. First off I would leave a minimum of .015" of material after making the initial cut. After making the initial cut make a second cut without moving the handle. This "spring cut" shouldn't remove more than a couple thousandths, but it should assure the tool has removed all the material intended to be removed without going past the target.

Now measure the stock and see how much still needs to be removed. If you originally intended to remove .015" and actually removed .017" after the spring cut you can expect similar results on each subsequent pass. If you need to reposition the tool to take a measurement I world back off the cross feed crank one full turn so you have the same reading on the dial only .100 or .125" further away from the material depending on the graduations on your dial.

Once you determine how much material remains to be removed reposition the dial one full turn to advance the tool to the exact same position it was in when you took the last cut. Now advance it the amount you wish to remove minus the additional amount that was removed on the "spring cut". This should get you to the target without going over it. If your machine has a lot of play in it you may need to make an additional cut.

The key when moving either the cross slide or compound lead screw is to back it off enough to eliminate any play between the lead screw and the nut when returning it to the cutting position. In most cases backing off the handle a full turn will be sufficient to eliminate the play and insure the dial is reading the true position of the tool when returning to the cutting position.
 
No matter what the dials say you need to develop an eye for material removal. When you have done lots of turning you will be able to "smell" the depth of cut with surprising accuracy
Especially true on a springy lathe like Atlas or SB
Mark
 
As a newby I often run into this same thing. The cut you think that you are making is different from what actually happens. And I have never been able to figure out where I went wrong. Now when I am coming close to the final diameter I never try hit it with a final cut. When I am at this point I take a minimum of 4 spring cuts and sometimes more until I don't see/feel anything being cut. I measure after each spring cut. Then I will try a .001 cut and measure after the cut. No more than this as I sneak up to the final diameter. I have found that you absolutely can't try to re zero your dials. You can't remove the work. Everything has to stay the same. If you change anything everything changes and you are basically starting from scratch. It is not unusual for it to take 15 to 20 minutes for me to sneak up on the final diameter. Even with this if I get within .001 of where I want to be I feel that I have done as good as I can.
 
When I need something precise (like what you describe) I rough turn the material down until I need 3 final cuts.
I carefully measure the starting point, divide by three, and dial in the DIV-3 number and take a cut.
I then measure the resulting cut, determine the difference between the dial and the cut, and the amount left.
Then I dial in 1/2 the rest and take the next to last cut.
Measure again, and dial again shooting at 0.001 too big. This allows a spring pass (with my tools) to hit the mark. Or some polishing with paper.
 
mickri said:
As a newby I often run into this same thing. The cut you think that you are making is different from what actually happens. And I have never been able to figure out where I went wrong. Now when I am coming close to the final diameter I never try hit it with a final cut. When I am at this point I take a minimum of 4 spring cuts and sometimes more until I don't see/feel anything being cut. I measure after each spring cut. Then I will try a .001 cut and measure after the cut. No more than this as I sneak up to the final diameter. I have found that you absolutely can't try to re zero your dials. You can't remove the work. Everything has to stay the same. If you change anything everything changes and you are basically starting from scratch. It is not unusual for it to take 15 to 20 minutes for me to sneak up on the final diameter. Even with this if I get within .001 of where I want to be I feel that I have done as good as I can.
Click to expand...
^^^ This is a very good post. ^^^
You are one step closer to becoming familiar with your machine.
 
This exact same situation happened to me when I took machine shop in high school. I had just finishing polishing the project with 400 grit, it measured right on. Proudly took it to the instructor and he said it's too hot. Imagine my surprise after it cooled, It was a couple of thousands under. When you doing a lot of cutting, a lot of heat is produce and your metal will expand. Before you make your final cuts. let it cool, then take your measurement.
 
If I have to hit a dimension, I leave at least 0.010 for my final pass and leave 1 or 2 for polishing to size. By the time I get to the final pass I have measured several times so I know how much I'm removing. I don't back the crosslide off between passes. I have never had much success trying to take a 0.001 cut and only resort to it when there's only a few thousand material to begin with.
 
I was taught by the "old guys" to go for it. Find a tool-speed -and feed that gives you an acceptable cut finish and then divide the remaining needed removal up into equal cuts. Like Mitch suggested. Measure after each cut. If you make several cuts and they all repeat, go for it. If your nerves won't allow this then back off last cut one or two thousands and file/sand. I try to make that a last resort. When I was machining for a living I used to bore couplings with HSS taking .100" to get down to where five or six cuts of .050 would finish it up and hit it dead on. Of course this was on a big Lodge and Shipley but the same theory/practice holds true on a small lathe. You want to do all your experimenting and wearing in of a newly sharpened bit or new insert when you have material to spare. You have the same tool pressure/spring on every cut so if you dial a value you get the value every time. If you get a combination of speed and feed and bit that works on your machine write it down for the next time. If you only have room for a couple cuts practice on another piece of similar stock to dial it in before you chuck up your piece. Machining stainless requires that you take big cuts to get below work hardening so this method works well. I ran a DoAll lathe that would get weird when the taper attachment was left on and not in use. Just the drag of pulling it up and down the ways would ever so slightly twist the carriage skewing the cuts. So would somebody stopping to watch and leaning against the back panel of the lathe pressing it against the carriage. (only about 1/8' clearance there). Another tip is to leave the lathe running when you dial in a cut. The vibration makes it easier to smoothly advance to a dial reading. Lastly a small pocket magnifying glass is just the ticket for touching off on a cut or dialing that half thousand. (even when your eyesight is good.) Practice makes perfect but unfortunately not every time.
 
Sometimes if your taking very small cuts or trying to hit a critical size it's worth putting a dial indicator on the cross slide so you can measure the movement when your close to size.

Stu
 
You must log in or register to reply here.
Back
Top