Shaper accuracy

[Pete301]

I wanted to check with you guys who have more experience with shapers.
I have a South Bend 7” shaper and I’m uncertain about how accurate a small shaper like this is capable of cutting.

On a 4” long cut on hot rolled 1018 steel, cutting .020 deep at a .006 feed, I get a taper of about 2 thousandths front to back. ( larger at front-smaller at rear). If I then follow up with a “spring cut” it trues up the part to under 1/2 thousandths of a taper. I notice some variance in the amount of taper based on how deep I set the shaper to cut. Deeper the cut , the more the taper. Should I just expect this? Or is this more of a taper than you get on your machine?
I’m not asking how to make it cut more accurate but instead unsure if the results I’m getting are in the normal or typical range for this size machine doing that deep of a cut.
My prior experience is only on a old worn lathe and it often requires a spring cut to true up the part.

Pete

 

[Nutfarmer]

On my old shaper the ways that the ram slid back and forth in wore more on the side next to the table. T hat gave the cutting taper you described. Rework of the ram ways and the taper went away. Just as with any older pieces of machinery it can be a challenge to hold accurate cuts. Some further investigation with a test indicator will give a better idea of where the taper is coming from. It could even be as simple as adjusting the gibs on the ram. If they are too loose that can give problems. The shaper should cut to an accuracy of . 001 . Any closer than that I would use a grinder.

 

[Ulma Doctor]

it's pretty common to have variations on shapers.
you'll have ram lift as well as box travel deviations that seem to be exacerbated by the smaller size on harder materials
sharp tools will help tremendously and a survey of the machine will show you wear in the travel axes, which leads to inconsistencies.
shimming of the part is often required to hit finer tolerances

 

[eeler1]

How accurate does the part need to be? If you are within 1/2 thou over 4 inches on that little machine you should be pretty happy, although it may be possible to achieve better with some tricks like Mike mentioned. My $0.02.

 

[projectnut]

This may be a dangerous thing to say, but I "assume" you have adjusted the ram gib using the set screws on the bull gear side. The recommended procedure is to start from both ends and work toward the middle. Given the age of the machine you may want to inspect the gib and mating surfaces for imbedded chips. In any case I think .0005" in 4" is well beyond the original spec.

South bend recommends a finish cut to be between .010" and 015".

 

[ThinWoodsman]

I’m not asking how to make it cut more accurate but instead unsure if the results I’m getting are in the normal or typical range for this size machine doing that deep of a cut.

I'd say your experience is pretty typical. It takes some time to tune the machine as others have said - before I took mine apart, I had a .002 shim under the vise (outside end) to account for the taper I was getting. I'd like to say that got me within a thou, but it really depended on tooling. A light finishing cut with a shear tool - .005 to .002 deep cuts no problem, within a thou of expected. Deeper cuts, harder steel, different tool (or same tool after a bit of cutting without stopping to sharpen) - tearing, maybe within .005 of what is expected.

My shaper is currently down. Not for any justifiable reason - I took the table and apron off to make DRO mounts, put the table (now with DRO !) back on but the apron is still off waiting for me to finish its DRO mount. At some point the sadness of not seeing the shaper run will provide the incentive to turn the wooden prototype into metal.

 

[Ulma Doctor]

It will be difficult to achieve .0001” tolerances with .001” dials

 

[ThinWoodsman]

It's the dial graduated 0 - 52 that always gets me

 

[Pete301]

I'd say your experience is pretty typical. It takes some time to tune the machine as others have said - before I took mine apart, I had a .002 shim under the vise (outside end) to account for the taper I was getting. I'd like to say that got me within a thou, but it really depended on tooling. A light finishing cut with a shear tool - .005 to .002 deep cuts no problem, within a thou of expected. Deeper cuts, harder steel, different tool (or same tool after a bit of cutting without stopping to sharpen) - tearing, maybe within .005 of what is expected.

My shaper is currently down. Not for any justifiable reason - I took the table and apron off to make DRO mounts, put the table (now with DRO !) back on but the apron is still off waiting for me to finish its DRO mount. At some point the sadness of not seeing the shaper run will provide the incentive to turn the wooden prototype into metal.

Thanks for all the responses! The pic below is marked to show the reading I measured at the four corners of my test workpiece.

So if this came off your shaper would you figure it’s about the normal?

If I then do a spring cut on the workpiece it trues up under a .001.

So part of this question is about how much the depth of cut effects the accuracy. Should I expect this much variance with a .020 cut. If so I can accept that and just run a spring cut when I want more accuracy.

 

[francist]

Have you tried the cut at slightly less depth, say 0.010” and another at 0.015” depth of cut? I have an 8” machine and find that it seems to prefer about 15 thousandths for steel so I’m wondering if you’re maybe just pushing the cut a bit much given the machine age and wear expectation. I would not expect a 4-thousandths variation of my machine over that length, it seems a bit much.

Maybe try a couple more tests at different depths, see what you can determine. And of course a sharp tool is important too for less push-back.

-frank