Tests for Shoptask tolerances?

[pfarber]

I've made some fairly simply parts on my shoptask. I have no background in maching and am simply learning as i go.

What measurements/items should I check to see if my Shoptask is up to snuff?

I can generally hold a +/-.005 tolerance (again mainly cause I am new and it takes me 2-3 parts to get to that level of accuracy) is there somethings that I should DEFINATELY look at to ensure its more ME makeing the mistake than the machine?

[norvil1]

are you using cnc or manual mode? if using cnc mode ,oo5 is not even in the
ballpark.

[gilchapa]

+/- .005 is a lot, you should check your measuring equipment. If thats not it, check the way you make your setups, your piece may be moving without you noticing...

[Pres]

If you have a DRO - use it!
If not, get at least a 2 axis DRO system asap.

They are pretty inexpensive and it takes out the guesswork of any machine "slop".

Pres

[dahui]

I'm new as well, but here are some things I have tested on my ST. Let me know if you need more procedural details. And yes, DRO is the way to go.

1) Check the backlash on all axes. I wrote them on my DRO screen with a silver sharpie. When I measure, my repeatability is will within .001, but when I'm making parts sometimes I end up scratching my head. As in, measure OD of a turned part, set DRO, take a few cuts, measure...and my DRO is off by a few thou.

2) Check/set the angle of the table perpendicular to the X axis.

3) Check the alignment of your tailstock (center, height, and parallelism to the ways)

4) use a 4 jaw chuck and dial indicator for turning. However, my three jaw seems to be more accurate that I would expect.

5) check the runout on the lathe and mill spindles. I found the 3/8 collet that came with my machine was junk. Buy some high quality collets. Remember, accuracy (or inaccuracy) is a compounding problem. You will only be as accurate as the least accurate part of your setup.

6) Also, make sure the mill spindle is perpendicular to the table. I'm not sure of any way to adjust that if it is in fact out of tram.

7) Adjust the gibs on the X and Y axis. I had to make a new gib for the X axis to get it to adjust correctly. The gib that came on my machine was signifcantly warped and I couldn't adjust out the play without getting a really sticky travel.

8) On that note, watch out for the table rotating around the Z axis when you move the X axis...adjusting the gib should minimize this. However, when I'm turning I try to set the Y axis travel at zero and set up the tip of my turning tool to be over the center. This should minimize the effect of any rocking that does occur.

Oh by the way it would be good to know if you are using CNC and what particular operation you are having trouble with. If you are not using CNC and have acme screws, it's also very important to set your position when traversing the axis in the same direction you did when you zeroed out...otherwise you will be off by the amount of backlash.

[pfarber]

I have 3 axis DRO.

Since the encoders for the DRO are not on the leadscrews then they should 'automatically' be 'ignoring' and backlash?

[sch]

The mill head is trammed with some difficulty by putting shims between the
the bottom of the quad cage and the lathe spindle housing. The four point
bolting gives you the necessary axes to tram the spindle in. I have never
even checked mine but the approach is similar to that done on a milldrill.
PITA compared to the 2min drill to tramming a knee mill.

One other comment, cutting closer than 0.005 is more than just DRO usage,
it requires a knowledge or the experience to know how much spring is in the system. Cutting something 3" in diameter on the lathe differs from something
1/2" in diameter. The final thousands tend to come off without any change in the tool setting at all, either lathe or mill, just repeating the last cut will frequently result in a frisson of chips flying off.

[scbvideoboy]

as a newbie my self, time on the machine and practice wasting parts will help bring down the tolerances. Spend some time on testing. The first 6 months I had my machine running I was doing more fixing and adjusting the machine vs making something.

Everybody will say the same about setup.

I've tighten everything I can for backlash, now it's more stopping and measure and reset 0, cut do the same again,even more so when you get to the critical measurement.

Slow going though. A long way to go.

Dave