I just finished tramming my new mill and I have it down to 0.0008" over 12". It seems to me that should be good enough for most of the projects I'm likely to do, but I'm interested in other opinions. My mill, a Precision Matthews 45M, is a 9x32 bench-type mill with a square dovetailed column. I had to shim the front two bolts that hold the column to the base using 0.001" shim. I'm using a test indicator graduated in 0.0005" attached to an indicol on the quill. As I rotate the spindle, the indicator tip sweeps out a 12" circle (I have to traverse the table fore and aft to keep it under the indicator). I don't have any shim stock smaller than 0.001" (not even sure if I could find any), so to get it any better, I guess I'd have to put some combination of shims under both the front and rear bolts. So what do you think, is 0.0008" close enough, and if not, does anyone have a better suggestion on how to reduce the error?
It's close enough. I use paper and mylar film for shims on mine.
To get around the traverse issue try this:
I normally put the brake rotor face on the table but thought this would be easier to see. Brakes are remarkably parallel when new.
Keep working at it and you can get it to .0002" or less.
It's only ink and paper
Are you sure your tables even flat to .0008" in 12"? Put one hand on the miller head and push while you recheck your tram. I bet your machine is less rigid than you think. We push for these fantastic numbers but they are pretty hard to hit and while most of the time I am the limit to accuracy sometimes it is the inaccuracy and lack of rigidity built into the machine.
Just under a thou in 12 inches- that's not too bad. If you could find some half-thou shim stock, you might be able to get it better, but you should check this with the table in different spots also. You may find it pointless to try for any better than what you're already achieved.
One should also measure that the table top is flat to the dovetails and that the Z-axis dovetail is square to the table movement, otherwise tramming to 0.000000...1 is quite pointless.
I am assuming by 12" you mean a sweep of 6" radius which would mean that from one point 180 degrees away from the opposite point you'd be out .0008 or a total of approximately .000066/inch. I would venture to say that the bearings and ways of your mill (or mine) do not have this level accuracy so the point is academic.
Your milling head tram or vise indication should be reasonably accurate for the type of work you do but I have chuckled to myself in the past over posts to this forum talking about levels of accuracy in the lower tenths or even millionths for the average home shop project. The key word is 'practicality'. If the accuracy of your set up suits the accuracy needed for your work that's all you need. An example of what I am talking about would be a typical flycut. If you are cutting on the backside of your flycut you should adjust your head if the results you are looking for require a PERFECTLY flat surface. If you are just squaring up stock for a general purpose project which doesn't require absolute flatness why waste the time. If I have .001 over six inches (Dia.) on the tram I am happy, even for most work requiring reasonably high accuracy.
Last edited by DATo; 11-02-2011 at 05:39 AM.
The exercise wasn't a waste. No doubt you've learned a lot and IMO got some quite impressive results. You also know a lot more about tramming. Something I don't believe was brought up though. Were your tram results measured with the table and head locks on or off? Those can change your results too.
No your mill won't machine to that level of accuracy, But you got it trammed to a much better condition than what you started with. That alone deserves an attaboy.
Well, as we used to say when I was but a wee lad, I didn't fall off the turnip truck yesterday. I'm not trying to tram it to 0.0000" (nor would I, at least not this particular machine), I just want it to be trammed to a point that is generally accepted as "pretty damned good" for the home shop. It seems to me that 0.0008" over 12" is pretty reasonable, and it wasn't very difficult to get it there. I think that getting it better than that may be quite a bit more difficult, perhaps in the area that I would call "diminishing returns". As far as locking the gibs, I had the X and the Z gibs locked. The Y-axis (front to back) I was traversing since the swing of the indicator was greater than the width of my table. The 0.0008" measurement is the average of 4 separate readings. Getting it much better would require locking the Y-axis as well and using some kind of parallel-surface plate of sufficient diameter, as suggested by DP. Based of all the responses I've seen so far, I think it is pretty good where it is and I am so far very pleased with the machine.
Carld, you are correct that at this point, I do not know that my machine is not capable of machining to 0.0008". It is brand new and I've never used one of these Chinese machines before. I also hear that there is a fair bit of variability from one machine to the next, even from the same supplier. I do know that I will never be able to repeatably machine to any tighter tolerance than the machine is trammed to. I also know that I've often gotten to within +/- 0.001" on a pretty worn old Bridgeport. I also know that if I assume the machine is incapable, or that I am incapable, of meeting a particular tolerance, that becomes a self-fulfilling prophecy.
Gunny - Let me hasten to add that I DO respect your desire to achieve the best settings that your machine is capable of, and as someone else pointed out the exercise in itself was worth the time if only as practice in tramming.
Congratulations on your new purchase and I wish you many HaPpy hours of machining with it !