Runout on y-axis travel

U

UMDJesse

Registered
Registered
Joined
Dec 14, 2021
Messages
28
I'm in the process of tramming out my PM-30. I have everything square off the spindle with my tramming fixture showing around .00025 on the x-axis over 12". The y-axis is square to about .001 with the column tilting slightly away from the operator...I might shim, I might not, but I know how to tweak this if I decide to. With everything adjusted, gibs snug, column locked, etc., I can crank the x-axis through its full range of motion and I get zero runout on X, it stays perfectly flat...great! Here is the issue, with everything still in place I go and crank the y-axis through its full range of motion and I get .005 (5 thou) runout over the width of the table...indicating the table sloping .005 (5 thou) away from the operator position. How can I resolve this? I've tried snugging up the gibs but this doesn't seem to change anything.

How can I adjust this table runout on y travel?

With the column fixed and locked, and the indicator fixed and locked on the spindle, I would expect zero runout (or close to it) on y-travel...even if the column were tilted I would not expect that to affect y axis table movement since the spindle and column are fixed.

TLDR
Tramming spindle = I have x-axis and y-axis satisfactorily square
x and y table travel = x is totaly flat, but I get .005 runout on y over the length of the table (about 8 inched)

Thanks!
 
Runout isnt the correct term, but I understand what your saying. There are some issues you'll probably just have to live with on a machine of that price point.

i believe, to fix the issue you are having, you'll need a good size surface grinder or do some scraping.

this would be a good post for the the machine rebuilding subforum.

there are others here alot more knowledgable than me about diagnosing issues like this, scraping ways, etc. Maybe one of them will chime in.

Do some searches in the machine rebuilding subforum. I bet you can find threads about similar problems.



As an example. Here's a previous thread..

www.hobby-machinist.com

Milling Table Unevenness

I got a PM-30MV mill and trammed the table. I set it up with a dial indicator on just the table surface and got pretty good, a couple of of tenths around the center of the table. I found out the left side X is .003 higher then the right at around 27+ inches. OK. Got a disk brake rotor around...
www.hobby-machinist.com www.hobby-machinist.com
 
Last edited:
Tramming correctly means adjusting the head of a mill so that the spindle axis is perpendicular to the ways. Tramming is done so that a facing operation doesn't create scallops in the direction of travel or steps in the direction perpendicular to travel. A common assumption is that the table surface is parallel to the x and y ways. It is just that; an assumption. And not always correct.

A table surface may not be parallel for a variety of reasons. Firstly, it may not have been machined correctly when it was made. Secondly, the table surface may be worn from use. And lastly, loose gibs may be allowing the table to lift as the center of gravity moves relative to the mill base.

It is fairly easy to determine whether the surface is parallel to the ways. Simply mount an indicator in the spindle and sweep the surface, recording the indicator height at various positions. I have done that on my Tormach mill with the following results.
Tormach 770 Table Profile.JPG


Each line shows table elevation at various x positions for a given y position from the home position (rear left on the table). The dotted lines show elevation before gib adjustment and the solid lines are after. Note that the home position has the table extended furthest from the column so the 0,0 position shows the greatest elevation in the y direction where the table has the least support. Also note the rapid increase in table elevation in the x direction. This is caused by the design of the mill which has a fairly large mass in the x axis drive on the left side of the table. Note the improvement made by adjusting the gibs. Because this is a CNC mill, the table gibs have to be adjusted to allow free movement. On a manual mill, the table locks can be engaged to oiveride the effect of loose gibs.

To properly tram a mill when the table is not parallel with the ways, a sacrificial plate can be fixed to the table and the surface milled with a small diameter end mill. The small diameter minimizes and scalloping or step effect. The spindle is then trammed to that surface. The loose gib effect can still influencethe facing operation. To that end, the location of the trammed surface should be chosen so that the table is best supported by the x and y ways. In my case, that would mean working in the area with the table furthest in and to the left where my elevation variation is around a half a thou.
 
Hello, I generally hang out in the rebuilding forum :) as I am a Machine Rebuilder by trade. Is your machine new or just new to you? How old is it? Can you attach some photo's with a sketch or write on the table with a marker. I would first Micrometer (mike) the table on all 4 corners to be sure it is co-planer or the same thickness front to back. New machines are built that way. Next take some .0015" thick feeler gage and see if you can slide it between the saddle and table on the 4 corners and middle front and back. It should be tight.

Then remove the saddle way wipers front and back and try to slide the feeler gage between the saddle and knee all 4 corners.
As RJ said tighten the locks. I would mount a dial indicator on the top of the table and tighten the knee, saddle and table locks and see if you get any movement. Another test would be to measure the thickness of the saddle by using a height gage and indicator. Move the saddle to the center of the travel, lock it and then set the height gage on each corner of the exposed knee top and measure the height to the top of the table. When the machine was new all these measurements should be zero or only out a few tenths. Check that and let us know. Good luck. Rich
 
Wow, thanks for the detailed responses guys.

Some back story:
The mill is a brand new PM30mv, hasn't even made a chip yet. When I received the mill, the table had a small casting imperfection. The very helpful folks at precision mathews offered to have the table reground. This thread now follows my setup issues having received the reground table.

I think the tram is good. I'm perfect on X, and maybe about 2 thou on Y. I was thinking I might shim the column but I'm not sure that's the issue now. For table axis travel, when indicating off of the spindle, I get no gauge movement on x. Throughout the entire 30" of travel on x, my .001 indicator will read 0.000. So x is darn near perfect and parallel with the gibs. When I repeat this test through the 8" of y travel I get .005. For y, even if the tram were off, I would expect close to zero since I'm not moving the spindle.

This seems to indicate to me that maybe this table was reground with a .005" taper. Does my logic add up?
Also, what would be the acceptable tolerance for this on a hobby bench mill? .001? .0005? .002?

RJSakowski,
"It is fairly easy to determine whether the surface is parallel to the ways. Simply mount an indicator in the spindle and sweep the surface, recording the indicator height at various positions. I have done that on my Tormach mill with the following results." I think this is what I've accomplished above.

Rich,
So I should have probably included this initially, but yeah the mill is brand spanking new. I've checked the fitment between the ways and gibs...and I can maybe slide a feeler just slightly in the gib for y, there does seem to be maybe a bit of a gap in there, but I have it adjusted as snugly as I can while still having smooth movement. I cannot get a feeler anywhere between the saddle and table. As for suggestion to mike the four corners, mike against what? could you elaborate on this? As for the other checks....I have an email in to PM tech support for this issue and I want to see how they reply before I tear the mill down. But I will get back to you on this.

So given the above measurements, is there anything else I can do to confirm my suspicion that the table has a taper? I'm somewhat in disbelief on this since they were able to get x-axis so perfectly flat and parallel, that y would have a taper...so I'm really hesitant to come to that conclusion. Unfortunately I did not take any of these measurements before sending the table for a regrind.

Thanks all
 
I just looked up he PM 30 mill.....https://www.precisionmatthews.com/shop/pm-30mv/

You need to Mic from the top of the table to the bottom thickness. 4 corners and move the table and check the middle. I have no idea how thick the table is. I have seen similar machines and the column was off. If it is a BRAND new machine does it have a test sheet in the back of the manual showing the specifications page showing the tests and the results of the tests when it was built. If it isn't there then when you call them ask for a copy. If you move the saddle (Y) in and out and you get .005" from front to back and if PM reground the table it should mic within .0005" in the width. If it mic's good then I suspect the saddle is off. PM Tech support should be able to walk you through the tests. If you had a Precision blade square and set it on the table and moved the head up and down on the Z it should be with-in .0005" in 12". If you don't have a square, do you have a 2,3,4 block to set on the end and indicate it for squareness. I know a fellow in Christiansburg VA who is a retired rebuilder who may come over to help if needed. If you end up shimming it, I would also add some dowel pins. to anchor it if the cap screws loosen. Did you check the tightness in between the column and the head with a feeler gage?
 
Last edited:
Rich,
Using a straight edge across the table I'm showing flat to at least .001 using your feeler gauge test.
I then Mike'd the 4 corners as you mentioned and I do see a taper there....
1648138270780.png

So if I'm reading your message right I should be within .0005, but I'm over 10x that at .006?

I looked up Christiansburg...thats a good 3 hours away so I wouldn't expect a visit. Though if your buddy did find himself in the northern VA area I could offer him all you can eat steak and beer.
 
You must log in or register to reply here.
Back
Top