[imp22b]

I just re-leveled and checked tram on my Tormach 1100 after I moved its location in my garage. I used a Edge Technology Pro Tram to get my measurements. I found that my X is out about .0005 and my Y is out about .0015 over 6". I am OK with the X axis but I would like to improve the Y axis. I have previously shimmed the base of my mill when I first bought my mill to tram in the X but I was never able to adjust the Y. It is my understanding that due the column being pinned it is difficult to adjust the Y tram.

Is there a good way to tram the Y?

Should I really be worried about the .0015?


Attachment 326902

[CadRhino]

I would refer to the original factory tram specs that came with your machine. You should be able to tram your machine to those specs. There are some videos on YouTube for tramming. I used brass shims.

I just re-leveled and checked tram on my Tormach 1100 after I moved its location in my garage. I used a Edge Technology Pro Tram to get my measurements. I found that my X is out about .0005 and my Y is out about .0015 over 6". I am OK with the X axis but I would like to improve the Y axis. I have previously shimmed the base of my mill when I first bought my mill to tram in the X but I was never able to adjust the Y. It is my understanding that due the column being pinned it is difficult to adjust the Y tram.

Is there a good way to tram the Y?

Should I really be worried about the .0015?


Attachment 326902

[109jb]

You can't just use something like the pro-tram to square and tram a milling machine like the Tormach. As good as a pro-tram is, it only tells you if the spindle is square to the table. If it isn't, it doesn't tell you why. It could be the column not square to the table, or the head could be tilted on its slide....

You should first be checking the column to table squareness by using a precision square clamped to the table with one leg vertical. You then mount an indicator on the head and run the head up and down to check the column to table squareness. You do this twice. Once with the leg clamped to the table aligned in the X direction, and then with the leg clamped in the Y direction. Once that is done, then you can use the pro-tram to check spindle to table squareness.

Also, make sure the gibs are adjusted properly. If the slide is sagging due to loose gibs it will throw the spindle out of square.

[CountrySmith]

You can't just use something like the pro-tram to square and tram a milling machine like the Tormach. As good as a pro-tram is, it only tells you if the spindle is square to the table. If it isn't, it doesn't tell you why. It could be the column not square to the table, or the head could be tilted on its slide....

You should first be checking the column to table squareness by using a precision square clamped to the table with one leg vertical. You then mount an indicator on the head and run the head up and down to check the column to table squareness. You do this twice. Once with the leg clamped to the table aligned in the X direction, and then with the leg clamped in the Y direction. Once that is done, then you can use the pro-tram to check spindle to table squareness.

Also, make sure the gibs are adjusted properly. If the slide is sagging due to loose gibs it will throw the spindle out of square.

I first checked my 770 spindle for parallelness to the z axis ways. I used Rollie's Dad's method for doing so (http://www.cnczone.com/forums/bencht...ist-posts.html). I mounted a 3/4" bar of uniform roundness and diameter in the spindle and checked the alignment in the xz plane and the yz plane.

A digital dial indicator was mounted on the table and measurements made at two different head heights about 6" apart. An adjustment in the xz plane can be made by pulling one of the tapered pins out and rotating the head as necessary. To facilitate that movement, I made a device for controlled rotation of the head. For an adjustment in the the yz direction, the head has to be shimmed to correct misalignment. Fortunately, the yz alignment was close enough to ignore.

Once the spindle axis is verified parallel to the z axis, the mill can be trammed. Corrections for tram will have to be done by shimming the base of the column.

R J

[109jb]

I first checked my 770 spindle for parallelness to the z axis ways. I used Rollie's Dad's method for doing so (http://www.cnczone.com/forums/bencht...ist-posts.html). I mounted a 3/4" bar of uniform roundness and diameter in the spindle and checked the alignment in the xz plane and the yz plane.

A digital dial indicator was mounted on the table and measurements made at two different head heights about 6" apart. An adjustment in the xz plane can be made by pulling one of the tapered pins out and rotating the head as necessary. To facilitate that movement, I made a device for controlled rotation of the head. For an adjustment in the the yz direction, the head has to be shimmed to correct misalignment. Fortunately, the yz alignment was close enough to ignore.

Once the spindle axis is verified parallel to the z axis, the mill can be trammed. Corrections for tram will have to be done by shimming the base of the column.

R J

That method will also work, but the runout of the bar mounted in the spindle can affect the outcome. I have seen very few spindles/collets/chucks that would not show some runout at 6" below the spindle face. You are essentially relying on the bar in the spindle to align the column. As said, it will work, but the first step would be to check runout of the bar.

I prefer the method of using a precision square to square the column first as this only relies on the precision of the square used to align the column. This isolates the column squareness and ignores head tilt until after the column is squared.

Then using the pro-tram type device introduces no more error because the indicators are zeroed at the same location on the table, essentially aligning it to the axis of the spindle bearings.

To each his own, but the method using the square is how I would and do align the column on my mill.

[SCzEngrgGroup]

Another problem you're likely to face is finding shims thin enough to allow a more precise adjustment. Since the column tram is adjusted by placing shims at the mounting flange, a very thin shim will have a relatively larger effect on tram at the spindle. So, a 0.001" shim may alter tram by several thou. The 0.0015" you're seeing now may well be about as good as you can get it.

Regards,
Ray L.

[CadRhino]

+1 I purchased a protram and several 2-4-6 precision blocks and a brass shim set.

That method will also work, but the runout of the bar mounted in the spindle can affect the outcome. I have seen very few spindles/collets/chucks that would not show some runout at 6" below the spindle face. You are essentially relying on the bar in the spindle to align the column. As said, it will work, but the first step would be to check runout of the bar.

I prefer the method of using a precision square to square the column first as this only relies on the precision of the square used to align the column. This isolates the column squareness and ignores head tilt until after the column is squared.

Then using the pro-tram type device introduces no more error because the indicators are zeroed at the same location on the table, essentially aligning it to the axis of the spindle bearings.

To each his own, but the method using the square is how I would and do align the column on my mill.

[CountrySmith]

That method will also work, but the runout of the bar mounted in the spindle can affect the outcome. I have seen very few spindles/collets/chucks that would not show some runout at 6" below the spindle face. You are essentially relying on the bar in the spindle to align the column. As said, it will work, but the first step would be to check runout of the bar.

I prefer the method of using a precision square to square the column first as this only relies on the precision of the square used to align the column. This isolates the column squareness and ignores head tilt until after the column is squared.

Then using the pro-tram type device introduces no more error because the indicators are zeroed at the same location on the table, essentially aligning it to the axis of the spindle bearings.



To each his own, but the method using the square is how I would and do align the column on my mill.

I initially tried your method but realized that there were some issues. First, I have no idea of the parallelness of the table surface the the xy plane, as determined by the x and y way axes. Second, I do not know the degree of flatness of the table surface. Third, I had no good way of verifying the squareness of my square and compensating for any errors.

The sheer beauty of the RDM method is it is independent of spindle runout. When you rotate a truly round bar to find the minimum and maximum and the average the two, you find the position of the spindle axis. When you move to a new position on the bar and repeat process, you again find the position of the spindle axis. In the case of a lathe, where the indicator is mounted on the carriage and the carriage is move on the ways, the difference in the averaged positions will give the degree of parallelness of the ways to the spindle axis. In the case of the PCNC, the spindle moves on the z axis ways while the indicator remains fixed in space.

Below is a drawing of measurements that would be made on a bar where there is both radial and angular runout. You will note that the the average of the readings defines the true position of the spindle axis.
Attachment 327110

Once the spindle is trued to the z axis ways, the tramming tool can be used to correct the column perpendicularity to the xy plane. Corrections would be done by appropriate shimming of the column base. prior to tramming, I would mount a plate to the table and face the surface. This insures that the tramming surface is parallel to the xy plane. I would use a small end mill to minimize any tramming error. It is not necessary to have a complete plane, only spot facing at the four cardinal compass points.

My reason for this approach is when I mapped the table surface, I found some disturbing irregularities. The graph below shows the table surface position relative to a plane parallel to the xy planes and zeroed in z at the x & y homing position. The different colored lines are for different y positions as measured from the home position.
Attachment 327112
As I was analyzing the results, I realized that at about x = 10", the weight of the table, stepper, etc. lifted the far end of the table against the play in the x axis ways. Adjusting the ways dramatically decreased that lift but it necessarily exists because of required clearance. Even with the improved status of the table, I still felt that the irregularities were great enough to not trust the square for column alignment which is why I did what I did.

I have not yet aligned the column as I needed to get a supply of shim stock in hand first.

Were I to use a square to align the column, I would mill a secondary surface to insure that the surface of the square was parallel to the xy plane. In my case, I would perform this as close to the left hand side of the table to minimize any lifting of the table. Additionally, it might be prudent to temporarily over tighten the x and y gibs to minimize any play and ensure that the way opposite the gib is in good contact with the table. I also would like to critically examine the table surface to see if the nonlinearity that I observed is due to way play, flexing of the table (there will be some), or the table simply has a bow in it. My approach to this would be to put bluing on my surface plate and look at the transfer pattern to the table surface.

R J

[109jb]

Verifying the X-Y plane of the table is trivial. Mount an indicator base on something other than the table and drive it around. If the table top surface is parallel to the x and y axis ways then the indicator will not move. If it does, you have bigger problems that need to be addressed before worrying about column or head tram.

As for the squareness of a square, they are made to specified tolerances. For example, a 10" x 6" x 1" thick granite square has a guaranteed accuracyof 0.0001". But if you don't have a granite square you can also check the squareness of any square using only one axis of a mill. Mount a straight bar, 246 block for example to the table with the edge aligned with the y axis. Clamp the square lying on its side to the side of the 246 block. Indicate the other edge of the square in with the x-axis and lock down the 246 block. Now just flip the square and indicate the edge . This will tell you how square the square is.

Attachment 327138

[MichaelHenry]

Another problem you're likely to face is finding shims thin enough to allow a more precise adjustment. Since the column tram is adjusted by placing shims at the mounting flange, a very thin shim will have a relatively larger effect on tram at the spindle. So, a 0.001" shim may alter tram by several thou. The 0.0015" you're seeing now may well be about as good as you can get it.

Regards,
Ray L.

Some guy on Ebay sells sets of paper shim in thicknesses of 0.0002, 0.0005, and 0.00075".

20 ft Precision Shim Stock 0002 0005 and 00075 Inspection Moore Jig Bore Etc | eBay

[rowbare]

Another thing you can do is play with how tight the base is clamped to the stand. A half turn on one of the nuts might just remove enough twist to get it where you want it.

bob

[imp22b]

Thanks everyone for the tips. I'll give them a try.