Tramming a small mill (Sherline mill)

[ttrikalin]

Mill tramming may be too mundane a topic for your graces, but what the heck, today it was mill tramming day and I think the first time I really got it OK.

Jerry G, webmentor to me and a bunch of younger and older sobs, suggested that all poor souls with a mill learn how to tram it properly... Everybody in class did it, but me. Now don't get me wrong, I'm not a slacker :-[ but life caught up, and a friend is visiting, and the car's sticker has expired, and a paper got accepted, and another is due, and a haircut is due even longer... so the mill mills happily, if untrammed, and me parts keep getting almost there but not quite there...

So today I decided to do what's most important and tram the little mill, a Sherline 5400. Here I follow what I understood from Jerry G, Dan D, and numerous other mates... I guess I'll blame them for what I did wrong (including misunderstanding what they wrote... I'm in me blaming mode tonight ).

First, I got me self a nice wheel bearing. Brought donuts to the guys who did the sticker renewal and asked for a scrap wheel bearing outer race. They were so nice and obliged... Gave me one with a huge scar inside the bearing's outer race... I shiver to think what happened to the balls...

I first filed some raised parts of the inner edge from the scarred side, and then went on using wet/dry paper. I moved the bearing in an "8" motion, on 600 grit paper after applying a mix of WD40 and some 3 in 1 oil... Gently...

After doing that on both faces and seeing that I had nice, "fresh" surfaces I miked to make sure that the surfaces are parallel. At first they weren't -- cause I was not holding the mike carefully and perpendicular to the bearing... oh: after making sure that I knew how to measure, I found out that the faces were perfectly parallel. Of course.

Then I mounted the 4 jaw chuck on the spindle and held my DTI so that it runs on the bearing race (diameter ~3").

And then I started indicated what was widely held to be a decently trammed mill (that is by me and Anastasia). Well, setting 12 o clock to be the most distant part of the y axis and the reference height, 3 o clock was +0.002", 6 o clock was +0.003" and 9 o clock was 0.001" - all points lower than the reference. I repeated the measurements 4 times.

BTW -- I have marked the internal part of the bearing at 12, 3, 6 and 9 o clock using a blue sharpie and the rotary table...

... and I had locked the Z axis and put a carton wedge so that there is no play and the rotation is only clockwise (looking from above).

I first trammed the X axis so that 3 and 9 o clock read the same number: 0.0015". This is done by loosening the 4 screws that hold the Z column to its base... (no shown). So the problem remained the Y axis... The front part of the base needs to be lifted up a bit... So let's shim the base. After trying paper, coca cola cans and what not I went to the hardware store (JJ Rounds) where Bruce gave me stuff, including a feeler gage...

Trying different thicknesses, it seems that a shim of 0.0015" does the trick. So I cut two small pieces from the feeler gage :hDe:

Who cares .

Here is the little piece under one side of the column base, in the front - barely visible. There's another one in the other side.

That did the trick. Here's 6 o clock (Sorry for the parallax error, I did not shoot the dial correctly) - 0.0000 -0.00025":

Here's 9 o clock. Again 0.0000 -0.00025".

Here's 12 o clock- through the dental mirror I got at Walgreen's.... Again 0.0000 -0.00025".

And finally, 3 o clock. A bit off -- 0.0005" - but who cares at this point... -- not me.

These measurements were repeated several times, and were stable within 0.00025". So I wore a smile on me face and went with my lovely wife for the beer I deserve.

If you see blunders let me know and I promise to put them in the booboos section

take care,

tom

 

[Deanofid]

We might call it mundane, Tom, but it's something we all have to do if we want to mill our stuff flat!. This is a good subject, and even if we might consider it "basic", it's not for people who don't know how to do this.

Using the bearing race looks like a good idea, as long as you know the ends are parallel. You can keep it forever as your master reference.

I would make one suggestion. Your DTI will be more inclined to show tiny variations if you position the probe nearer to a 90 deg angle to the work. Just a thought.

Thanks for starting the thread. Maybe others will chime in with their method.

Dean

 

[Russel]

First of all I want to state that I'm no expert. I've had my Sherline equipment for a few years, but until recently, really haven't used them much. So, be sure to correct me if I'm wrong.

Anyway, as I understand the procedure spelled out in the Sherline instructions, the first thing that you should do when tramming the mill is to sweep the table in the Y axis to measure tilt of the table. Apparently and error of .001" or so is not unusual and must be accounted for when adjusting the Z axis. I know the last time I drilled a hole (that was supposed to be straight) at an angle, I carefully went through the process a few times to confirm repeatability. My Sherline 2000 mill table is .0015" higher in the back than it is in the front. What I ended up doing to adjust the Z axis with regard to front/back (Y axis) error, was to move the table all the way back and tram off of the base below the table saddle. That worked well for me with repeatable results.

Using a bearing race is an excellent idea. What I'm thinking is that if I carefully grind groves in the bottome edges of the bearing race, it could sit on the mill base with the Y axis lead screw passing through the grooves in the race. Then I could sweep the top of the bearing race 360° and use the base, not the table, as a reference.

Russ

PS I almost forgot: Thank you for posting such a well illustrated explanation of tramming!

 

[ttrikalin]

I would make one suggestion. Your DTI will be more inclined to show tiny variations if you position the probe nearer to a 90 deg angle to the work. Just a thought.

Dean,
this is probably correct... Not sure how much of a practical difference it would make - but no reason not to do it neatly. I have yet to make a small thingy to hold the DTI properly. This was a let's-try-to-do-with-what-we-have-here (and finish today) job...

The bearing is not my idea... But bringing donuts to the sticker guys and then ask for it was my idea . Well I kind of cheated there too... I heeded advice from Click and Clack from car talk...

take care,

tom

 

[ttrikalin]

First of all I want to state that I'm no expert. I've had my Sherline equipment for a few years, but until recently, really haven't used them much. So, be sure to correct me if I'm wrong.

Russ, I'll join you in being non expert ;D ;D

Anyway, as I understand the procedure spelled out in the Sherline instructions, the first thing that you should do when tramming the mill is to sweep the table in the Y axis to measure tilt of the table. Apparently and error of .001" or so is not unusual and must be accounted for when adjusting the Z axis. I know the last time I drilled a hole (that was supposed to be straight) at an angle, I carefully went through the process a few times to confirm repeatability. My Sherline 2000 mill table is .0015" higher in the back than it is in the front. What I ended up doing to adjust the Z axis with regard to front/back (Y axis) error, was to move the table all the way back and tram off of the base below the table saddle. That worked well for me with repeatable results.

But Russ, work is mounted on the table and thus it is the surface of the table that we should use as a reference... no?
At any rate, tramming using the base of the mill as a reference will work as long as the table surface is reasonably parallel to the base, and it is. So I'm not surprised that it works fine (i.e., within tolerances for practical purposes).

On a related point, in my machine the areas of the table outside the T slots are ~0.001" lower than the area in between the T slots. Just an observation in support of using the bearing...

Using a bearing race is an excellent idea. What I'm thinking is that if I carefully grind groves in the bottome edges of the bearing race, it could sit on the mill base with the Y axis lead screw passing through the grooves in the race. Then I could sweep the top of the bearing race 360° and use the base, not the table, as a reference.

See above, may not be necessary... Also a difference of 0.0015" over 3" travel is not a deal breaker I think...

My two cents...

tom

 

[ttrikalin]

Jerry G agrees with Dean on refining the process... see below.
I change the title of the post to "tramming a small mill (Sherline)" - to be more generic

Hi Tom,
Better late than never....
only thing I saw that needs refining is the angle of the indicator to the bearing.
Can introduce the cosine error, depending on the make of indicator.......
Later,
Jerry G
P.S.
From the photo , it looks like the balls of the bearing seized up.....

 

[D.C.Clark]

I use a piece of plate glass, 8 x 8 x 1/4 inch thick, clamped to the mill tooling plate, and a plunger type indicator attached to a piece of bar held in a collet. The larger the radius you indicate, the more sensitive the measurement will be.

David Clark in Southern Maryland, USA

 

[ttrikalin]

Nice setup.

did you mic the glass to be sure it's OK? I guess a good way to see that it's surface is nice is to assert that it mirrors in an undistorted way.

Love the bar -- will make one too.

Love the base!

 

[D.C.Clark]

Thanks. There's much more to the base than is visible in the above photo. I've made my Sherline mill a hybrid 2000/4000 machine that can be quickly configured in a number of different ways. I don't believe I've posted anything about it on this forum -- will do so in a new thread. Have to find, or take, some more pics.

The glass I'm using is a piece of borosylicate plate which is optically flat -- about a tenth of a wavelength of light, or a few nanometers. (Originally purchased for a much more exacting application.) But, pretty much any piece of 1/4 in plate glass will be flat to a few microns or better.

Plate glass is also known as "float glass". It's made by floating the molten glass on a pool of molten tin. Makes a wonderful, light weight, inexpensive, surface table. Place it on a scrap of short-pile carpet to support it evenly. A sheet of 1500 sandpaper on top and you've got a lapping and polishing table.

David Clark in Southern Maryland, USA

 

[Russel]

But Russ, work is mounted on the table and thus it is the surface of the table that we should use as a reference... no?
At any rate, tramming using the base of the mill as a reference will work as long as the table surface is reasonably parallel to the base, and it is. So I'm not surprised that it works fine (i.e., within tolerances for practical purposes).

The Z axis should be aligned with respect to the X and Y axis. If the top of the table isn't in line with the Y axis and you use it (without accounting for the tilt) to tram the Z axis, the Z axis will be off.

Here is a very exaggerated diagram to illustrate what I mean:

I am speaking theoretically, if you are making really small parts then the error doesn't really matter. I'm splitting hairs here, just to be sure that I understand it correctly.

On a related point, in my machine the areas of the table outside the T slots are ~0.001" lower than the area in between the T slots. Just an observation in support of using the bearing...

See above, may not be necessary... Also a difference of 0.0015" over 3" travel is not a deal breaker I think...

My two cents...

Good point. Also, much as we care for our machines the table sometime acquires a few scars. The a bearing race can help to deal with that.

[edited to add diagram]