Mini Mill chamfers out of square....or not?

[ssdesigner]

So I have the LMS 3990 mini mill and I've been using it for about a year now. I trammed it as a novice but felt like I had it pretty dialed in. Putting the finishing touches on this solid block of cast iron that I'll be using as a solid tool post with a 3/8 chamfer end mill bit. I did one side using the y-axis as my feed, kept the part in the vice and moved the spindle to the other side to finish both edges. One side looks really good but the other side is off by a noticeable amount. Thankfully this is merely cosmetic, but I'd like to correct. Seems odd that moving 2" on the x axis to do the other side would have it off by this much. Was there something else that could have caused this?

Pics below of both edges.

 

[markba633csi]

You are talking about the width of the chamfer changing from one side to the other?
Or the width being different between the two pictures?
I'm thinking either backlash issue or workpiece not being square- maybe climb cut vs. regular cut

 

[ssdesigner]

The width of the chamfer changed from one photo to the other.

Here is a side by side with reference lines drawn in:

 

[RJSakowski]

A likely cause is an out of tram condition in the direction of the cut. It doesn't take much of a tilt to create a visual defect in a bevel or chamfer.

 

[RandyM]

Are you absolutely certain the block is square?

 

[markba633csi]

Backhanded by backlash?
When you cut on one side of a workpiece then move the X axis to cut on the other side, you have to take up the X backlash
in the opposite direction than you did for the first cut- in one direction you are pushing on the leadscrew, in the other you are pulling it.
Even so, errors creep in. On some machines the Y axis will shift very slightly depending on which way it's traveling
The Induma "sidewinder" machines were sometimes known to do this especially with increasing wear

 

[homebrewed]

If the fixed axis of your mill wasn't locked down, it could have drifted over due to the combination of backlash and cutting forces.

Another possibility is that the work moved during the procedure. How was it held in place?

Another possibility is that the chamfer tool was slowly pushed up into the tool holder while being used. How was it held in place?

 

[PLASTICULTURE]

Taking into account that you mention that you are a "Novice" and I am going to assume that you used a vise to hold the workpiece. You mentioned that this is for a new tool post for your lathe which leads me to believe that the metal block is not tiny. Thus, you may have just put the block into the vise without using parallels under the block when securing it in the vise. The use of parallels is fundamental to help indicate that the material is square in the vise once tightened to hold it. If you can still wiggle the either of the parallels, means that the block is not square in the vise. Don't ask me how I know this because I have never done that before and heard of it from a friend of a friend.
BTW... I 100% agree with the comment from "RJSAKOWSKI" "It doesn't take much".

 

[ssdesigner]

Ok, to answer a few questions:

This started out as a 2” x 3” x 2” block of cast iron. I chopped the 2” dim in half because I only need 30mm height on the tool post to keep all of my quick change heights the same.

I do use a vise on my mill. I do use parallels to hold my parts in the vise. I used a fly cutter to square the largest face first. Then proceeded to square up all sides.

 

[Parlo]

Did you take into account the chamfer size when moving the x axis to the other side? The chamfer width needs to be added to the block width, if this is not accurately known then it could present a chamfer with double the error. Working from the centre of the block with equal +/- X axis shifts should give equal chamfers. In this situation I would consider locking the Y axis and rotate the part 4 times in the vise using the fixed jaw as the cutting side. Climb milling is great for chamfers.