Versaframe out of spec?

We have been busy with machining, and on some lightening cuts that go .075 into .1 versaframe, we are having issues with it being fine in some places, and going all the way through in others. Is anyone familiar with versaframe ever being out of tolerance?

“Out of tolerance”…which tolerance?

Vex doesn’t really provide a wall thickness tolerance. Let’s say that’s not the issue though, since aluminum extrusions can easily be within 0.010".

Vex also doesn’t guarantee a straightness per foot tolerance, so while the tube may not be perfectly straight, that’s also not a spec.

So…you’re asking for a resulting 0.025" wall thickness. How flat is your table or how true is your spindle? Are you sure it isn’t your setup that is causing the overcut?

1 Like

We had it on multiple pieces and it’s being done on a nice machine so that’s not the issue. (If that’s your question)

How nice is the nice machine? Router? Mill? Do you have a surface plate and indicator to check it? Is there a pattern to the overcuts? How sharp is the tool?

Just a few details that will help you get help on the matter.

1 Like

Can you provide more info? Which dimensions were off and how off were they?

I would check if whatever you are using as work holding is level with the spindle. Also, it doesn’t matter how nice or expensive your machine is it can still be out of calibration.

$30,000 Tormach 1100. Using a digital probe to zero and check levelness.

We checked every parameter that could cause an issue. To make sure it wasn’t us we took a piece of .125 and made cuts .115 into it across the entire xy axis area with no issues.

Did you take a caliper and measure the wall thickness?

Seems like the wall thickness is off. Tomorrow we are going to cut cross sections of those pieces and measure the wall thickness where the cuts went through

Something makes me doubt your setup for 0.125" and the versaframe were the same. When you say you cut 0.125" across the whole area, how were you holding it? Clamps to the table or some waste board?

When you hold the versaframe, how is it being held? Single vise? Multiple clamps to the table?

I can imagine a scenario where a fast feed rate combined with a possibly less-than-sharp tool causes the versaframe to bend, then as you machine the area it springs back as the material thins and hence cuts through.

I can also imagine a scenario where an overhanging piece bends and causes a similar overcutting problem.

Also, did you consider the parameter of area moment of inertia when accounting for “every parameter that could cause an issue”? Bending of a sheet against the table can be very different from a tube in a vise or even a tube clamped to a table.

The .125 was 2x1 put in a vice.

Feedrates-slow as hell, tool-new

I do not do the CAM. It’s done by a mentor and if he says he accounted for everything I trust him on it.

We’ve got a big machine shop near us that water jets our bellypan and all the gussets. I will give him a call and I’m sure he will be able to either tell us we are doing something wrong or it’s not our fault.

If you checked the levelness on a Tormach and still had issues, I’ll bet it was the tube. I’d be interested in seeing the results of the caliper check tomorrow.
The difficulty of machining Versaframe and getting holes to all line up made us move to unmarked 2x1s this year.

1 Like

I’d say check the vise trueness to the spindle and how far you are overhanging the work.

You still haven’t told us what is over cut and what isn’t. If the problems are far from your vise on both sides, bending is something to look at. If the problems are to one side of the vise, then vise trueness (parallelism) to the table is something to look at. If the problems are all over, then feed and speed or actual wall thickness are things to look at.

Hey Alex, our team also had similar issues with some tubing being out of spec. We emailed pro support with measurements and pictures and they were very helpful. I believe they have a new supplier and are working out the kinks with them.

1 Like

@AlexG how long was the piece you were machining? The longer the piece the more you’ll notice changes in wall thickness and straightness.

As stated above, any extruded dimension (so everything except the holes) are going to be with 0.010".

So with a material thickness tolerance of .010 and taking a 0.075 cut on a .1 extrusion. With everything else being perfect you have potentially 0.015 left. Add in the “waviness” (Variation if the 1" dimension) of 0.021 (If am reading the standard for aluminum extrusion correctly) You are going to mill through in places even with your perfect mill.

Out of curiosity, exactly how much weight do you expect to save with this operation? You would need to remove 68 square inches of material at that thickness to save half a pound of weight. Honestly, that seems like a lot of work for very little gain.

Further, what do you expect the 0.025 wall thickness to do for you? It might look cool, but I doubt it’s actually going to provide much, if any, structural support on your robot. That’s almost 1/3 the thickness of 1/16 material… and I can bend that by hand, with no tools needed. If your set on losing that small amount of weight, just go all the way through the piece and be done with it.


Can you measure the actual cut at the points where the .075 DOC is going all the way through? You can’t rule out problems with your manufacturing setup, and since you have a cut right at the area of concern you can just measure with a depth gauge or something right there.

Things that can explain what you’re seeing:

If you are machining outside of the immediate vicinity of the vise it’s possible that the cutting forces are pulling the tube up into the cutter. Or, frankly, any compliance in the part setup.

Extruded tubing can have the inner die wander relative to the outer die, creating wall thickness variations.

Any chips present on the parallels or whatever you’re using for a z-stop.

Straightness tolerances of typical extruded aluminum is 1 to 5% of the bar’s length.

Z-difference if using multiple vises.

Z-probe caliper miscalibration.
Improperly set z-zero.
Improperly chosen tool in tool library.
Improperly entered tool offset in tool library.
Chips/dent/contamination on TTS holder.
Loose TTS holder.

I’ve got a Tormach 1100 in my garage, they’re not fool-proof :slight_smile:

More to the point: the published Vex drawing for this tubing has no tolerances on it, so the material can’t be out of a tolerance that does not exist.

1 Like