[bearer]

So I've had one of these 3040 machines sitting around for a while. Upgraded with a G540, so it runs pretty well after that.

But I'm having trouble with accuracy, and I've probed out the table and found it to be almost +/- 0.5mm.

Attachment 294550

As the picture shows, it seems the x/y plane is twisted with respect to the z axis. (30x30mm grid, 5 samples, 0.03mm std dev, so data should be ok i think)

I recon this will be the start of me learning how to do this, but I don't speak fluent CNC so I'm not sure what i'm supposed to search for to find out how to do it. I'd much appreciate some pointers on what I need to do, and where I can find more information on how to get it done.

Input will be much appreciated!
-Bjorn

[Fixittt]

first thing I did was lay a straight edge across the table surface. left to right my table is very wavy. High and low spots. This is because the extruded table top has bots about 2 inches apart holding it down. Every where there is a bolt, there is a low spot. Every where there is not a bolt there is a high spot.

Lay a piece of glass on the table and re probe it. my gantry was severely tilted on the right side. I had to use .030 inches worth of shim stock (Feeler gauges) to raise the right side. The shims where places under the right upright support.
Also notice in the spindle picture above the 4 black bolts, there is a piece of shim stock horizontal to bring the top of the spindle outwards to help get it square to the table.

Used this method to check to see if the spindle was squared to the table. Shimmed where needed. When doing this I had a plate of glass on the table top due to the wave in the table.

https://www.youtube.com/watch?v=6tMw7GUpWvY

[bearer]

Well, it took a year, but a glass pane did indeed smooth out the measurements, but the same basic warp was still present. Made made some holes and pockets in a plastic sheet (chopping board), bolted it down and milled it flah-ish.

Then I hotglued my indicator to an Allen key and put it in the collet and placed some feeler guages around the spindle mount to get it as tram as I could against the glass while spinning the Allen key. Removed glass and cut another pass of the cutting board, and its alot better. Think I need to look at my feeds and speeds to make the surface finish better now.

Wondering if I could mirror the front section of the machine and try to cancel out the warping. Its not quite symmetrical, but it could almost cancel out. Any thing to look out for if trying to remove the thick alu plate in the front?