Mini Mills, Bench Mills. Cutting aluminum extrusions square.

Hi all. I have come to seek your collective wisdom prior to buying a manual milling machine for the purpose of precisely cutting 2020, 4040 extruded aluminum profiles and possibly some small milling of brackets/ peripherals for 3d printers and CNC building, or making precise cuts on soft metals that are not possible with a bandsaw.

Over the past year I have struggled with ordering extruded aluminum frames to specific lengths only to have them arrive 0.2mms out of spec or not cut at perfect right angles leading to bad times when building cubes. I've basically bought all the tools necessary for a small fabrication shop twice (bought cheap then learned about starrett, mitutoyo, using center finders, transfer punches). I have been avoiding buying a mill and lathe but the time has finally come in my maker journey to get the tools that can make other tools.

I am looking at the mini mills from the littlemachineshop due to their non-tilting head columns (and transparent business practices). Would the cheapest one (the 3990 hitorque minimill) be suffice or overkill for this task? or could I get away with one of the micromills for the task of squaring the ends of 2020 extruded aluminum ends, such as the one from proxxon? If anyone has any experience with these mills for the purpose of small hobby fabrication/protoyping I would like to hear your advice.

My plan if the item im building exceeds the area of the mills build table is to use blue layout fluid to pre-mark all my cuts and just take my time.


I look forward to any responses.

I always recommend taking a look at the g0704 mill because it is so popular. It sits at at price/performance sweet spot, and there is a ton of support for it. The mill you linked will also be fine. I wouldn't go as small as a micromill since you will want to use endmills long enough to clean up 4040 without flipping it over.
The main drawback of machining is that accessories (cutting tools, workholding, metrology) will often cost as much as the mill, and fully equipping yourself from scratch can be even more expensive.

I'm not sure that your expectations are realistic though. Cutting a length of extrusion to 50cm +-0.2mm (for example) is a measuring problem, not a cutting problem. Reliably scribing and then hitting that dimension is not trivial. You could skim it off on the mill, measure the length with calipers, and then trim the excess; but then you need a pair of calipers long enough to measure every piece you cut.


One of the advantages/disadvantages of extrusion is that there are so many degrees of freedom during assembly. Your goal should really be to have everything close enough to look nice, and then tap it all into alignment during assembly. Even if your extrusions are all perfectly sized you will still need to perform alignment during assembly anyways since the brackets and such are not perfectly repeatable.

If you want to build a perfect cube from extrusion you should be looking at surface plates, squares, levels, rubber mallets, calipers, straightedges, indicators, etc... (IMO) On a three axis robot you generally have about five alignments that actually matter, and you should measure and adjust them directly when possible. Everything else is just cosmetic.

Edited 1 time(s). Last edit at 12/12/2017 12:47PM by 691175002.

Usually absolute length of frame pieces isn't critical, but matching lengths is - that's how you get square assemblies (that and square milling of the ends). If the mill has a long enough work table, you put a stop on it and use that to mill pieces to matched lengths. I set up stops on a bridgeport mill to finish t-slot pieces and get them to matched lengths easily within 100 um.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

You will find having a minimill extremely useful. I can vouch that LMS is a good company to deal with and are basically the only company with a stock of replacement parts for their mills and others. I have purchased parts and accessories only. The Grizzly is fine too.

In machining repeatability is more important than accuracy in most cases. The extrusions themselves make great gauges. You can easily make a gauge that will measure multiple parts to great accuracy to ensure they are all the same.

Quote
691175002
I always recommend taking a look at the g0704 mill because it is so popular. It sits at at price/performance sweet spot, and there is a ton of support for it. The mill you linked will also be fine. I wouldn't go as small as a micromill since you will want to use endmills long enough to clean up 4040 without flipping it over.
The main drawback of machining is that accessories (cutting tools, workholding, metrology) will often cost as much as the mill, and fully equipping yourself from scratch can be even more expensive.

I'm not sure that your expectations are realistic though. Cutting a length of extrusion to 50cm +-0.2mm (for example) is a measuring problem, not a cutting problem. Reliably scribing and then hitting that dimension is not trivial. You could skim it off on the mill, measure the length with calipers, and then trim the excess; but then you need a pair of calipers long enough to measure every piece you cut.

Yeah definitely exagerated the minute-ness in differences in length from the extrusions i received from open builds, they were probably an entire milimeter off on some of the pieces. And they were parts that were intended to run parallel to each other.

Quote
691175002
One of the advantages/disadvantages of extrusion is that there are so many degrees of freedom during assembly. Your goal should really be to have everything close enough to look nice, and then tap it all into alignment during assembly. Even if your extrusions are all perfectly sized you will still need to perform alignment during assembly anyways since the brackets and such are not perfectly repeatable.

If you want to build a perfect cube from extrusion you should be looking at surface plates, squares, levels, rubber mallets, calipers, straightedges, indicators, etc... (IMO) On a three axis robot you generally have about five alignments that actually matter, and you should measure and adjust them directly when possible. Everything else is just cosmetic.

Thank you for the reply I had not considered the g0704 due to random people on machinist forums complaining about grizzly and harbor freight in general but in recognizing your user name and knowing you are knowledgeable on these matters I am definitely taking a good look at it. It looks like a really good deal at that price, and the shipping charges make it a steal (even if i have to risk dealing with some quality control issues). From what I understand they are all made by sieg anyways. And Yes I have most of the tools your have listed except for a calipers long enough to measure 50 cms. My plan to get around owning calipers that big was to use some video microscopes I have around and measure extrusions from an anatomical superior view and then laterally with two separate video microscopes at the same time. not sure if it will work but I'm going to try it.

Quote
the_digital_dentist
Usually absolute length of frame pieces isn't critical, but matching lengths is - that's how you get square assemblies (that and square milling of the ends). If the mill has a long enough work table, you put a stop on it and use that to mill pieces to matched lengths. I set up stops on a bridgeport mill to finish t-slot pieces and get them to matched lengths easily within 100 um.

I wish i could buy a bridgeport, but that would probably also require owning a fork lift, getting 3 phase power, and paying divorce fees and a bunch of other stuff I don't want happening. How are you verifying the 100 microns?

I don't generally. The DRO on the machine is good to +/-0.0005", and my frames come out square when I bolt them together. I always check the squareness of the end finishing by standing the pieces on a granite slab next to each other after deburring them. I've never had to recut any of them.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]