MPCNC: Corner Post Shims

The rail height measurements suggested three shims could level the MPCNC rails:

MPCNC Rail Height - 2017-12-23

MPCNC Rail Height – 2017-12-23

The numbers inside the lower square give the additional height required to sorta-kinda level the result, keeping in mind we’re not dealing with a particularly stable mechanical setup.

The figures in the lower right translate sensible metric values into mils. I plucked those sheets from my brass shimstock selection, taped them together into a 42 mil stack, and introduced them to Mr Bandsaw:

Sawing MPCNC Corner Post Shims

Sawing MPCNC Corner Post Shims

The sacrificial sheet underneath the stack prevents bending. Using the saw (with a 24 tpi blade), rather than tin snips or scissors, produces a nice clean flat cut without any curling or bending.

A brief conversation with Mr Drill Press created screw clearance holes:

Drilling MPCNC Corner Post Shims

Drilling MPCNC Corner Post Shims

N.B.: Brass is fiercely grabby, so don’t use an ordinary twist drill. Blunt ’em if you have a spare set of drills, but a step drill works for my simple needs, shallow holes, and infrequent drilling. In any event, don’t hand-hold the sheets, because they can turn into whirling knives without the formality of warning you first.

I bandsawed the holes into slots, so I could slide the shims under the corner posts without completely removing the screws, in the hope the posts would stay more-or-less in the same place. Probably doesn’t make any difference:

MPCNC Corner Post Shim

MPCNC Corner Post Shim

Looks like I overtightened the post clamp screw a bit, doesn’t it? So it goes with 3D printed parts.

Another round of measurements with the shims in place:

MPCNC Rail Height - 2017-12-25

MPCNC Rail Height – 2017-12-25

The numbers on the outside of the bottom set give the difference from the lowest rail in each direction, the inner numbers are the average of the two differences in each corner.

All of which seems to indicate the pen height now varies by a smidge over 0.1 mm across the span of those 16.5×14 inch plots.

A plot with all the legends and traces at Z=-0.25 came out OK:

MPCNC - Leveled plot at Z -0.1

MPCNC – Leveled plot at Z -0.1

The legend in the upper left looked slightly faint:

MPCNC - Leveled plot at Z -0.1 - legend detail

MPCNC – Leveled plot at Z -0.1 – legend detail

The upper right legend looks about the same, suggesting my average of differences probably isn’t meaningful.

Lowering the pen to Z=-0.25 should darken the traces a bit and reduce the effect of any inconsistencies in the tool length probe switch.

Not, of course, that this will make much difference in actual use; a router will probably shake the whole thing out of alignment in a matter of seconds.



  1. #1 by Jason Doege on 2018-01-05 - 10:15

    What material are your rails made of. I go looking at home depot and what I see in this category looks like galvanized steel with a pretty bad surface finish. I was hoping for extruded aluminum but they have none (at least, not in Texas.)

    • #2 by Ed on 2018-01-05 - 11:50

      Ordinary galvanized steel EMT, nothing fancy at all. The surface finish is terrible, there’s no roundness spec, and you’d never use it for a precision device.

      For a medium-size router frame, though, it works fine.

      The MPCNC doc (or forum?) discusses why aluminum cylinders aren’t stiff enough. Basically, EMT hits the sweet spot of good-enough, available-enough, and cheap-enough.

      • #3 by Jason Doege on 2018-01-06 - 16:59

        Ah. I missed the aluminum versus EMT discussion. That is useful to know. Thank you.

        • #4 by Ed on 2018-01-06 - 18:29

          There’s also some notion the high point loading from those bearings will mash flats along aluminum rails and produce a sloppy fit. If the loading just mashed down the high spots, it might produce a better fit, but I can see a definite problem if you started with a good overall fit.

  2. #5 by Vedran on 2018-01-05 - 11:12

    Don’t bother Jason. Pipe-for-linear-guides style machines don’t make sense now that you can buy semi decent linear rails from China for really low price. You can even go with HiWin style rails for not much more. No matter how crappy they make them I’m sure they’ll outperform non-precision tubing from home improvement store any day of the week. Printed parts don’t stand a chance when you bring router into the picture.
    I was milling down an aluminum heatsink the other day and caught the thin (60 thou) fin by mistake. Bit drew it in, crashed, then proceeded to move the table along the locked down axis, move my vise bolted to the table, tilt the milling head about 2 degrees off level and chipped the bit just for good measure. Vise was secured with 6mm screws, head with two 8mm screws. All that on a 400lbs cast iron machine, with only 200W milling head at around 1400RPM.
    I really like what Ed’s doing and as a learning platform for pens, drag knives or lasers pipes can work. On the other hand you can get the aluminum extrusion laser cutter (with 2.5W laser unit) in DIN A3 size for around 250$ delivered. Omit the laser and price drops under 200. Unless you’re in it for the building experience it just doesn’t make sense.

    Ed, do you have BOM or at least the idea of how much MPCNC set you back?

    • #6 by Ed on 2018-01-05 - 12:00

      The two parts kits came in a bit over $400 direct from Ryan, the bench was $200 from Lowe’s, and various bits & pieces like the Protoneer CNC shield and the (yet to be seen) Dewalt router added another $200 or so. If you were willing to print the parts, scrounge the other hardware, and build a scrap bench, figure well under $400 all in.

      Two pairs of meter-long linear rails would set you back a kilobuck, which is the charm of EMT: it’s Good Enough for the purpose.

      • #7 by RCPete on 2018-01-05 - 14:54

        For a notch further up, look into aluminum extrusions. I used some surplus extrusions from a wafer fab conveyer for a sander frame. Angle connectors and proprietary t-bolts (from the original) worked well enough. Hex head cap screws could be used, too.

        An extrusion from a solar panel mount system (Iron Ridge XR 1000) is running about $5.50/foot from sellers like solar-electric dot com. Lengths run 5.5′ to 17′. Might be good for the bottom rail, with something else for the cross rails. I’d wonder about drill rod for those, maybe with linear bearings. OTOH, I’d not want a router near those bearings. (I’m using the I-R rails on my solar system; no problems, though I’ll clear snow if it gets too thick.)

        • #8 by Keith Ward on 2018-01-05 - 16:35

          Linear rails would be pricey for sure from many high-end companies but apparently they are not as high as they used to be from even McMaster. Ordinarily I would lean towards using drill rod instead, but looking at the prices on various types of actual linear rails is very comparable to drill rod and probably worth the extra cost as they would be consistent all the way to the ends. I have found drill rod to often be out of round when cut on the ends and not to be relied upon. They must shear them instead of sawing in the smaller diameters … that is unless you buy it from Starrett :-O

          • #9 by RCPete on 2018-01-05 - 22:15

            I was told (maybe 20 years ago) that the roundness of drill rod can be spotty for high precision work; they are (or were) produced by centerless grinding. Haven’t bought anything large in a while, and the fine stuff (1/8″) was good enough.

            For a router application, I’d be worried about keeping flying wood crud out of the works..

      • #10 by Vedran on 2018-01-08 - 09:45

        I’m not talking nice stuff here :)
        12$ / meter 20mm rails with supports – and similar for the bearings or
        51$ /1.5m MGN9 knock offs

        I still think it must be way better then pipes – way less bendy at least:)

  1. MPCNC: Sakura Micron Pens | The Smell of Molten Projects in the Morning

Spam comments vanish. Comment moderation may cause a delay.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s