you say that you got the QCW Frame because you were not confident you could roll the bench around and keep the worksurface coplanar. But you don’t mention whether the QCW Frame did fulfil this expectation. When QCW Frame was released I was unsure whether it would be strong and rigid enough in case its underground was not coplanar. I asked myself, if it wasn’t, was that the reason that they released the Any Surface Leveling System at the same time?
You don’t mention whether you checked coplanarity of the four QCW feet. Did you apply one of the methods here and here? E.g. the fishing line method Tom mentioned here the other day is very simple and accurate (note that if the two lines touch each other at their crossing point in the center, you got to swap the two lines so the other is on top. In the end they have to touch each other in both dispositions).
If you say your floor isn’t flat (I know such conditions) you could move the bench around and check if fishing lines distance at their crossing point in the center changes (or by simply putting a piece of veneer under one table foot).
If you lift a corner you do get twist… But we are talking minimal and frankly not unexpected with the span and weight of the journeyman. I think the leveling feet are a great as on for this and wish I had added them.
did you check it on your QCW frame? If you do get twist then, @woodsmoke will still get twist when moving the table around on the floor not being flat, and then you still need to provide a rigid machine base. I find it important to remember that Onefinity CNC is sold without one. Before release of QCW Frame there was simply none offered with it, and since release I was not yet able to check QCW Frame, thus my question. It makes a difference if QCW Frame is the rigid machine base that was missing on the Onefinity CNC or if it is just a wasteboard holder.
It is important to understand that surfacing the wasteboard or the table top with a surfacing bit is in no case able to provide coplanarity of the surface (i.e. removing twist). On a table where one foot of the machine is not in the same plane than the three others, every workpiece that you work on will result in having a twisted surface, even when your check for rectangularity of router axle to table top shows a perfect right angle. The problem is, if you move your swinging arm with the dial gauge to front and back and it shows perfect right angles on the right end of X travel, and with the gantry slid to left end of X travel it shows perfect rectangularity to front and back on this end too, still the table can be twisted and both Y rails not in one plane (not coplanar). On such a table, every workpiece you produce will be twisted, and surfacing will not change anything on this.
This is why I pledge to always do at least the fishing line test over the top rails.
Not directed at Aiph5u, just adding my thoughts to his comment.
Maybe do it first for the QCW frame. Level the frame using the adjustable feet or shims and secure to the table. Then repeat the fishing line (or piano wire) check with the CNC mounted on the QCW frame.
For shims adjustments if you don’t have shim stock a cheap way to get a selection of shims is buy a couple cheap feeler gauge sets. You can find a 32-leaf feeler gauge set with a range of 0.0008 to 0.040" (0.02-1.00MM) for as little as $6 online. Good for shimming your Z axis too. I know there are many choices for shim assortments available, but none are as cheap as this. The shims are larger but for this size machine I don’t think you need the 2"x2" or larger shims.
PS: I said above that I yet did not check the rigidity of QCW Frame. But from how it is made I would assume that it is not able to retain coplanarity by itself considering the weight of the machine (and the load of forces exerted during milling). Its third dimension is not high enough, it is too flat to have enough rigidity (assumption, until I really test it). But I would not be deceived it it is in fact not much more than a wasteboard and T-Track holder (remains to judge on whether the price is justified). What I believe is that the Onefinity CNC from beginning addresses to a target group that is able to build a rigid (anti-)torsion box from wood and I think we should not omit this important step. I also think that a machine base that is rigid to perfectly retain coplanarity if you boost one of its feet with a lifting jack would not only cost much more than the Onefinity costs, but would also weigh multiples of Onefinity weight and would not be able to be shipped and set up that easily. If there exist a target group of users that are able to build a rigid (anti-)torsion box, it makes sense to offer a machine without a machine base. It is one of the reasons we bought it and not another CNC. You make the part that you are used to do yourself and what you buy is the part that you cannot make easily, as a woodworker.
By the way I find the third dimension of some (anti-)torsion box examples seen a bit little in my opinion. I think while I plan I will concentrate on a design of a torsion box of at least eight inches height for Journeyman. On the other hand, it does not really have to be a box of same height everywhere. Eight inches in the center, but the wings could taper to a decent height on their ends.
My opinion is an 8" deep torsion box for this size (L x W) table is
greater than required. Four inches face-to-face should be more
than enough for the loads imposed by the Onefinity and any work
piece placed on the machine.
Almost all of that load is transferred to the table top (torsion box) at
the corners where the feet are. The load is transferred directly down
to the floor by the table legs or whatever your support structure is
under the torsion box. You have very little if any cantilevered load so
minimal load on the skin of the box once it is setup and immobile.
The box only serves to keep the four corners in the same plane.
It will work harder if the surface on which the table rests is uneven as
most floors are and it has not been properly leveled. And having your
table on wheels implies you plan or have need to move it on occasion,
and this is when the torsion box will be asked to step up and keep that
top surface in a single plane and not let it warp or contort. To me ANY
movement of the table would necessitate checking the spindle travel
is still coplanar.
I choose to set my machine up without wheels and placed it so that
hopefully I never have to move it once the table is leveled and dialed
in but I can still run a full 4x8 sheet through if I should need to.
Yes, I know I used the word level or leveling and that has a different
meaning than coplanar. But I believe level is easier to obtain for people
with the tools most have access to. You need to achieve both, at least
that is how I have set up mine.
Having the table level makes all other measurements that much easier.
And maybe it keeps your bit from rolling off the table and crashing to
the concrete floor.
My plan I have in mind differs from this concept. I don’t have legs at the corners. The leverage forces in my plan are higher because they are caught in a middle portion that has only half the width and length of the horizontal CNC machine dimensions, therefore the double vertical dimension of the upper part. But I will use even less material and at same time achieve high torsion resistance.
THIS is exactly what I want to eliminate with my concept. My table feet system is not connected to the corners of the table top by legs.
Sorry if I present a concept here but have no drawing prepared, it will take a little time before I implement my idea but I have no problem to share it then.
Leveled coplanarity is my favorite coplanarity
If you precisely level two lines you have ensured their coplanarity at same time
Thanks to you and Tom for mentioning the fishing line method to test if the Y axis rails are coplanar. I first saw it a long time ago in an article about flattening workbench tops with a router sled. It is a simple and elegant solution. In my case, I realized the rails are magnetic and used some small bar magnets to pin each end of the fishing line(s).
The Journeyman arrives today (4 weeks earlier than I expected). So my torsion box isn’t built yet. But I made one as an outfeed table for my saw, following an 8-part series: SawStop Outfeed Table Pt 1 - YouTube My plan is to adjust the dimensions to make another one for the CNC.
This one rides on casters, but the corners are adjustable. I plan to keep the CNC in one place, but if/when I get to tiling, I will have to move it.
My shop floor is also uneven, and I notice that the table legs follow the floor (losing the planar upper ends at the legs). But the table stays planar. A wobble tells me I have to adjust the support under its corners.
By the way, the video is by a fellow from Georgia (USA). If you’re a northerner (like me) increase your You-Tube speed to 1.5. You won’t miss anything, and you’ll stay awake.
