8020, how to ensure squareness at joints?

[averageJoe576]

Hey guys,

Building a device which may become a Router in future, but just a modular setup table for now, and wondering about some general 8020 tips. The tabletop perimeter and some intermediate beams will be 1530, 3" face vertical.

What tips/hardware do you suggest for getting good squareness out of 90deg joints? (3" face inside)

What can I expect if I just do a combo of inside connectors like #4304 (or gusseted like #4438) plus right angle flat plate like #4481 on bottom?

Permanent bottom plates necessary? Or just clamp something flat while bolting inside corners?

Any advantage to something like ground CI angle plates and some similarly flat material for bottom?

I have a Taig CNC mill and a horizontal bandsaw, fwiw.

Similar Threads:

• How do I ensure a flat torsion box?
• use q setter to ensure drill is broken
• How to ensure supported rails are even?
• Just IN- Ensure Radial Integrity and Maintenance-Free Operation with Square Bearings
• What should I do to ensure the glass tubes not to break?

[wizard]

It all starts with cutting the materials square! That requires a precisely set up saw of some type. Ive used a miter saw set up with a precision machinist square. I believe 8020 at one time used a table saw with a orecision cross cut sled.

In any event a square structure starts with cuts that are square on the ends. It is the critical first step.

As for things like gusset plates and tie plates they do nothing for squareness. Most of the plates are nothing more than bar stock or sheet goods with some holes drilled in them and as such do not index anything for alignment. Angle brackets can achieve some level of squareness but one alone they are not structurally sound.

Im on my cell at the moment so those part numbers mean nothing to me. However what stands out here is the machines size. You didnt mention that because 1530 may or maynot be suitable for a router or even a setup table. By the way setup table isnt defined either. So we can only talk about T Slot contruction in general terms. The first thing here is that every joint benefits from reinforcement usually in the form of angle brakets or gusset plates. The gusset plates can be supplied by the manufacture or you can make your own. Angle brackets are a bit harder to do on your own but not impossible.

One thing you might not be aware of is that you will be nickle and dimed to death for much of these ""accessories"". The T Slot nuts, bolts, brackets and so forth add up to alot of money if you do a solid build. Your mill and bandsaw has the potential to save you some money here if you want to spend the time doing brackets. In any event the cost of T Slits can be high compared to doing the same structure in steel (subject to change). As such i almost always suggest that people look at steel tubing for a machine frame. Steel might not work for a setup table though if those T Slots will be used for setting up whatever.

So we need more info here especially about this table size and what is being setup on it.

[averageJoe576]

The 1530 is already bought and paid for (got it at %50 discount) so accessing its viability is moot at this point, lets just go on the assumption that its stiffer and straighter than what I was planning on using, and that I want to get the best out of what I got.

As I mentioned, I was hoping to keep my t-slots free, so I was thinking something like this:

view from below

view from above

In this setup, the dimensions of the angle/flat plates would control the squareness, and the cut surface wouldn't matter, unless there's something I missing.

[wizard]

How well those joints will work really depends upon what you expect to do with the unit as an assembly table. As a router I'd be very reluctant to go with the joints you are showing above.

Personally if access to the T-Slots is that important I'd come up with a stronger weldment for each corner. Use plate steel heavy enough allow leg mounting if this is to be a floor standing table. Basically if you want the table to resist twisting and remain square you will need fairly beefy corner brackets. The brackets also need to be large enough to get plenty of grip on the extrusions T-Slots.

[skrubol]

What are those braces, about 3mm thick? As shown, assuming good enough attachment to the extrusions the stiffness will be about that of a somewhat damp noodle. Without a top plate to pair up with the bottom plate, the bottom plate isn't of much use, and without gussets the inner bracket isn't going to do much.
Or is that 1530 inch? (confusing nomenclature, 1530 could be 15x30mm or 1.5x3.0") Either way, brackets so thin in relation to the extrusion won't be very stiff.
A corner bracket like this (for 1530 inch) will be much better.
https://8020.net/shop/4338.html

Not sure how square it will be though.

[averageJoe576]

1530 is a part number used by 8020, they are 1.5" x 3.0". All the parts I linked are for 3" extrusions.

I mentioned in my OP I was trying to decide between several brackets, including the #4338. As an alternative, to get higher precision, I mentioned getting ground flat cast iron angle plates.

The ones in my drawing are just to show the general idea, not actual dimensions.

[skrubol]

Sorry, I didn't go back to the OP when responding to the post with the models.

If you've got places where you can put 45 degree braces in, that will add a ton of strength, but also give you the ability to tweak the squareness.

[averageJoe576]

Ah, that sounds like a good approach. Focus less on enforcing top-view squareness at the corner and fashion some of the 1530 (or other members) into an long (12"?) adjustable 45deg brace and dial the squareness in and adjust at the cornet for top-surface flatness.

Any suggestions on how to gauge squareness/flatness on a that large a scale? Don't currently have access to any large flat or square surface.

I guess, as wizard mentioned, the standard approach would be to just have cross members cut precisely and butted flush. My bandsaw won't deliver on that, but I guess I could stack them in my mill and face the edges. I plan for my non-perimeter cross-pieces to be modular, so I don't really want them tight fit, but I guess I can fashion some "fitment" members and use them temporarily for setup.

[skrubol]

Rough squareness is relatively easy, make sure opposite sides are the same length, then make sure the distances from corner to corner are the same. Depending on how accurately you can measure you can get it pretty square that way.
Flat.. I don't know, usually comes down to a reference surface or straight edge. Machinist levels can get you level (and if everything is level, it should be flat,) within well under a degree, but I've never used one. I'm sure someone with better skills than mine can chime in.

[averageJoe576]

Yeah, I considered buying a 6" machinist square and clamping it on top as I bolt the corners down, but I do link the idea of a precision level much better, would have several other uses for it.

[wizard]

Unless i missed it we still haven't gotten this tables size and use from you. This is important as it impacts how we would address some of these issues.

You call this a setup table but all we appear to be discussing is the outside frame. We should also be concerned about any T-Slot members filling the space between the outside rails. Depending upon machine size you will need something, especially with light 1530 extrusions. We dont even know if this is a bench top or floor standing machine.

Ah, that sounds like a good approach. Focus less on enforcing top-view squareness at the corner and fashion some of the 1530 (or other members) into an long (12"?) adjustable 45deg brace and dial the squareness in and adjust at the cornet for top-surface flatness.

Building up something like this without a flat surface to start from is a challenge. Again size is an issue but on a significantly large machine you may need a temporary platform or frame to build the machine up on. Smaller machines might be doable on the kitchen table.

Getting things flat and Square is why some resort to welding up a frame, having it normalized and then having the critical surfaces machined. Epoxy leveling is another approach people have used. If you are full of energy hand scrapping the frame flat is also a possibility.

Precision levels are also a possibility and the use of T slits can give you some adjustability. I say some because it depends upon the assembly techniques and hardware used to bolt everything together.

Any suggestions on how to gauge squareness/flatness on a that large a scale? Don't currently have access to any large flat or square surface.

Precision measurement and some trig can help. The simple no trig method of measuring diagonally has already been mentioned. To be effective thought you really need precision scales. Dont dismiss trig though because it can help with determining error.

I guess, as wizard mentioned, the standard approach would be to just have cross members cut precisely and butted flush. My bandsaw won't deliver on that, but I guess I could stack them in my mill and face the edges. I plan for my non-perimeter cross-pieces to be modular, so I don't really want them tight fit, but I guess I can fashion some "fitment" members and use them temporarily for setup.

The above leads me to believe that you would be better of buildin this fixture table and then building a separate router machine. The reason is this you dont want a router with loose fitting modular pieces. Especially when witking with small cross section extrusions like 1530 where your internediate cross members can offer a lot of support to the outside rails. On the otherhand i can see where adjustability in a fixture table might be very desirable. Id go so far as to suggest that special attention be paid to how you make those interior rails adjustable.

A few more thing to consider. We font know the size of this frame but understand this it will flex and twist under its own weight even if you do somthing better with the corner joints. Provisions for jack screws should be added if this table will sit on a bench top. If the final goal is a free standing bench then the frame underneath is just as important as the table frame. It comes back to the question of how will this frame be used.

[wizard]

Yeah, I considered buying a 6" machinist square and clamping it on top as I bolt the corners down, but I do link the idea of a precision level much better, would have several other uses for it.

Precision levels are extremely handy to have! However when it comes time to build a CNC you will still need a precision square. Also depending upon you routers specific design, absolute squareness of the X axis frame "MAY NOT" be as important as having the linear rails parallel and in plane.

[rowbare]

Buy 4 angle plates of an appropriate size, bolt them in and leave them there.

bob