Machine Frame Assembly

[Jieve]

Hello all,

Question: I have a machine frame made out of aluminum profile rails (from a company called Kanya) as shown in the link below. Although we were able to completely assemble it, we've had some issues with the straightness, in that the frame does not sit flat on the ground. Unfortunately, the floor on which we assembled it was not exceptionally straight. A plate (12mm) is going to be screwed down to the top (like a table), as well as to the lower ledge, so it needs to be as straight as possible (not sure what the length tolerances are on the cuts).

Machine Frame

My question for you guys is: In what order should I assemble and torque down the frame pieces in order to keep it as straight as possible? The plates are to be screwed down using countersunk phillips screws. Would it be best to first screw down the plates, then add the legs afterwards? And add the angled bracing pieces last, after everything else is torqued down, or what do you think? The holes in the plates were going to be drilled after the frame is fully assembled, but now I'm starting to wonder if maybe the plates should be drilled and installed first, in order to ensure that they are actually flat and to help alignment.

Any opinions about this would be greatly appreciated.

Thanks!

 

[MintJulep]

Shim under the legs.

 

[MikeHalloran]

Assemble it upside down, on the table top, unless you have a stiffer, flatter surface on which to build.

That implies predrilling the tabletop, of course.



Mike Halloran
Pembroke Pines, FL, USA

 

[tygerdawg]

Your link to image doesn't work for me on two different computers. I'll assume it looks like any other aluminum extrusion frame.

You say "as straight as possible" but don't give any tolerances. Some things to consider:
(1) Use levelling footpads and a bubble level to make the 12mm plate & final assembly "straight", relative to the floor.
(2) Because AL extrusions are bolted connections, you are likely to impart twist to any elements, thus giving non-straight mating surfaces. If "straightness" is really important to you, then provide assembly jigs or other fixturing to aid the assembly process. It may be as simple as solidly clamping the mating elements between two plates with a C-clamp, then making the bolted connection.
(3) Typically AL extrusion framing elements have a couple degrees of distortion built into the design to provide "lock-washer" function for the bolted connection. You must be prepared for that, and how it affects subsequent connections and the final assembly.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[hydromarine]

Use chockfast on the base will take out any flatness issues

 

[Jieve]

Thanks for the responses.

For some reason the link worked last night, and now it's removed the colon from the address. To see the image, open the link in a new tab/window, and add a : after the https, refresh and it should be viewable.

I have remeasured the assembled but not yet completely torqued down frame at different locations, and it is actually much more accurate than I thought, I think the largest deviation is somewhere around 0.5mm. My concern was not necessarily that the plates are level with the ground, since this is easily fixed with shims. My concern was the fact that the countersunk holes in the plate don't allow for much sideways play during assembly, and that the frame was originally rocking on the floor (distorted). The slot nuts (whatever you want to call them) into which the screws are threaded can move in one direction (say X) but are fixed in the Y direction. The screws screwing down the plate are essentially double fixed fasteners so the holes in the plate need to be accurately located. If the plate is screwed down first, then the frame fasteners tightened, this could result in a bending load on the screws if they are not completely accurately located. A colleague recommended just drilling the holes and countersinks slightly larger to accomodate, if necessary in certain locations.

Initally after assembling it and placing it on the floor, the frame rocked back and forth. I moved it to another section of the floor, and it was fine. Since our floor is apparently really uneven, I'm thinking the suggestion of screwing down the plate first with the holes pre-drilled is the probably best solution.

Does this sound reasonable?

 

[Jboggs]

My first thought would be to contact the frame system manufacturer and ask for their help. Have you done that? What did they say? They are the ones with a personal interest in making sure their product is useful for their customers. I have my own thoughts on how to do it, but I suspect you would not be the first of their customers to have this issue. I would give their recommendations more consideration than my own because they have encountered the question before.

 

[Jieve]

Hello,

The frame is actually my own design for a machine prototype. The assembly questions started coming up after we received the parts. I wouldn't really call them problems or difficulties at this point, however I was curious as to how the more experienced in this board would go about assembling this in case I am overlooking something. So far the answers have been pointing me in the right direction.

Thanks.

 

[hydromarine]

Also not able to open your attachment, sounds like you need location dowels fitted to your frame prior to bolting down the plate .This would solve any misalignment , but you would of course need accurate alignment prior to drilling the frame for the dowels.

 

[Jieve]

To open the Link, right click On it, select open in another tab/window, then when you get the error page, add a colon : after the https in the address bar. For some reason it was deleted when i posted the Link. That should work.

 

[mfgenggear]

Jieve

late in the game here. can't open file to see your assy.
I have assemble projects in the past.

The following sequence may help.
1#) to achieve the final assy straightness the details must be machined to achieve final assy stack up tolerance.
as suggested before precision dowel pins will maintain the alignment & straightness.
a precision plate will assist with maintaining alignment, squareness & straightness.
The machining must very close true precision.
2#) loosely assemble (if possible) if weight allows, use clamps to hold it in place while screwing in place. temporary fasteners also help, something similar to clicos, which allows for the actual fasteners to be installed. then torque after final assembly. snug each screw in sequence,
use a precision level .0005 or better. to make sure it level & straight. a precision square will help.
if it is small enough for CMM, if not a jig optical transit will help .001"
do not torque until all screws are snug. then after all the assy completed then do the final torque in a rotating sequence.
the material of the screws & parent material will set the require torque values.
the precision dowels will maintain the required precision.
but with out viewing the actual size & weight of the assy this is just swag.
" can you try to post the frame assembly again"

Mfgenggear

 

[Jieve]

Ok here is the frame out of 6063-T66 extruded aluminum profiling. A 12mm thick plate (not shown) screws with countersunk screws onto the top. There is also another 12mm plate (not shown) the same size except with corners cut out which screws with countersunk screws onto the top of the lower cross rails. Profiles are 40mm x 40mm. Overall size is L 884mm x W 636mm x H 640mm.

Overall length tolerances of cut profiles given by the manufacturer of the profiles are based on ISO 2768-m.

Thanks so much for the help and responses.

 

[mfgenggear]

Jieve

What I mention earlier should be close to what needs done.
except for production tools can be designed & developed to mount on &
expedite assembly.

Hope that helps

Mfgenggear