Top Beam attachment to frame

re: Top Beam attachment to the frame.

Splitting off of the 800+ post thread

How should the top beam be attached to the frame

For the standard frame, the top beam will not be adjustable.

The version 1 approach of attaching the motors to a piece of plywood and attaching that to the frame is something we want to get away from, the plywood flexes and the attachment to the frame has proven to be less than rigid (even with a lot of screws driven into the joint). And finally having the motors stick out gives them leverage to flex the frame.

We’ve talked about having several layers of 2x4 spacing the motors out from the top beam, but that has alignment problems and still gives the motors the ability to flex the frame

As a result, we are wanting to attach the motors directly to the top beam roughly as shown in the pictures of Bar’s work, for example

For the standard frame, the forward face of this top beam should be about 6" forward of the face of the leg.

We have looked at multiple ways to do this, none of them is obviously right, and all of them have problems

One approach is what is pictured in the picture above, with short lengths of 2x4 attached to the legs and the top beam attached to these supports.

  • The difficulty with this approach is how to attach the top beam to these supports.
    We’ve discussed many opttions
  1. (current thoughts), lay the board flat so it’s easy to fasten (there is a little movement as a result)
  2. have the board upright, with a block behind it (how do we keep it from being tilted as it’s fastened, the front of the beam needs to be parallel to the legs/chains/etc)
  3. have the board upright and drill through the long dimension of the 2x4 to fasten it to the supports (how do you drill a good straight hole)
  4. variation of the last, drill through the support and have the bolt/screw from below (similar problem but only one end of the hole needs to be in the right place)
  5. use a heavy duty angle bracket and piece of unistrut (trivially modifyable to be adjustable) like this, but attached to a wood leg and top beam
    P1080915 (cost of the brackets, more stuff to ship, people in metric countries would need imperial tools to attach)
  6. drill through the thick side of the leg and add spacers, as per the first picture in this post (how do you make all the holes straight and keep the beam level, it does make it easy to have the face of the beam parallel)
  7. use threaded rod to space the top beam out from the frame,adjustable4

What other ways can people think of to attach the top beam to the frame?

@Bee is talking about having a horizontal 2x4 at the top of the frame and the top beam fastened directly to that. That makes the top beam very rigid in both directions, but doesn’t get the motors in the right place.

with our current approach of laying the board flat, we have more flex than we would like, but attaching a board or piece of plywood to the back of the top beam would make it significantly more rigid.


My take on this. I was looking for bolts but it hit me you can buy all thread and use nuts and washers as well.

It uses 4 1/2 threaded parts (bolts or all thread) to make 2 pushers to adjust outward and 2 pullers to join the front and back.

Thank you

As it’s fastened or as the glue cures?

Toenail a screw into it or add a very inexpensive and readily available mending plate.

I know it would add additional weight and cost but would using a 4x4 work for the top rail. It would be extremely ridgid.

I personally am leaning towards the unistrut solution for my build.

a length of LVL would also work well. as the beam.

For my machine, I’m going to use the unistrut angle to hold the top beam (I’m trying to find out if there are enough people in southern california who want to do this that we can order a box of 20 without having to ship them)

I didn’t think about lvl. Aren’t lvl more stable then dimensional lumber.

yes they are, and a good bit stronger, also available (via special order) from big box stores.

but right now we think that they are overkill, and so are looking at just using standard stick lumber

Yup… I allways tend to go overboard on things like this.

1 Like

if it’s worth doing, it’s worth overdoing :smiley:

I have to fight against this all the time.


Last night I ran a test to see how much the beam moves as a result of a possible “expansion” of the glue as it dries. I used gorilla glue and made the top beam attachment to an arm using the block as described in another post. I used two screws to secure the top beam to the block.

To test the expansion effect, I intended to measure the distance between the top of the beam and the bottom of the arm after I glued and then 24-hours later after it set. However, because my calipers are limited to a 6-inch range, I couldn’t measure from the top of the beam to the bottom of the arm (2x3.5=7 inches). So I marked off a spot near the middle of the top beam and measured the distance to the bottom of the arm.

When I made the joint last night, the distance as read by the calipers was 5.265 inches. Just now, I went and reread the measurement (approximately 20-hours later) and the measurement read 5.254 inches. This is a smaller distance than the original so there wasn’t likely any expansion of the glue that would lift the front of the beam. It could just be measurement error (i.e., can I put a caliper on a point within 11/1000ths of an inch precision?).

Based upon the test results, in my opinion, there’s no issue with the block technique. Also, the glue makes the joint immovable as well.

1 Like