Assembly using turnbuckles and wire rope

[theycallmevirgo]

The attached assembly represents a piece of furniture intended to be assembled with wire rope tensioned by turnbuckles. My limited understanding of assembly constraints and joints is insufficient to the following tasks; 

1) If necessary, align the turnbuckle (top left) to lie on the work axis (currently being used to position the wire rope, center)

2) Insert the hook of the turnbuckle into the loop of the wire rope

3) make length of wire rope a parameter equal to (distance between bolts-(2*length of turnbuckle)).

PS - I can't give a specific example right now, but it seemed as if some assembly constraints resulted in moving parts of the frame around. I can't seem to see any degrees of freedom on the frame at all - how could that be?

Many thanks in advance

Joe

[3D4Play]

I work with wire rope and turnbuckles a lot, so have a few generic thoughts and suggestions for you:

1. the frame isn't fully constrained, so there are certain parts that will move around when they are "flexed". I don't think you intend to constrain them by using the turnbuckle and cable.
2. In the turnbuckle assembly, a few parts are adaptive. I think this is because you want to set the turnbuckle's length based on adaptivity, once it's constrained to the cable eyes. If my guess is correct, then I suggest doing this as a flexible component.
3. The YZ origin plane isn't centered on the hook's axis. This will cause you a little problem when constraining, unless you have intentionally offset for some reason. You should constrain the turnbuckle with the eye/hook axes constrained in the same plane (YZ in this case), their respective swept profiles constrained in the plane 90-degress from the YZ (XY in this case).
4. Nice job on the wire rope modeling, but honestly you are just adding bloat - and likely even taking a slight performance hit - by modeling it realistically in this way. I suggest changing the rope to a single, untwisted swept circular profile.
5. The key to aligning turnbuckles and cable assemblies to constrain easily, is ensuring that the turnbuckle eye axes and the wire rope assembly eye axes are in the same planes, respectively, so that they can be mated together (per item 3 above). Note that in your model the cable turnback eyes are offset from the XZ plane because they are turned in opposite directions. Flip the direction on one of them and they eyes will then be aligned as I'm suggesting, but aligning them to the XZ plane won't align with the cable centerline axis, and therefore must be individually constrained.
6. I disagree with your desired cable length, which really wants to be two equal lengths, more or less equivalent to the distance between pins/2, minus the desired FIXED turnbuckle length, hook-to-eye. This assumes two equal cable lengths with a fixed-length turnbuckle in between them. For realism, remember that the length of the turnbuckle and the cables is not axis-to-axis, it's point of tangency to point of tangency. Sooner or later you'll probably want to compensate for this.

Hopefully, you find these suggestions helpful.

[3D4Play]

My apologies, I misinterpreted your intent 😉 The original would apply if you wanted two equal cables with a turnbuckle in between. However, I suspect you only need a single cable and a single turnbuckle. So, item 6 should have read like this:

I disagree with your desired cable length, which really wants to be two equal lengths, more or less equivalent to the distance between pins/2, bolts, minus the desired FIXED turnbuckle length, hook-to-eye, of a single turnbuckle. This assumes two equal cable lengths a single cable with a fixed-length turnbuckle  at one end in between them. For realism, remember that the length of the turnbuckle and the cables is not axis-to-axis, it's point of tangency to point of tangency. Sooner or later you'll probably want to compensate for this.

[theycallmevirgo]

I'm not 100% clear on what you mean by (3). Do you mean, a flush between the eyebolt yz plane and the turnbuckle assembly yz plane, as well as a flush between hook yz and assembly yz?

I'm trying your other suggestions right now.

Thanks again

Joe

[theycallmevirgo]

Another weirdness; if I try to create a flush constraint between the turnbuckle part xz plane and the eyebolt part xz plane, it never prompts me if the constraint already exists. What's the point of making more than one flush between the same 2 planes?

[3D4Play · Accepted solution]

Yes. Actually, I think you want the turnbuckle body, the hook axis and the eye axis all to be in the same plane. There's a lot going on with what ostensibly seems like a wire rope, its turnback/eye (thimbled or not), and the three components of a turnbuckle - in your case, the body, the hook, and the eye. An easy way to constrain them all is to simply align the turnbuckle assembly and the wire rope along a single axis, but in reality you'll probably want to display them in a view that shows everything nicely aligned in a normal plane. Play with the attached parts and see if there's any additional insight.

[johnsonshiue]

Hi! Inventor supports partially constrained and redundant constrained workflows. If you want to avoid redundant constraints, you can simply to to Tools -> App Options -> Assembly -> check "Enable redundant relationship analysis." Then any redundant constraint will be marked with an exclamation sign.

Many thanks!

[theycallmevirgo]

Thanks so much for this. I ended up using your assemblies with two modifications. First, I projected a circle into the middle of the hook curve to give the rotation joint something to grab on to. Second, I constrained yz plane of eyebolt to xy plane of hook. This enabled rotation of the rope in a plane parallel to the walls of the assembly, which will hopefully allow me to dimension rope length dynamically. 

Thanks again

Joe