4
$\begingroup$

I've been playing around with 3D printing some compliant mechanisms, mostly to get precise linear motion, and I've noticed that many commercial compliant mechanisms are often designed like this: XY Flexture

Where the flexible parts are only thin by their joints. Is there any benefit to this over simply keeping it thin across the whole flexible part? All my designs use flexible parts which are thin all the way, granted I'm mostly working in PLA so I need the parts to be long and thin to handle the stresses without breaking, but what is the reason behind keeping those parts thick in the middle?

$\endgroup$
2
  • $\begingroup$ One thing they would do is prevent crumpling from torsion. Are they pinned to the body to resist torsion from the retainers and tubes? $\endgroup$ – Phil Sweet Mar 4 at 16:55
  • $\begingroup$ @PhilSweet beyond that, the pictured item is for precision optical alignment, so if torsion tilts the bore with respect to the mounting rails, it's not very useful - and the tube screwed into the bore may be a bit of a cantilever $\endgroup$ – Chris H Mar 5 at 10:25
4
$\begingroup$

In this case its because they are emulating a bar mechanisms. This makes it possible for the designer to think of the problem in terms of revolute joints which allows them to use age old design principles with parallelograms and all.

enter image description here

Image 1: Equivalent bar mechanism

In this case the bar is rigid so the equivalent compilant needs to be rigid too. Doing it thisway makes the thing much much easier to design.

$\endgroup$
2
  • 1
    $\begingroup$ Ah, This makes more sense than the other answers. I thought it would be equivalent to just have a thin part, but with the bar example it becomes obvious that a flexible bar would allow rotation of the central part which is not allowed with a stiff bar :) $\endgroup$ – Beacon of Wierd Mar 4 at 16:04
  • $\begingroup$ Yes, the center does not rotate just translate. The benefit of compliant joints here is that they dont exhibit stick slip. Which allows you to be very precise indeed. $\endgroup$ – joojaa Mar 5 at 0:14
3
$\begingroup$

I'll start from the particular example you are asking about. It is a positional adjustment device. It is heavily used in laser optical tables to provide micrometer adjustments. And it falls in the category of compliant mechanisms like you suggest.

There are a few requirements for these devices (most have been already pointed by others). IMHO, the main two are

a) resistance to vibration (the added mass serves this purpose, because it lowers the natural frequency).

b) spring back repeatedly to original position. This is a very important aspect for these types of devices. They do not rely on the screw to set the position, they passively press onto the screw, thus minimizing any backlash that might be present.

Now regarding why compliant mechanism need to have thick sections:

In my understanding is that for the same force levels, the deformation of thin sections provides the range of movement, while the deformation of thick parts is negligible, thus allowing to maintain a functional shape.

If the thickness was everywhere small/flimsy, you would have a structure that would deform everywhere significantly. That may not always be a problem. The problems occur when you are trying to move/displace this structures a lot. Then their mode of displacement becomes unpredictable. (you might have experienced the problem when you are trying to use a tape measure in non straight position like the following)

enter image description here

$\endgroup$
3
  • $\begingroup$ Having done a lot of optical work including designing flexure mounts, I'd hope there isn't too much vibration around, but if there is, this will be better than some designs. The springing is very important. We're using an aluminium (6061-T6) prototype, because we can just get away with the elastic range that offers. We'd always thought we'd have to have the final one made in steel. But it mustn't be too stiff a spring as tiny adjustments need to be made in awkward positions $\endgroup$ – Chris H Mar 5 at 10:30
  • 1
    $\begingroup$ Stiffness is a function of the cross-section and/or the mass moment of area, so you can fine tune it within some limits. I've been around a few of these components and I'm always impressed by their workmanship.... even the cheap ones. Its not an easy undertaking so my hat off to those that can pull it off. $\endgroup$ – NMech Mar 5 at 11:05
  • $\begingroup$ I'm lucky to have access to a very good workshop. Our design was rather different ( a tripod arrangement so conceptually simpler) because of access and the need for the thermal expansion to be as symmetrical as possible (between about 40K and above room temp). It was quite a fun/challenging project $\endgroup$ – Chris H Mar 5 at 14:48
2
$\begingroup$

This looks like a wire edm part for those fine interior features. My guess is the design on the thin cuts was optimized for that process - i.e. constant width cut except at the entry holes, which are drilled.

Where the "thicker" hexagon-ish zone is, it is essentially not flexible. Essentially all the flexibility comes from where the sections adjacent to those which are "thin". The resulting stiffness of the mechanism is controlled by the length of those thin sections.

The question of why "thin-thick-thin", vs "medium" the whole way on each side, may be because they needed some thickness for the adjustment screw to tap into. The opposite side would then become the same to preserve symmetry. There may be a vibrational consideration too as Solar Mike suggests.

I have a feeling whoever designed this had some aesthetic objectives too. (I.e. looks cool)

$\endgroup$
3
  • $\begingroup$ "Looks cool", may be more likely to avoid natural frequency due to vibrations... $\endgroup$ – Solar Mike Mar 4 at 14:03
  • $\begingroup$ maybe, but the screw adjustments seem like they would constrain it. I.e. I think this might not be a "suspension" part, but a position adjustment for whatever goes in the center axis (lens?) ... Not sure tho $\endgroup$ – Pete W Mar 4 at 14:05
  • $\begingroup$ @PeteW you're right. I recognised it as from Thorlabs who make optics and opto-mechanics $\endgroup$ – Chris H Mar 5 at 10:26
1
$\begingroup$
  1. For traditional parts you still need some thickness for subtractive manufacturing processes to prevent the piece from deflecting too much during material removal to get repeatable parts.
  2. The time it takes to remove additional material costs money, if less material can be removed the part can be made for cheaper.
  3. Usually you still want some rigidity at mounting surfaces or you lose the precision of the location of your initial joint.
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.