Flexible Printed Circuit Geometry

[DanBovinich]

I have a geometry question I can't solve and I can't google. I don't even know what to google or what to search in this forum.

I am making a flexible printed circuit (FPC) to connect current collectors in a battery pack. This FPC has traces on it that go back to a circuit board that senses the voltage of lithium batteries.

We are riveting this circuit to the current collectors. We want the FPC to be longer that a straight line to the rivet holes so that the FPC will bend upwards. This will form a "wave" or "bow" "arch". We want the bow to be a certain height and width. The bow allows tools to go into the module. If the FPC was dead straight the tools wouldn't be able to go in and tighten some screws with out damaging the FPC.

So, Imaging taking a piece of paper, put your two fingers on the paper starting with your fingers parallel and wider than 30mm apart. Now slide your fingers closer together, the paper will bow upwards. Keep moving together, so that an arch/bow/wave forms that is exactly 5mm high. Your finger edge should be 30mm apart. See attached pic.

I am on SW 2021.
The sample part is uploaded.
This part MUST be formed in sheet-metal feature, it MUST flatten. Otherwise it would be hard to manufacture. we need the flattened pattern.

The question is, how, in SolidWorks, or in geometry, would the shape of the bow be formed? Yes, I can just guess, but that's not why I'm here.

Problem, I tried forming a spline, but sheet-metal will not work (at least spline don't work in SW2021).

If a spline can be formed, I could make it a construction line and then fit radii into it to make the sheet-metal feature work. But I don't know how to know what kind of spline would form?

I have some dimensions that must be kept. See attached drawing. The hatched area on the FPC will have stiffeners in it, so consider them as "fixed geometry". The 30mm & 5mm dims are the important dims.

Also, there will be other arches on this FPC that may have different heights and widths. So, I need to be able to "scale" the geometry.

Thank you for any help you may give me....

[SPerman]

I'm not a sheet metal guy, but my assumption is that the shape is determined by several factors that you will only determine through real world testing. It could probably be calculated in simulation, but it would require you to know the materials properties.

[DanBovinich]

That's a thought, I wonder if the motion manager or simulation could make a model? Not sure. The material is some kind of plastic.

Thank you,

Dan

[zwei]

The question is, how, in SolidWorks, or in geometry, would the shape of the bow be formed? Yes, I can just guess, but that's not why I'm here.

I am a bit rusty on my structural mechanics but maybe a buckling analysis can give you some idea on the deformation/bowing?

[Frank_Oostendorp]

I can share, out of experience, that a FPC that needs to arch within certain tolerances, needs some guide to do this in a reliable way. Either a carrier of plastic inside the arch, or a limit of space on the top of the arch, including a layer of insulation, if needed. If all traces are parallel running across the arch, it might form more predictable, but if damage is done when the arch is not exactly 5mm....

[TTevolve]

Do it as a sheet metal base flange. Have to use arcs/lines and use fillets to connect them. It unfolds, probably will need to play with the K factor to get the correct size you want in the flat.

Here is the sketch I used

Unfolded

[DanBovinich]

That's the way I did it. The question is, what dimensions and shapes to use? The dimensions you used were just a guess? Right?

[gupta9665]

Have you tried using arc length dimension to fix the arc length?

I'm a gif, click on me :P

[Frederick_Law]

Shape doesn't really matter.
Flat length is important. Whatever shape it is, flat length stay the same.
Set k to 0.5, sketch the profile so total length: arc and straight equal to flat length.
Extrude model midplane. Or contour flange midplane.

[DanBovinich]

Have you tried using arc length dimension to fix the arc length?

FCB.gif

Deepak, is that you? My old buddy?

That works, but the sheet metal feature will add radius's in the corners. You have to tell it what size R. So, its back to guessing again....

[josh]

Will the flexible circuit be completely free in the air other than being clamped on both ends? And it’s all assumed to be elastic deformation, right? You’re not pre-forming it with creased bends? And you’re showing it the same width all the way across too. All the shapes I’m seeing so far have small radius bends upward and then a large radius arch downward. Assuming it is symmetric through the thickness, I’m pretty sure there is no reason to think that the upward bends will have a different radius of curvature than the downward bends.

[Frederick_Law]

With k=0.5, flat length is: 2*(d6+d7)+d5

Set all radius to be the same:

[gupta9665]

Deepak, is that you? My old buddy?

Yes my old buddy, miss the days we had worked together

That works, but the sheet metal feature will add radius's in the corners. You have to tell it what size R. So, its back to guessing again....

You can keep a small bend radius if that helps.

[DanBovinich]

With k=0.5, flat length is: 2*(d6+d7)+d5
FlexPCB-01.jpg

Set all radius to be the same:
FlexPCB-02.jpg

Hi Frederick,

I ended up making it very similar to what you did, except that I don't have any flat lines. I made them all radius' with equal & tangent relations.

[Frederick_Law]

Flat hole might be easier to assemble.
Could setup equations to drive radius.