Unbending a section of a cone 1

[Guest0527211403]

Hi folks,

I am trying to unbend a section of a cone in Pro/Sheetmetal.

To give you a visual idea of what I am trying to do, imagine a cone which has had two angled, planar cuts through it. The result is a section of the cone bound by two ellipses. A series of these sections are going to be welded together to create an expansion chamber for a two-stroke engine. The problem area is in the diffuser section, where it is making a full 180 and increasing in diameter at the same time.

The technique I am using involves designing a theoretically perfect expansion chamber using a variable section sweep, slicing it up into curves and then creating boundary surfaces between these curves which I plan to use to drive the sheetmetal parts. I then want to nest all of these parts on a single sheet and send a DXF to a laser cut shop.

When I bring the surfaces into sheetmetal, I create my first wall as an offset wall from the imported quilt. I then add a rip down the inside (where the seam will be). When I try to flatten this though, Pro/E tells me that there is "No Geometry to Unbend". argh.

I'm certain that the surfaces I'm bringing in are conical, as they only have curvature in one direction and they definitely aren't cylindrical.

When I make a cone on its own in Pro/Sheetmetal, I can rip it and unbend it without a hitch. But I don't want to use this bottom up technique on this, since a design change would be a nightmare, and at this stage (I'm restoring this scooter) I'm certain that I'm going to have to make some changes as I go along.

Does anyone have any ideas as to how I can unbend these parts? Or, if anyone has another design approach, I'd love to hear your ideas. So far my brother (who I'm building it with) wants to try and hydroform it by welding two half sections together, sealing it on both ends with a valve and a nipple, and pumping it with pressurized water, while banging out the imperfections with a hammer. I've seen it done before, but I'm worried that the end product won't look anything like what we want. It also just seems inherently dangerous.. but this is the same guy who convinced me to ride my bike off of the roof of our garage when we were kids.

Thanks in advance

 

[Heckler]

I haven't used the sheet metal module in Pro/E so I can't comment on its functionality. But the project is of interest. I would have to side with your brother's technique of hydroforming then weld the two half sections. This is the way mass produced chambers are made. Get your bike ready for another trip off of the roof . How bad were you injured? I did something similar with a sheet made into a parachute.

 

[Guest0527211403]

I came out of that with only a sprained ankle and an new-found distrust for my brother's bright ideas. But that's what happens when you're the younger brother. Mind you, that wasn't the last stupid thing I've done on his suggestion. Aren't siblings great?

Here's a link showing what he wants to do:

http://www.two-stroke-addicts.freeserve.co.uk/mick/hydropipe.htm

The chamber is for an old 90cc Vespa. Luckily we have a lot more work to do before we start fitting an exhaust, so hopefully we will find the best way to do it before then.

 

[jeff4136]

You might try offsetting a surface 1/2 T (or some distance to guestimate neutral axis) and (Insert, Advanced,) Flatten Quilt.

 

[3dlogix]

I worked for a packaging company a few moons ago, and we developed hundreds of of funnels, some simple and some complex. The most complex ones involved transitions from round to oval sections. The trick with any sheet metal part is understanding a very basic concept. If you were to take the part you wanted to unbend, and analys the surfaces, you must have perfectly flat sections (except for the individual bend lines). If you were to take four points in space and define a surface through each corner, the final surface must be flat. If there is a twist in the surface, it won't unbend. One way is to develop sections that have 3 points, overextend the surfaces and make cut before the sharp points.

This post is intended for most new Sheetmetal users, I am sure you guys have much more experrience with it.

Steve

http://www.sprdesign.com

http://www.3dlogix.com

 

[Guest0527211403]

Jeff

I tried out your method, and after taking a bit of time to figure out how the flatten quilt feature worked, I got what I wanted, thanks.

I just got an quote on waterjetting the parts... I think I'll stick to printed templates and tin-snips

 

[jeff4136]

> printed templates and snips

That's the way I'd do it. If its a developable shape I'd try construction paper before cutting metal. If you try something like the hydro-formed piece, with compound curvatues, it's all a guess to start with.

You might play with the flatten quilt "resolution". Depending on the shape (amount of compound curvature?) you will probably see some differences. I can't tell you from practical experience how useful the results are. A good starting point, anyway?

 

[Guest0527211403]

I tried it today on a simplified model with cardstock out of my inkjet, and it seems to be close enough for my needs.

Just one more note- I had to add a small cut to the quilt, so that I could add a datum point to a corner. This sets the tangent plane for the flatten quilt feature.

I might have a technique for developing the blank if I'm going to hydroform it. My Variable Section sweep is driven by a graph feature which controls the diameter of the chamber through the trajectory (see here and here for a few good sources on design information for expansion chambers). I then made a separate variable section sweep which, instead of a circular section, is a straight line with the trajectory an axis of symmetry. I then control the length of this line with a relation that ties it to the same graph freature, multiplied by pi/2. This gives me a flat surface which is half the circumference of the pipe in width at any given point on the same trajectory.

I'm thinking this would work if I keep the rads on the trajectory large enough to prevent major overlapping. The only way I can really try is to weld it up and try and blow the steel sheets up like a balloon... If anything, it's a starting point.
Thanks again for the idea

 

[dgallup]

Hey, thanks for that link to Gordon Jennings Two-Stroke Tuners Handbook. I had a copy 30 years ago that I used to make a pipe for my Rickman 125! Haven't been able to lay my hands on it for decades. Never thought I'd find it on a PTC forum!