Solidowrks: Rolled Stuctural Channel? 2

[mdoerner]

Hi All,

I just completed a Solidworks training course. I work for a Manufacturer of Conveyor components. I'd like to use my Solidworks educational license before it expires in November on a little problem.

I'm aware the sheetmetal component of Solidworks will allow you to flatten a sheetmetal object. What I'm looking for here is something similar, but for structural steel. I have a conveyor frame that is to be bent on a constant radius (outside) of about 5'-6' out of C4 channel. There will be hex-axle holes drilled through the side channel for tapered rollers prior to being rolled. What I'd like to is build my conveyor curve with the hex-axle holes, the "flatten" it like sheetmetal to give precise hole locations for the machinist prior to rolling. Is this possible in Solidworks? Thanks in advance.

PS I suppose I could use the "thickest" sheetmetal available in solidworks, as that would be similar to the C4 channel, but the "stone and tablet" guys might not get it.....Thanks again.

Mike Doerner

 

[CorBlimeyLimey]

The sheet metal mode is not suitable for bending and straightening of the channel.

Your best bet would be to create the straight C4 channel in one config, and then use the Flex function in another.

 

[MikeHalloran]

I haven't tried this, and I'm at least a decade behind CBL in SW, so do consider the source...

I'd try to superimpose (within the channel) a sheetmetal part rolled to the radius of the channel's neutral surface, put holes in that, and use it as a template to locate cut/extrude holes in the formed channel. Then the blank for the sheet metal could be used as the machinist's template.

I'd be expecting a lot of warnings and cautions from SW about not being able to make sense of it all..



Mike Halloran
Pembroke Pines, FL, USA

 

[mdoerner]

Flex function? 1st I've heard of it......do tell.

Mike D

 

[CorBlimeyLimey]

http://www.archive.org/details/BenEadieFlexCommand_0

http://www.youtube.com/watch?v=IKIGU7YMR3w

 

[kellnerp]

I would not suggest using the flex feature for this. I am going to assume that you are bending the C4 the hard way for one of the channels and the easy way for the other. The location of the neutral axis in the bending operation will be different and this will change the overall length of channel needed and the position of the holes.

When the sheetmetal module performs a bend in flat sheet it applies a bend allowance to account for the stretching that occurs and this is based on a "k" factor that represents the location of the neutral axis during the bend. A "k" of .5 indicates the neutral axis stays in the geometric center. Among other things the neutral axis in a C4 is not in the center of the flange and it will tend to move quite a bit depending on the bend direction.

You are going to have to come up with a "k" factor for your bending operation in order to do what you want. You might be able to find this out by measuring the change in hole spacing before and after a bending operation. This change will be greater as the bend radius changes so you will have to gather data and chart it to be able to handle all design cases that come your way.

Once you know the "k" factor you can model the C4 in it's curved state and dimension it using arc length dimensions. Transfer those to the straight C4 model directly and then scale everything in the long direction according to the "k" factor you have determined. You could do this in equations or with a design table to maintain documentation.

The flex feature is hard to control and may not perform the "bend" in a way that is true to real life because it doesn't account for either plasticity or shear effects in bending.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[CorBlimeyLimey]

kellnerp ... Agreed, but once the K-factor is determined, the Flex Triad can be positioned to suit. It still may not be exact, but should be as close as the Sheet Metal mode would offer. Even the SM function does not reflect true plasticity and deformation. Thinning and buckling is represented.

Whichever method is chosen, two configs would be required to depict the straight and rolled profile.

 

[kellnerp]

mdoerner,

I made one misstatement. I said," Bent the easy way or hard way" Should be "leg in or leg out."

CBL,

The flex triad maintains plane sections as plane.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[CorBlimeyLimey]

Correction ... My last post should have read, "Thinning and buckling is not represented."

kellnerp ... "The flex triad maintains plane sections as plane."
I'm sorry, I don't understand that statement. Would you expand on what you mean please? I have used the offset Flex feature in the past, with excellent results.

 

[kellnerp]

It's an FEA thing. It means no plasticity is accounted for.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[DiegoLGraves]

I've used two configurations for rolling structural steel shapes. One shows the finished part with holes; the other configuration of the straight length is in the same file, but has no common features or sketches, it's completely separate.

The "k-factor" on structural shapes changes dramatically with the equipment being used, how tight or loose the rollers used follow the metal and how much the channel legs are trapped (kept from buckling.)

If this is a short run and the hole placement is critical you'll be better of putting them in after rolling. If you are running a larger qty you'll probably need a sample run; just make sure the sample is rolled using the same method used for the production run. Also, unless there are straight sections at each end of the channel, you may need to allow for cut-off, as it's difficult to roll all the way to the end.

FWIW, Regards, Diego

 

[CorBlimeyLimey]

kellnerp ... SOrry to keep on, but are you suggesting that the Flex feature does not change the inside and outside arc lengths (in effect allowing for plasticity) when the Triad is offset?

 

[Overworked1]

I also work for a conveyor manufacturer, and although I do not typically deal with our curved conveyors, I do know that other people in our company are using the flex feature for rolled channel sideframes. I believe there was alot of testing done in the early days to make sure the k-factors were done correctly.

Also, the straight and curved configs like Limey mentioned is how we do it.

Aaron
SolidWorks (x64)/PDMWorks 09 3.0
CADKey 99 R1.0 (Yes, still using it!)

 

[mdoerner]

Wow guys, thanks for all the info. I'll see if I can't play with this over the weekend.

Mike Doerner

 

[Overworked1]

I lied. I just pulled up some of the curved parts the other guys have made at my company and I guess now they are creating their curved channel parts using a circular sweep.

Aaron
SolidWorks (x64)/PDMWorks 09 3.0
CADKey 99 R1.0 (Yes, still using it!)

 

[kellnerp]

CBL, NO I am not suggesting that.

Read this and see if it makes any sense to you.

http://www.google.com/search?hl=en&source=hp&q=plane+sections+remain+plane&aq=f&oq=&aqi=

Try the first wikipedia article.

BTW, changing the inside and outside arc lengths is something that happens whether or not there is plasticity. We need to know the length along the neutral axis wherever that ends up to be and the flex feature does not control that. As Diego mentioned, that is something highly dependent on how and what you bend. There is no one size fits all solution, flex feature, sheet metal or as Overworked found out just modeling two different shapes.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[CorBlimeyLimey]

Thanks for the link. I will read that tonight.

I fully agree that no one-size-fits-all solution exists. One of the beauties of SW is being able to employ various methods of creation. There is often no right or wrong way, just different ways.

I was more curious as to why you disapproved of Flex.

"We need to know the length along the neutral axis wherever that ends up to be and the flex feature does not control that."
Actually the neutral axis always passes through the Triad centre, so it's length is always known. Knowing where to place it is the trick.