How can I manufacture a reliable all-steel structure with cross ribs????

[hmalfs]

Hello,

So I want to manufacture a structure with cross ribs, as shown in attachment. My question is: how can I manufacture it? I think it is better to manufacture the ribs and the plate separately. I was thinking about using angle soldering, but the surfaces have much different dimensions, and those must not exceed a ratio of 3:1. The main problem is: how can I manufacture a structure that is rigid enough, even with a small contact area on the intersection of the cross ribs, due to the low thickness, which is approximated equal to 2 mm. The small contact area is a problem mainly for adhesive bonding. Also, furnace brazing is not an option, as It will surely create a gradient of chemical composition, and could certanly invalidate experimental testing results. Also, I was thinking of bulding one diagonal, and then bonding or soldering the other two parts that belong to the other diagonal rib, one at each side.

Any suggestion appreciated.

Thanks,
Hugo

 

[MikeHalloran]

What experimental testing?
Why would anyone care about chemical composition of a structural part?
Since you haven't given us a clue about what you're trying to do, we can't help much.

I.e., you've told us a little about your solution, and noted correctly that it may be difficult to manufacture in certain ways, but you haven't told us what problem you're actually trying to solve.



Mike Halloran
Pembroke Pines, FL, USA

 

[berkshire]

hmalfs (Mechanical)
You do not give us much to work with here. You do not say what material you want to use, and you do not give us the size of the part.
If you put fillets in the corners and at the base of the vertical flanges, you could machine it with a CNC mill. You could EDM the part with a sinker die. You could also Chem-Etch the part. all of these methods have their advantages and dis-advantages.
Or as you stated in your OP you can join the part from separate components.
The questions un answered are: What does it have to do, how much are you prepared to spend?

You are judged not by what you know, but by what you can do.

 

[hmalfs]

Hi Mr. Mike,

Thanks for your reply. Below I'm answering your questions:


What experimental testing?

It is for simple bending and torsion loads, both separately and coupled.

Why would anyone care about chemical composition of a structural part?

Chemical composition is important, because the Young's Modulus depend on it.

The problem i'm trying to solve is actually how to bond the ribs to the plate, and the several parts of the cross ribs between them, in such a way that the structural reliability is not compromised.


I hope I told enough to get a solution. Please let me know,

Thanks,
Hugo

 

[berkshire]

My bad,
I just noted in your title that you are using steel.
B.E.

You are judged not by what you know, but by what you can do.

 

[hmalfs]

Hi Mr. Berkshire,

Thanks for your reply,
I'm answering your questions below:


You do not say what material you want to use

Structural Steel, can be a Dual-Phase or another

and you do not give us the size of the part.

The plate shall have dimensions of 2*300*1000 mm. The ribbed structure is to be later incorporated into a beam or another geometry, with a total width b of 150-230 mm

What does it have to do, how much are you prepared to spend?

It has to be reliable in experimental tests, mainly in transverse loads bending and torsion, both coupled and separated. The experimental tests may only be performed in the elastic domain. I can spend on the low mid xxxx USD, but please bear in mind that in this application, price is a very very important factor (it is not a performance application, where expensive materials are commonly used)


I hope I told enough to get a solution

Thanks,
Hugo

 

[berkshire]

If this is a test fixture, which has to be stiff, why are you limiting your plate thickness to only 2mm.
Is weight and inertia a consideration?
B.E.

You are judged not by what you know, but by what you can do.

 

[MikeHalloran]

In that case, I'd suggest cutting flat blanks by laser or waterjet, with a fair number of tabs and slots, with the tabs not quite extending all the way through the slots, and puddle welding the tab edges to the slots, working entirely from the flat side.

The rib corners and intersections are easily accessible to a TIG torch, from at least one side.

Virtually the same parts could be furnace-brazed or dip-brazed.

You'll probably find that buckling of the ribs will limit the resistance to the loads described. Are you allowed to fold the free rib edges to stiffen them?



Mike Halloran
Pembroke Pines, FL, USA

 

[berkshire]

If you make your flat plate a little thicker, say 2.5 mm, then you can machine flat, after all of the tab welding and plug welding is done. The two side flanges could certainly be folded up if the outside bend radius is not a problem
B.E.

You are judged not by what you know, but by what you can do.

 

[Tmoose]

How perfect does the 'other" side of the plate need to be?
I was thinking flanges on several edges of each ribs, then spot welded or bonded and riveted

http://fsims.faa.gov/wdocs/alerts/MAY2006_files/image016.jpg

http://bmwvmca.org/joomla/images/images/NewsV8N2/picture13.jpg

http://www.millerwelds.com/education/articles/images/Canarrow-TIGspot1low.jpg

 

[hmalfs]

Hi Mr. Mike,

Are you allowed to fold the free rib edges to stiffen them?

No


Can you post a textbook or something about what you suggested? And can you detail more your explanation?


thanks,
Hugo

 

[hmalfs]

Hi Mr. Berkshire,

After extensive study, we found out that the lower the thickness is, the better will be the behaviour of the part in relation to unribbed parts. This was tested numerically with very fine mesh.

 

[BrianE22]

Any welding will likely warp the structure. If you need a flat top surface you may want to leave some machining stock. It looks particularly vulnerable to bending in one plane. Can you add faces to the two open ends?

 

[hmalfs]

Hi Mr. Brian,

Thanks for the reply.

It is not really a problem, because this ribbing pattern will be applied to several structures, all of them already tested.

 

[dhengr]

Hmalfs:
Seems to me that you’ve asked this same vague question several times in your Ops, in only slightly different ways, and never given enough meaningful problem info. to really generate anything other than a guessing game. Isn’t the question how to design and build a closed box beam section, of unknown length and loading, of light gage steel material, and of fairly small width or depth so as to make welding and final fabrication very difficult? Internal stiffeners may be important/required to reduce outer skin buckling, and to improve longitudinal, lateral and torsional stiffness.

Have you tried some of the suggestions from the earlier threads? Without knowing more about the actual problem, we could go on guessing and suggesting slight variations forever and not really get near a practical solution. Then you ask for references and text books on subjects like the moment of inertia, the center of gravity of a section..., and one can’t help but wonder why you aren’t asking your teacher for help on these matters, or looking in your own school text books. Then you use CAD and other computer software to start to do problems which should be second nature to an aspiring young engineer. What’s going on here? What is your actual engineering background and engineering position? There are a bunch of smart people here on E-Tips, most always willing to be helpful to a young engineer, but we are not in the business of teaching beginning high school or collage science and engineering classes.

 

[ornerynorsk]

Perhaps a good foundry could cast it for you? Does it need to be of wrought steel, or would a casting be permissible?

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[MintJulep]

You've stated several times that you've done testing.

How did you manufacture your test samples?

 

[berkshire]

hmalfs

After extensive study, we found out that the lower the thickness is, the better will be the behaviour of the part in relation to unribbed parts. This was tested numerically with very fine mesh.
That is why I was suggesting that you machine back to the desired thickness after any welding operation.
The other thing you can do is to bend flanges on your ribs and spot weld them into place. you can also make your ribs thinner, and spot weld, two of them back to back, before spot welding them into the tray. The other thing I would suggest is to disregard the FEM and physically test a part.
B.E.

You are judged not by what you know, but by what you can do.

 

[tbuelna]

hmalfs-

If the model shown in your picture is proportionally accurate (ie. the ribs are 2mm thick), then it can easily be investment cast using rapid prototype tooling. Investment casting will allow you to optimize your design for stiffness and stress by using fillets and tapering of the rib/base section thicknesses. Investment casting is also very dimensionally accurate, so the casting would require a minimum of finish machining.

 

[ornerynorsk]

There is also a rapid sandcast process that directly "prints" the cope and drag from a sand/polymer mix.

http://www.youtube.com/watch?v=SgrGhJ5u5oM&feature=player_embedded

No connection to the company portrayed in the video.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.