Is Welding Worth It?

[Clayton Summerall]

Originally Posted by SystemS

welding is worth it if your design requires more customization than can be done to a kit bot frame. When a student is capable of welding a chassis correctly, the first time, it is quick and not a bad idea. One year a sponsor did it for us it was done wrong, after 2 weeks. That year it would have been better for a kit bot frame.

Why do you have to weld to have a not kitbot frame?? This year we had an 090 bent sheet metal chassis that had about 600 rivits in it with no welding.

[GeeTwo]

Originally Posted by SystemS

welding is worth it if your design requires more customization than can be done to a kit bot frame. When a student is capable of welding a chassis correctly, the first time, it is quick and not a bad idea. One year a sponsor did it for us it was done wrong, after 2 weeks. That year it would have been better for a kit bot frame.

Originally Posted by Clayton Summerall

Why do you have to weld to have a not kitbot frame?? This year we had an 090 bent sheet metal chassis that had about 600 rivits in it with no welding.

+1. Our 2014 Aerial Assist chassis was c-channel and tubes bolted together using flat angles and Ts as gussets. Even as violent as AA was, frame strength was not one of our problems.

[Gdeaver]

We weld, rivet, bolt, bond our robot assembles. Isn't engineering about evaluating and choosing the best method to solve problems?

[Spark]

If you use thin sheet alum it can sometimes be better to weld. Though usable, rivets and bolts could rip out or bend the metal. A long weld, not just a spot weld, has more surface area connecting the two pieces. However, if you need precise pieces it's easier to rivet or bolt. I'm not saying keeping welded pieces flat is impossible, just harder.

[Rivet Man]

Originally Posted by Spark

If you use thin sheet alum it can sometimes be better to weld. Though usable, rivets and bolts could rip out or bend the metal. A long weld, not just a spot weld, has more surface area connecting the two pieces. However, if you need precise pieces it's easier to rivet or bolt. I'm not saying keeping welded pieces flat is impossible, just harder.

Hi Spark,

Going to pretend I'm Weld Man here for a second...but if you're talking about bending the metal, a continuous weld is always going to produce more warping in a part than riveting if all else is equal.

Mainly what happens is you heat up the metal and it gets softer (more malleable, and a bunch of other science jargon). So what happens is it'll distort and actually pull to a side. Afterwards there will be additional stresses inside the material (as with all fastening systems, that's how they work) that keep the material bent out of place.

Even if you preset your material and had it exactly in place, a long continuous weld is 100% going to distort a thin sheet of aluminum. Fortunately, rivets have your back and could be used to hold the sheet in place to help reduce warping.

Here's a link that has some good tips on how to mitigate these warps.
http://weldingdesign.com/archive/avo...ion-thin-sheet
Main points from this link are things like, backward walking, non-continuous/stitch welding, and presetting.

It's true that rivets can bend a thin piece of aluminum, but that would mean you have other issues in your setup. If your pieces aren't mated correctly (air gaps, bends aren't proper) when you're about to rivet then yes, rivets will bend your rig...they're trying to pull the pieces together after all! And if we're talking about the retaining mandrel or the head impinging on the surface of the metal...chances are you're using too much force on too soft a material, that ones on you and not the rivet!

Rivets are only as good as the person/design that's using them, and they're definitely not the bad guy!

Best Regards,
RM

[JamesCH95]

There is a school of though that suggest stitch welding may be superior to seam welding for certain applications. If a crack starts in a stitch weld it will break through one stitch and then stop, providing a good visual cue to the operator/maintainer, and easy access for repair. If a seam weld cracks or un-zips it can propagate a large distance before it is detected and may be more difficult to properly repair.

What makes seam weld cracks harder to repair is guessing where to drill the stop-hole for the crack, because a crack will propagate further than is visually detectable. Of course X-ray or magnaflux (and other technologies) can help, but that's a whole other level from FRC. With a broken stitch weld the whole stitch can be cut and re-welded (if not already broken) with no need for a stop hole.

I'm going to stop myself here on the 'seam vs stitch' discussion unless there is a debate or questions that arise.

Originally Posted by Rivet Man

Hi Spark,

Going to pretend I'm Weld Man here for a second...but if you're talking about bending the metal, a continuous weld is always going to produce more warping in a part than riveting if all else is equal.

Mainly what happens is you heat up the metal and it gets softer (more malleable, and a bunch of other science jargon). So what happens is it'll distort and actually pull to a side. Afterwards there will be additional stresses inside the material (as with all fastening systems, that's how they work) that keep the material bent out of place.

Even if you preset your material and had it exactly in place, a long continuous weld is 100% going to distort a thin sheet of aluminum. Fortunately, rivets have your back and could be used to hold the sheet in place to help reduce warping.

Here's a link that has some good tips on how to mitigate these warps.
http://weldingdesign.com/archive/avo...ion-thin-sheet
Main points from this link are things like, backward walking, non-continuous/stitch welding, and presetting.

It's true that rivets can bend a thin piece of aluminum, but that would mean you have other issues in your setup. If your pieces aren't mated correctly (air gaps, bends aren't proper) when you're about to rivet then yes, rivets will bend your rig...they're trying to pull the pieces together after all! And if we're talking about the retaining mandrel or the head impinging on the surface of the metal...chances are you're using too much force on too soft a material, that ones on you and not the rivet!

Rivets are only as good as the person/design that's using them, and they're definitely not the bad guy!

Best Regards,
RM

To put some 'science jargon' there...

The metal will change shape due to extreme localized heating of the weld pool driving thermal growth of the material. Any softening is secondary this CTE (coefficient of thermal expansion) driven growth.

Warping can be managed by proper design, small or non-existent root gaps, fixtures, and sufficient tack-welding prior to seam welding. If a part comes out warped there are many ways to correct it that include bending, re-heating welds (sometimes with fixtures), cutting, stress-relieving (sometimes with fixtures), or any combination of the above.

[FrankJ]

If you want to see lots of examples on how to weld thin sheets, google automotive sheet metal repair. It is mostly done with steel, but lots of YouTube video. Welding in such a way to minimize distortion is an art and science. For heavier weldments you have to calculate and allow for shrinkage. There is only so much that stress relieving will do.

One large scale example of thin sheet frames are airplanes. They are mostly riveted or glued and riveted.

[mwmac]

Originally Posted by FrankJ

If you want to see lots of examples on how to weld thin sheets, google automotive sheet metal repair. It is mostly done with steel, but lots of YouTube video. Welding in such a way to minimize distortion is an art and science.

Some eye candy of the art of welding sheet aluminum...(click on build photos)
http://www.p4bynorwood.com/

[FrankJ]

Originally Posted by mwmac

Some eye candy of the art of welding sheet aluminum...(click on build photos)
http://www.p4bynorwood.com/

I think I saw some rivets... Anyway I want one. I think it is more than we are allowed to spend on an FRC bot though...

[mwmac]

Originally Posted by FrankJ

I think I saw some rivets... Anyway I want one. I think it is more than we are allowed to spend on an FRC bot though...

There are lots of rivets but the welded seams on the front and rear fenders in particular are quite impressive imo. WRT the rear fenders, you can see the use of several holes to enable clecos to hold the panels tight while welding the seam.

[FrankJ]

Originally Posted by mwmac

There are lots of rivets but the welded seams on the front and rear fenders in particular are quite impressive imo. WRT the rear fenders, you can see the use of several holes to enable clecos to hold the panels tight while welding the seam.

I didn't mean to imply there wasn't welding. I still want one. Or a GT-40, 917...