Cracked Rear Cross Frame 17 Footer

[LBaranyai]

As a side note, my husband transports his motorcycle on the back of our 17'. We have a motorcycle rack that fits into the rear hitch. He takes it on his hunting and fishing trips and drives a fair distance on both highways and gravel roads. We have not seen any evidence of cracks or deterioration from the additional weight load and we would since we are rather OCD when it comes to maintaining our home and vehicles. This particular "incident" serves as a good reminder that once you bring your trailer home, it's all up to you. Kudos for Thane's due diligence.

[ruscal]

This would make for some fun testing. Get a piece of 2x2x1/8" x12" tube and a couple of 2x2x 1/8" plates to weld on the ends. Drill and tap a pipe thread into one of the plates. Fill half full of water install plug and freeze. Observe results. If it doesn't split, thaw water and fill completely and re-freeze. Observe again. This could also test your welding skills. I would imagine something is going to give! Frozen water must be able to apply tremendous force. I have no idea if the tube could store enough energy to explode, so don't use your mother's best fridge for the test. I'm guessing it will crack slowly since we're not dealing with a compressed gas.
Russ

[Brian B-P]

Originally Posted by ruscal

This would make for some fun testing. Get a piece of 2x2x1/8" x12" tube and a couple of 2x2x 1/8" plates to weld on the ends. Drill and tap a pipe thread into one of the plates. Fill half full of water install plug and freeze. Observe results. If it doesn't split, thaw water and fill completely and re-freeze. Observe again. This could also test your welding skills.

I don't think this is a test of welding at all - you could just duct-tape the ends. Because there is a huge air space above the water, it won't build any pressure, so you just need to keep the water in until it freezes. When it freezes it will bulge out the ends to some extent, but that's not going to change the fact that the ice can only burst the tube if expands horizontally against the tube walls with great pressure even though the top surface (and the ends in my duct tape version) is unconstrained.

To make it really realistic, take a sandblaster to the lower corner on one side, then leave it to rust for a few years, before the freezing test.

Water freezing in the bottom of an open bucket can certainly split the bucket, and that's essentially what is being expected in this test. I just doubt it's the primary cause of the failure in the crossmember.

[Ron in BC]

Originally Posted by Brian B-P

To make it really realistic, take a sandblaster to the lower corner on one side, then leave it to rust for a few years, before the freezing test.

Water freezing in the bottom of an open bucket can certainly split the bucket, and that's essentially what is being expected in this test. I just doubt it's the primary cause of the failure in the crossmember.

Yup and add in some prolonged flexing from torsion and I bet it'd crack.

Ron

[ruscal]

Originally Posted by Brian B-P

I don't think this is a test of welding at all - you could just duct-tape the ends. Because there is a huge air space above the water, it won't build any pressure, so you just need to keep the water in until it freezes. When it freezes it will bulge out the ends to some extent, but that's not going to change the fact that the ice can only burst the tube if expands horizontally against the tube walls with great pressure even though the top surface (and the ends in my duct tape version) is unconstrained.

To make it really realistic, take a sandblaster to the lower corner on one side, then leave it to rust for a few years, before the freezing test.

Water freezing in the bottom of an open bucket can certainly split the bucket, and that's essentially what is being expected in this test. I just doubt it's the primary cause of the failure in the crossmember.

Brian,
I ran your duct taped version of the test last night. I was like a kid on Christmas morning anticipating the amazement I would experience upon opening the freezer. Dang! the tube was totally in tact! In fact I calipered the tube in both dimensions and it hadn't expanded at all. I had a piece of 2x2x 1/16" x 18", and duct taped the ends up half way, filled it half way and froze it overnight. Do you think the water has to be constrained in all directions? I tend to think so. When I get time I'll weld the end caps on and fill it completely to see what gives.
Fun stuff,
Russ

As to why that xmember failed, it's a good question... Rust certainly helped. It is strange how the tears in the metal look "puckered" out. By design, that member shouldn't be that highly stressed under normal loading. When I added the receiver to the back of our trailer I didn't have much space to spread the two tubes apart from each other. That of course increases stress to the tubes. To help stiffen the system I mitered the front tube in a gentle "V" so it wouldn't bow up and down so much under load. The Escape trailer layout is superior to mine due to more spacing between the tubes providing more leverage. It appears they also have 2 longitudinal tubes going perpendicular to the cross tubes which would help carry some load and stiffen the system.

[ruscal]

I re-ran the test by constraining the ends with caps welded to the ends. I also wanted to compare two weld filler metals, using 309 stainless steel on one end and 70, 000 series steel on the other. I installed a brass plug in one end plate for filling with water. I filled it completely and froze it overnight. 0 degrees.
Upon examination in the morning I found the tube swollen, and the endcaps also swollen. No splits in the tube, no failed welds. Kind of anticlimatic.....
A box tries to become a sphere when pressurized, so really is not constrained, even with the caps. With enough freeze/thaw cycles you would expect the corners of a square tube to crack if filled with water and constrained on the ends. It is unlikely that it happened to Thane's trailer, since water level was probably very low inside. Water pipes would split when full of water and frozen because they are round to start with and can't expand like a square tube.
First pic is empty tube
Second pic shows swollen surfaces
Third swollen end cap

[Ron in BC]

There's nothing like empirical testing.

I've welded 316 tube to mild steel using 316 filler and even in tough situations such as a steel exhaust manifold to a s/s injection elbow it's never been an issue.

I understand the theory that it's not the best practice but from a practical point of view it does work.

Now, who's going to weld up a large Tee out of square tube and subject it to repeat torque loading?

Ron

[Brian B-P]

Thanks Russ
No surprises, but good confirmation.

[ruscal]

Originally Posted by Ron in BC

There's nothing like empirical testing.

I've welded 316 tube to mild steel using 316 filler and even in tough situations such as a steel exhaust manifold to a s/s injection elbow it's never been an issue.

I understand the theory that it's not the best practice but from a practical point of view it does work.

Now, who's going to weld up a large Tee out of square tube and subject it to repeat torque loading?



Ron

Ron,
I attempted to do a torque test, but had a SNAFU. I'll post pictures in another post.
I used a 5" long piece of 3/4 x 3/4 x 1/8" tube, which was welded to two levers.
One lever was clamped to the bench, and the other was free to use to pivot the tube. I had no real way to apply millions of torque and release cycles to the fixture, so decided to try deep cycling the tube until failure. First I torqued it about 5 degrees for 200 cycles. There was no signs of distress, so started torquing past the yield point. The tube was torqued until it held a 10 degree rotation, then returned back to 0 for another 367 cycles until it failed. That tube could really absorb way more abuse than I expected! 10 degrees is a lot of twist. The SNAFU that I mentioned above was that the tube decided to fail at the heat affected zone, screwing up the test. I would have had to have inserts pressed into the ends of the tube to apply the torque. One thing I did learn though was that when torqued to yield the tube sides were concave, and when returned to zero rotation the sides were flat. That means the corners of the tube would eventually crack at some future cycle number. With a round tube I would expect the cracks to spiral like when a skier breaks a leg from catching a ski tip.
How does all this explain Thane's trailer cracks? I have no clue, but something attacked that poor tube.
Russ

[ruscal]

Pics of the test fixture.

[Thane]

Well I sure caused a lot work for some fine folks. Well done!

[Ron in BC]

Originally Posted by ruscal

Ron,
One thing I did learn though was that when torqued to yield the tube sides were concave, and when returned to zero rotation the sides were flat. That means the corners of the tube would eventually crack at some future cycle number.
How does all this explain Thane's trailer cracks? I have no clue, but something attacked that poor tube.
Russ

Maybe you found the smoking gun. That coupled with compromised structure from road debris sandblasting and rust might be the answer. Great effort at solving the puzzle.

Ron