[Chips O'Toole]

As a highly unskilled welder, I am in no position to judge, but I am starting to have certain suspicions about the people who teach welding on Youtube, and I wonder if anyone here would like to confirm or refute them.

Youtube is full of people making very long, continuous demonstration welds on coupons as well as assemblies that will never actually be measured or used. Nearly all of the time, the welders don't even mention the problem long welds cause: severe warpage. They're welding things they're just going to throw out, so they don't worry about warpage. My suspicion: they ought to be telling us not to make long welds unless we absolutely have to, because a) short welds warp much less, and b) long welds are nearly always totally unnecessary for the strength of what we're working on.

I saw a video guy putting 6"-long stick welds on scrap steel. When does it make sense to make a 6"-long stick weld without stopping? I would think that would be reserved for big projects that have to be extremely strong but don't have to be very straight.

I've been putting half a square inch of 70,000-psi weld on joints that will never see more than 50 pounds of tension. Seems like a stupid idea in retrospect.

Youtube is also full of people making very pretty uninterrupted welds, but in reality, they should be telling us to make interrupted welds, a short segment at a time, and skip around, no matter how ugly the welds are. We should get used to making ugly welds that don't warp much and then grinding later to make them cute. True or false? To watch Youtube, you would think grinding was a sin, but it seems like it's the correct and normal thing to do.

I started building an arbor press stand, and it warped even though I had it clamped down when I welded it. I had long welds in various places. I cut it apart, added some bracing, and put it back together with short welds. Now it's much better, but the welds are really ugly, so I'm grinding the ugly off. It will look great when I'm done, and it will last forever, but I didn't weld it like the Youtube guys.

Here's another question: what do people think of making short welds and then filling in the gaps with something like JB Weld? I don't want to have big gaps on my arbor press stand, because they look bad and are hard to paint over, but 10"-long welds are a complete waste of wire, gas, and time. It occurred to me yesterday that I could just cram some kind of epoxy or putty in the gaps and paint over them. Genius or terrible idea?

Seems like there are a lot of guys on the web teaching us things that look nice on video but don't make any sense when you're actually trying to get a project built. On the other hand, I know nearly nothing about welding, so I could be wrong.

01 16 20 arbor press stand after disassembly and new welding underside small.jpg (207.71 KiB) Viewed 3414 times

[Bill Beauregard]

50 years now distortion has been my biggest challenge. I find myself struggling between the "I'm not an engineer, how much weld is just enough?", and putting too much weld, then figuring out how to straighten.

If you find the formula, let me know.

[Poland308]

There are times when making a longer weld is advantageous. IE less start stop points, these are possible points for leaks on pressure vessels or containers, also start stop points are often focused points for stress failures or torsional forces. Warpage is a concern that is more or less a trade off. If a little warpage isn’t acceptable for your application then you might want to consider other means of connection. If it has to be welded and it has to be warp free, then there are common ways to over come this. Over build, make it out of a base material that is thick enough to be machined flat or too big to be warped by the stress of the needed weld. Most times you don’t need a total penetration or need a part to be welded on all sides to be strong enough for the intended use. Intended use is a loaded variable of its own, its often coupled with the risk of catastrophic failure.

[Oscar]

One word: Scope. You can't cover everything in one video or book, or college course, or degree, or even 50 degrees.

"I've been putting half a square inch of 70,000-psi weld on joints that will never see more than 50 pounds of tension. Seems like a stupid idea in retrospect."

You don't need 35ksi of stress/tension to fracture ½in² of a 70ksi weld joint. depending on joint geometry and load application, it can and usually is substantially less than that. Again: Scope.

[Chips O'Toole]

One word: Scope. You can't cover everything in one video or book, or college course, or degree, or even 50 degrees.

You don't need 35ksi of stress/tension to fracture ½in² of a 70ksi weld joint. depending on joint geometry and load application, it can and usually is substantially less than that. Again: Scope.

Sure, but you're talking about special cases where there would be twisting, metal fatigue, shock, or some other problem. The welds on my stand are going to have very easy lives.

[JayWal]

Like Oscar said it depends on the application. For example if you weld 2 coupons in a t joint and slap a 1" weld on one side only, that piece is strong is some ways, but can be easily broken if loaded other ways. So you can either put a full weld on one side or 2 shorter welds on both sides. Sometimes you need a full pen weld, other time tig fusion tacks is all you need. Depends on application and loading

Sent from my SM-G970W using Tapatalk

[Oscar]

One word: Scope. You can't cover everything in one video or book, or college course, or degree, or even 50 degrees.

You don't need 35ksi of stress/tension to fracture ½in² of a 70ksi weld joint. depending on joint geometry and load application, it can and usually is substantially less than that. Again: Scope.

Sure, but you're talking about special cases where there would be twisting, metal fatigue, shock, or some other problem. The welds on my stand are going to have very easy lives.

True, but the issue is again scope. You have to "limit" the conversation somehow, or you'll never end up covering everything because even what we do know, is enough to fill a few lifetimes' worth of conversation.

[Poland308]

Some torsional stress is derived from welding on steel, especially square or round tube, just from welding on it. Most tube steel has a welded seam just from the manufacturing process. You look close at a full 21ft stick and even untouched it may have a twist in it. Cut unto it even only partially through and you will see the twist change. It may even pinch your saw blade. Point is, there are almost always torsional stress even before you use your steel. Even temperature changes will often cause enough contraction or expansion to create stress that can be enough to cause weld failure. But like others mentioned this can come down to joint design. Every welds success or failure is a direct result of specific design criteria and only after that is it a weld maker issue. Even the best welder in the world fill have a joint failure if the joint design is inferior.

[tweake]

I saw a video guy putting 6"-long stick welds on scrap steel. When does it make sense to make a 6"-long stick weld without stopping?

a lot of gear is welded with huge long runs. especially with mig when runs can be many many meters long.
6" weld can considered a tack weld in some situations.
it all comes down to what your dealing with.
however thats kind of out of the scope of welding and more into the fabrication side of things. you have to keep in mind welding is only 10% of the job. learning all the fabrication skills like fitup or how to combat distortion, is another thing again.

edit: one of the problems of being self taught is many people tend to leave important information out by accident. its finding all that info and putting it all together takes a long time.

[Bill Beauregard]

There is a glue advertisement where a man glues his hat to a hoist fitting. He glues himself using 1 square inch of surface. He then hangs high in the air. A tensile strength of 200 psi isn't hard with most glue. If you add any of several stressers, It'll separate. You have a lot of load calculation before you know the factors. Butt welding a horizontal I beam, you need all you can get.
Holding deck plate to that I beam, you need less.

[homeboy]

Chips: Referring to your first post regarding the "short,ugly" stitch welds I have a few areas in my "hobby" trailers I build that are a flat or tube Butt or T joint stitch welded. My main concern is sealing the joint from water penetration but also for neat looks. I simply clean everything up, prime and use paintable automotive seam sealer chalking ( need to be or know someone who's handy with a chalking gun) and paint over. Do it neat and anyone will think your the best welder ever.

[Chips O'Toole]

I'm thinking I may go get some Bondo. Then I'll show the beautiful Bondo arbor press stand to other people and make fun of their welds.

[Poland308]

While using body filler is a common tool in a mass production setting. It’s important to be upfront about using such a tool with any potential customers. Silicone is my tool of choice if I know a part is structurally proven strong, but it needs to have cavities sealed up from the elements. This is where joint design can excel. Welding only enough to be solid keeps time and filler costs to a minimum, and reduces stress and warpage.

[homeboy]

While using body filler is a common tool in a mass production setting. It’s important to be upfront about using such a tool with any potential customers. Silicone is my tool of choice if I know a part is structurally proven strong, but it needs to have cavities sealed up from the elements. This is where joint design can excel. Welding only enough to be solid keeps time and filler costs to a minimum, and reduces stress and warpage.

For me the main criteria is that it's durable, paintable and flexible. For places out of sight I mask off the area and use a paintable undercoating ( similar to stone guard) in a spray bomb to seal the seams.

[DennisCA]

Here's a guy who did infact talk about distortion and weld short stretches randomly to get around it:
https://www.youtube.com/watch?v=a_RlL1O-bK4

Around 4 minutes in for the relevant stuff.

[arenjenkins]

There is no one size fits all for any welding. There are completely acceptable and desirable 6' long welds, 6" long welds, and 1/2" long welds. It really all depends on the procedure, process, and weldment. Sometimes the angle is getting less warpage, sometimes its to make a weld seam that will be leak proof to liquid, sometimes it's to hold 1lbs, sometimes 100,000lbs, sometimes interpass temperature is the highest concern, or haz, or joint ductility, is the material .030" thick? 30" thick? Lap joint, open corner joint, ext...Point is, to say that it's just plain wrong to make long welds is kinda like saying "The best color is orange."

[snoeproe]

The tests for AWS D1.1 and CWB csa w47.1 are on 6” long plates.
Other than the stop and restart required for the root passes, you don’t want to stop in that 6” weld zone. If you do stop in there because your worried about distortion, your a fool and haven’t been properly trained.

I did a job one time for a fellow who moves buildings. I took 2 40’ long I Beams and joined them to make 1 box beam. I made welds much longer then 6” on that project.