Why you shouldn't cool your welds with water & info on weld strength 

I could definitely butcher a fine dining meal 😅

Aluminum has a real bad “hot short” problem where if it cools fast the welds will crack. If you run a fusion weld with no filler down a thick plate of aluminum, and have the machine set to pulse, you can literally watch the material crack every time it goes off the pulse current to background current. Pretty wild stuff lol.

I will have to drum up something. Maybe burritos in the rod oven segment 😅

@@makingmistakeswithgreg I'm envisioning the thumbnail 🤣

I will definitely give that a shot to see what happens. To me what’s scary is the thought some idiot is out there welding on things and dumping water on the welds to get the job done faster lol. You would never know looking at them but a decent shock load and it’s toast. I expected them to more or less tear at about half a bend, but that second one literally blew up once the plate bent at all.

Definitetly, it would be very interesting to see how normalizing/annealing may or may not mitigate the effect of quenching, and I would go a step further and run a set of unquenched pieces through the oven as well to see what if any difference might be there. If I recall, 450°F might be just at the lower bound of annealing temperatures depending on the steel. So, if possible, a higher temperature might be desirable. (But I suppose it depends on the oven; I used my apartment oven for some school project parts years ago, never was sure if it actually worked or not, lol) Another thought is to test all pieces to failure using different geometry that will allow a break to occur at some point. My thinking here is that the unquenched and annealed piece may actually end up being the strongest, but it would be interesting to see the absolute limit of each combination of variables. As I saw mentioned in another comment, time before quenching could add another dimension to this. Obviously, this could easily exponentiate out of control with all the variables though.

@@GimpGladly The process of tempering is well known. I would recommend that you look into post weld heat treatments. For example, on C-Mn steels, a PWHT is commonly performed at 1050F to 1100F to temper the transformation products in the weld and HAZ. This is fairly common in thick sections (> 0.5") where the cooling rate is sufficient to cause transformation to martensite and retained austenite. In the oil and gas industry many fittings are now provided in a quenched and tempered (Q&T) state to provide for higher strength. hth

@@makingmistakeswithgreg The ductility of martensite (the principal component in a quenched steel) is almost nil. Compare this to fully annealed steel which can have a ductility (elongation) of up to 30%. Quenching any steel is not a good idea unless you plan on tempering it. Ideally, tempering of quenched steel should be done at 500F to 900F and PWHT at 1000F to 1100F. The higher temperatures will result in a more ductile weld. 450F is about the lowest temperature and I recommended that as you can use your kitchen oven for that. You will need to leave it in at temperature for at least one hour. Figure 30 minutes for the piece of steel to come to temperature. I have a convection oven that I use for preheating large components especially aluminum prior to welding. PS: Doing face bends is better for examining the weld. This does tend to always break the weld and allows examination of the depth of fusion and for porosity.

Good info. Once I can locate a shop I am Looking forward to experimenting with blacksmithing, I know very little about it. Speaking of temps on plates, I am planning on chilling a plate to -30 atleast and welding it while cold. Any predictions on the results? I have welded outside when it was -14 but I preheated the material to 150 first. It’s amazing how much heat it takes to heat up even 1/4 in plate that’s below -10.

It seems like that weld had zero flex to it. The second the plate started flexing (and some pressure was put on the weld) it instantly gave up. I expected the weld to tear slowly and fail, not instantaneously fail. It would have been fun to hit one with a hammer because I bet it would have broke instantly too.

I would imagine they could have used an induction heater coil run off an engine drive. Considering the temps, amount of metal, and circumstances I would imagine they must have heated the pipe. I am going to chill down a fillet weld to -50 and weld on it cold to see what happens. I would think it would fail like it did in this video, it will be interesting to see what happens lol.

And they very often use an old Dr Pepper (Other brands are available) bottle to squirt water on the weld, very cringe-worthy indeed.

Some of the rods those guys run seem to be longer than what we can get, they may be home made lol. I think they bend to rod to make it easier to weld around something. When it burns off close to the middle they just break the rod in half and keep welding lol. I love watching those guys, they will fix anything with those rods lol.

Even with a part that's loaded below 5t you have to watch out for mulch quicker fatigue failures. It's been a while but Blodgett talks about it in one of his books specifically about welding in bad weather causing welds to fail at super low cycle counts

@@kevin-pk6hd Good point! Fatique failures are hard enough to predict that even aviation level engineering cannot guarantee perfect results.

Tempilaq?

That's kinda cool!

The amount of time wasted over repairs is definitely no joke, and not wanted lol. Defects are generally not acceptable especially if it lead to a failure after it is buried. I don’t doubt they would have xrayed everything, I can’t imagine the cost for the repair bill for a failed weld, probably far more than the X-rays cost lol.

I will have to do that to see if there is a visual difference 😀

No problem 😀. I would imagine when the temp of the weld is less than 300f degrees it would probably be less affected. I will have to do some research on this to get a better idea. I have a feeling even 20-30 seconds before water would drastically change the results. More tests will need to be done 😀

Especially the ceramics are not used as heatsink. They, in my opinion, prevent the weldpool physically from dripping off. Aluminium or copper backings may have a decent heat dissipation but you should grind these roots out anyways and reweld them from the back because if you alloy steel with these, the results may be unpredictable (brittle). Just try tigbracing with to much heat - the alloy is not even scratchable with a file....

So with a backing bar that pulls heat out I would think it would be slow enough not to matter. A ceramic trough strip like they use on big welds (typically in the flat position on a butt weld to keep the backside from dropping) would insulate the molten pool enough that it would slowly cool. Where things get super crazy is aluminum. Aluminum is very hot short sensitive and the faster the molten pool solidifies the more likely the weld is to immediately crack. This is why you don’t leave craters at the end of aluminum welds, it will crack before it’s cool at that point.

Thanks for replying me! I've actually held the ceramic backing today, but the dinosaur colleague said it is never used because it's too expensive. And unnecessary, since we repair shipping containers which are patched up more than your grandmas quilt, and it's mostly vertical and horizontal butt welding on the walls, and low amount of bottom perpendicular wall-to floor welding, 4 and 5mm max (at least for me, so fare). I've overseen that the ceramic is actually pure alumina which has high IR reflectivity and poor conduction. I thought it was intended to quickly cool the welds, my bad. As stated, I've not seen it before, as I'm a novice. I did see plenty of coworkers, including seniors, at my previous shit-show workplace, use aluminum plates for the same purpose- preventing dripping. I guess that's my in situ proof of your previous video's point - you've been doing it forever... But wrong. As for aluminum plate cooling, I've quenched many a blades between them. I made a simple vice attachment for two plates. In goes the hot stuff, I squeeze, it cools, and comes out straighter than Berlusconi. I can only imagine what the microstructure of the so rapidly cooled weld looks like. It's twice as hot, or more, than my knives were. That's some serious stressing. Thanks again! You could cover it in a video :)

It sure is. When playing with higher strength steel it can be a big no go to allow the temp to reach above or below interpass temp limits.

So I have done a tremendous amount of stick welding in the elements and I am not joking when I say it is very difficult to have 6010 or 7018 to put down poor welds with defects from wind, rain, etc. Stick welds can produce porosity from surface/metal contamination, typically from oil or grease. I have never tried, but I have a feeling if someone were to weld on 3/8th steel that was say -10 degrees, the finished weld may be very brittle. That’s something I will definitely have to try. Anytime I have welded in temps like that I preheat the material to atleast 150 degrees then weld it.

After plasma cutting, it is always wise to clean off the cut with a grinder to remove the millscale/oxidation layer anyway. That being said, if you did weld straight after plasma cutting without cleaing off then the surface is being melted again to a liquid state. The millscale/oxidation would be the contamination that would weaken the weld I reckon.

@@JonDingle I totally agree, and I am one who really preps any area to be welded down to shiny metal. I didn’t do a good job of explaining my question, and I’m not sure how really describe what I was thinking, so I will put it this way: the plasma cutter rapidly heats the metal along the cut line, and then it cools fairly rapidly by the remaining metal along the heat affected zone acting as a heatsink, so you rapidly heated and cooled that metal thereby changing the crystalline structure and essentially hardening the base metal along that heat affected zone (I’m guessing say an 1/8” to 1/4” in from the cut line) to a certain degree. I have noticed when grinding or drilling & tapping that affected metal that it seems considerably harder. Assuming that metal to be welded (as suggested earlier) as a fillet weld was properly cleaned and prepared; I am thinking that boundary of hardened metal will behave differently. Perhaps the weld process will anneal that as it goes molten and alloys with the filler and parent metal and then cools much slower than a plasma cut would. Perhaps I’m over thinking it, but my concern would be for a high liability weld it could be a factor to consider. In my ca se, I have a heavy duty mobility lift on the back of my truck that is under a tremendous stress when fully loaded, especially going over bump. It has cracked before, and quick repairs made (not by me). I am going to rebuild the lift and reinforce it with plasma cut gussets.

Same happens with laser cut parts yeah. But welding that hardened edge will reset the grain structure and it will be metallurgically just fine and dandy. especially if you clean off the oxide first.

@@analogplanet9675 Thank You - that is good to know and hear from someone who has experienced similar hardening along cut lines.

Not all rebar is weldable. If you weld high carbon non weldable rebar the welds will break. I know this from experience lol.

For that it comes down to your personal skill level. Depending on how strong the base material is I would use 7018, but that’s not the easiest rod to work with. My guess is most companies would weld something like that with 6011. It runs far better out of position than 6013. Flux core wire would also do a decent job easier than stick.

@@makingmistakeswithgreg Thanks for your reply. I don't have flux core so stick with have to do. I'm on a tight budget as I have recently given financial help to a friend recently divorced. A lot of people here in Europe recommend 6013 but I have been looking at your tests showing the 7018 welds holding up in your press. I'm not the world's greatest welder and I'm doing a lot of practice on scrap pieces at the moment in preparation for this project. I'm about to buy my first 7018s. Wish me luck. Thanks again. Brian.

Multi pass on thicker non high strength steel I would say sub 200 wouldn’t be an issue. On some steels it’s best to keep the temp above 150f (or even much higher) between passes. I will do some research to get specifics. The biggest issue is if your 3 passes deep on say 1/2in steel with 7018, and you start seeing half your weld glowing red the grain structure will be poor. I will shoot some videos on this and test stuff to demonstrate it 😀

It’s possible it would absorb hydrogen into the weld because the tail end of it was red hot when I dumped it. I don’t think that would directly cause the failure though because the material itself wouldn’t drastically be affected by it. I have welded with soaking wet 7018s (that were in a bottle of water for over a day) and they had no issues passing a bend test. The leidenfrost effect would likely keep the water from touching the molten metal until it was solidified. My assumption is that the grain structure of the weld was drastically affected by how fast it cooled. I am hoping a metallurgist could shed some light on this for all of us.

I might try that. My guess is it will still fail. Letting it air cool is the way to go.

Haha I got a good laugh out of that because I know it’s true lol. I have fixed a fair share of higher strength steel that broke and clearly the guy previously put down some big fat caulk beads (likely with no preheat) and had no idea why it failed. Luckily those types avoid watching videos or reading to find out ways to do things right. That’s good, it will keep you busy forever lol

That would be worse because the spray from a can is much colder than water in a bucket or out of a tap.

That’s an interesting question. The oil would likely not evaporate as fast, but I bet the results would be similar.

Well, that was interesting. The only things I've dealt with for hardening are O1 and W1 (aka 1095), but I've got some D2 for future projects. The W1, which is just high carbon steel, requires a water quench, and bigger pieces are prone to wrapping and cracking. O1 will harden in an oil quench, and is less prone to wrapping and cracking. D2 is air hardening, and supposedly has minimal distribution during quench. We'll see. At any rate, whenever you quench harden steel, it needs to be tempered afterwards or you get something that's glass-hard and glass-brittle, i.e. yield strength (bending) and ultimate strength (breaking) are very close together.

The book learning is optional, but making mistakes isn’t lol. That’s welding: 99% mistakes for a long time before you get to about 50/50 mistakes vs things go well lol.

your channel taught me more than any other channel on welding and i follow all of them all. you and have a way of explaining things like no one else. you point out your wrongs and rights and i have become a better welder since following you. for some reason i just like to joke around about the book learnins but I really do appreciate them and always look forward to your content book learnins and all i appreciate it . keep up the great content hopefully ill make enough in this business to become a patreon soon

Considering how hot it’s been it’s almost unavoidable lol.