how do i no if my welds are going to be strong enough ??

[FAAace09]

im not new to cars or motorbikes and always needed welding here and there as you do, im now on a land rover 300tdi project and got alot of rust as usual and need/want a safe sound strong landy, but welding prices are silly, ive got a good garage who does cheaper cash jobs as family friend but still, last chassis weld was 200 pound and got thinking i could buy a welder and do it my self alot cheaper as hasnt got to be the prettyest welds as its underneath and not the body but needs to be strong and safe, there welds are strong and good but not the tidyest i feel i could do as good, as doesnt take me long to get used to things hard, ill give it all my time and heart as i love my projects, and it will help in next projects loads , so how will i actualy no the welds will hold as it will be in the back of my mind are they strong enough ?? help please

 

[shenion]

Run some test welds then put them in a vice. Beat the crap out of them (bend in half) to see if the weld fails before the metal does.

 

[grumpygramps-2007]

thats the same hi-tech test i was told to do.

 

[cumbriasteve]

A 'destruction' test is the technical term or as shenion states 'beat the crap out of them' which is a much clearer term I prefer...but it's actually a hell of a good way to get an idea of your welds.
I was so chuffed when one of my first tests resulted in the metal tearing before the weld did, gives you the confidence you need.

 

[darran sandwich]

Sorry to hijack the thread a little but with this test, how can you be sure the failure isnt premature because of rapid cooling in the HAZ around the weld? or is the fact that it is allowed to air cool and not a quench enough to negate this risk?

 

[keith19]

Rapid cooling of the heat affected zone with m/s using mig isn't likely to be a problem. Some pro welders even use wet rags and cold air guns to cool the job quickly. Youtube is full of them.
I'm never in that much hurry myself, so I leave them to air cool.

 

[darran sandwich]

Oh right thats good news, is that true on all thicknessess of steel then?

 

[hotrodder]

Sorry to hijack the thread a little but with this test, how can you be sure the failure isnt premature because of rapid cooling in the HAZ around the weld? or is the fact that it is allowed to air cool and not a quench enough to negate this risk?

You can't but generalizing it's not something that'll be an issue with mild steel and on materials where it is an issue the damage will more likely be done by say forgetting/ignoring preheat than getting bored waiting for it cool

While the BFH test is better than nothing it doesn't tell you much. A 'nick break' is more useful imo and easy to do at home. Basically you score the weld metal with a hacksaw before reaching for the BFH, the idea being that the 'nick' guides the fracture. The sample should break through the weld metal (if it doesn't then you have issues) and you then get a peak at the inside of the weld.

Bend tests are another that can be done at home with little/improvised kit. More detail here... http://www.twi.co.uk/content/jk73.html and from this site... http://www.mig-welding.co.uk/wiki/Weld_Testing A mickey mouse version of macro etching can be done with acid from say a car battery. It's nowhere near as clear as the results gotten from the proper reagents but will show the fusion zone

 

[darran sandwich]

This was my thought really pre heat is effectively just increasing your cooling time by reducing the temperature differential around the weld. I guess medium and low carbon steels dont have much carbon ti migrate anyway so thats probably where the risk is mitigated.

 

[outofthefire]

At college we used to cut a section through the weld with a hacksaw polish it up with a file and emery cloth then put a bit of nitric acid on the weld it would etch it and you could see the penetration it also show any inclusions, after that we tested it like hotrodder describes. One thing that I learnt was nice looking welds dont necessarily mean there good welds.

 

[Paul.]

A very simple test used on sheet metal welds is the cupping test, it used to be quite a common C&G sheet metal welding test but isn't used much nowadays, all you need is a ball pein hammer and a piece of tube, or even an old socket. this is just a couple of gash pieces of 16swg, butt welded with the tig, for mig I would grind off most of the penetration and the reinforcement. it stresses the weld and the heat affected zone in all directions
the face

the penetration

Dimpled

That'll pass