metal fatigue

[JNewman]

Here is a question for the metalurgically inclined. Will a normalizing cycle eliminiate or reduce the effects of metal fatigue? I often break the spring on spring swages that I use frequently or a lot for a particular job. The break is often right beside the weld, rarely in the weld unless the spring has been repaired a number of times and a big section is nothing but weld. I have found running a couple of short beads of weld along the length of the spring where it attaches so there is less of a sharp transition helps. Normalizing near the weld after welding often helps as well. But today after breaking the same spring for the third time on this job I wonderered if I took a torch to the stressed area of the spring and warmed it up to aroung 1500 every now and again if it might eliminate the breakage?

I can see this job being reasonably hard on the spring as I am swaging 1 1/8" tool steel to 3/4" and I am taking a smallish bite so the swage is getting a twist as well but to break it 3 times with only about 90 done is getting frustrating, if I break it again tomorrow I will be replacing the spring. Maybe a longer one will help.

[781]

At the BAM conference last month Bob Patrick gave the answer to stop this from happening but for the life of me I cant remember the answer

[pkrankow]

I suspect that normalizing after welding will help as it will even out the heat affected zone, but normalizing after a period of use would only help if you believe work hardening is occurring. A longer spring with a loop in it to make it act softer is where I think you are going. (or a hinge.)

1500F seems hot for removing work hardening, but I don't know what steel you are using.

Phil

[jeremy k]

I would move the spring towards the "little bite" side more so that you dont get as much of a dramatic "twist" during use. Also the swage top and bottom can be ground off on the "less bite" side to make it swage more evenly were the preasure points are or need to be, I mean. - JK

[bigfootnampa]

It sounds like you really use your spring swages and therefore I think you'd be better off eliminating the welds altogether. As J. W. Lillico describes in his wonderful book "Blacksmith's Manual Illustrated" you can use 5/8" rod for the spring and drill 3/4" holes in your die blocks then upset the spring rod to 3/4 on the ends... after bending the spring it is inserted into the 3/4" holes and center punched all around to tighten the spring in the die blocks. This is one of many great ideas in this succinct little book!

[Timothy Miller]

I have stopped using welded spring swages almost all together for this very reason. I now if the tool will be used a lot forge out a tang on the side on the swage and drill two holes in it. I then rivet on the spring. I found this failed too because the soft iron rivets wore after a wile but it still lasted longer than welding. I did once forge rivets from 4140 and these held up but it was a pain to do. My next step was to forge out the spring out the side of each swage blank then harden and temper the two half's. I use old fork lift forks for my stock seems to be almost the ideal steel for this job and cheep too. The top and bottom swages with integral springs are rivited together well actually one rivet and one bolt. This allows the tool to be unbolted and swung apart for any dressing that it might need. I only go threw all of this work if I know I will be using this tool a lot or it is a high value part that I want to be sure the tool wont fail mid heat and ruin it. I tend to look long term with my tooling I intend to be still doing this in 25 years so I want my tools to last. I will try to remember to take a photo of the tool and my tool holder that uses a wedge to lock the tool in.

[JNewman]

It is only A36 or 44W, is fatigue caused by work hardening?

[JNewman]

I have stopped using welded spring swages almost all together for this very reason. I now if the tool will be used a lot forge out a tang on the side on the swage and drill two holes in it. I then rivet on the spring. I found this failed too because the soft iron rivets wore after a wile but it still lasted longer than welding. I did once forge rivets from 4140 and these held up but it was a pain to do. My next step was to forge out the spring out the side of each swage blank then harden and temper the two half's. I use old fork lift forks for my stock seems to be almost the ideal steel for this job and cheep too. The top and bottom swages with integral springs are rivited together well actually one rivet and one bolt. This allows the tool to be unbolted and swung apart for any dressing that it might need. I only go threw all of this work if I know I will be using this tool a lot or it is a high value part that I want to be sure the tool wont fail mid heat and ruin it. I tend to look long term with my tooling I intend to be still doing this in 25 years so I want my tools to last. I will try to remember to take a photo of the tool and my tool holder that uses a wedge to lock the tool in.

I will have to try this next time I make a high usage spring swage.

I got another 20 pcs. swaged today before it broke again. I only had 2 more to swage and they are just to go on the shelf. I am going to finish the other end and heat treat the ones I have swaged so I can complete the order. I only have an order for 80 and have swaged over 100 now. Once the job is done I will replace the spring. Thinking about it today I have used this tool for about 500 of these and 400 other tools that have 4" of 4140 swaged, I guess the spring steel is just shot.

[thingmaker3]

Work hardening and fatigue are not the same thing. Work hardening can and does change the fatige limit. Whether it increases or decreses the fatigue limit depends on how much work hardening and what kind of steel.

I'll re-read the relavent chapter of Dieter's book tonight & try to be a little more clear about this.

[Timothy Miller]

Here you go. The first pictuer is of a solid steel 7/8" spring swage forged from forklift fork. I heat treat as it as 4140. My die holder has a 1" square hole forged from solid and milled to receive a wedge. I welded a small plate over the milled slot to contain the wedge. This wedge or key locks in the shank nice tight so there is no slop or play. Before I used a set screw and I was constantly tightening it. The swage holder is a piece of 1" hot rolled with a small clamp on the top this clamp allows me to move the swage around on the die where I want it. it is held with grade 8 bolts and locked with double nuts. The the 1" bar is removable and can be positioned anywhere on the spring of the swage. The swage can also be used free hand or on the anvil. I can unbolt the two halfs and just use it as a bottom swage as well. The second is a swage with the spring riveted on. But the rivets tend to elongate over time and then fail. I personally think common hot rolled steel is not the appropriate material for this kind of application.

[thingmaker3]

First off, thanks for the opportunity to go back through Dieter's book!

Fatigue does not leave any changes to the structure. Work hardening does.

Fatigue takes place within the elastic limit of the material. Work hardening requires us to exceed the elastic limit. Permanent deformation is required for work hardening to take place.

Fatigue failure of mild steel requires 100,000 to 1,000,000 cycles at 40 or 50 ksi. Alloy and spring steels require tens of millions of cycles for fatigue failure.

Now let's talk about cracks. Microscopic cracks can form in the heat affected zone (HAZ) when welding. (Note the HAZ is not just the bead, but everything near it.) The more carbon in the steel, the more probable a crack will be present in the HAZ upon cooling. When we smack on something with a crack, the crack propagates (gets bigger) by some amount. The amount might be even more microscopic than the crack, but the crack does get bigger. And the bigger it gets, the faster it propagates.

Two ways to help prevent cracks in the HAZ: 1) preheat. 2) post-heat. Both are needed.

Less work to just use those nifty rivets the folk are talking about.

[basher]

welding spring steel will harden the spring rite next to the weld as you are getting it above critical and allowing the mass of the spring to cool it quick enough to harden , so tempering would help .
also making the spring parabolic (or at least tapered ) so that it shares its load along its length in proportion to its distance from the fulcrum will help spread the concentration of flex over the whole spring.
ideal is to not weld at all.

[JNewman]

Thanks guys. I do preheat the die before welding as it is always alloy steel usually either 4140 or 4340 but not the spring (mild steel). I guess I should preheat both from now on. I have had the springs break outside the HAZ but it normally is within. 100 000-1 000 000 cycles sounds like a lot but with the hammer hitting at 140 blows per minute and vibrating much faster between blows it does not take that long. Sounds like heat treating after use will not help. I will have to try Southshore's method out. I like the idea of being able to swing the top half off for dressing and to use as a bottom swage.

[monstermetal]

I will quantify this statement with the preface that I am not a metallurgist, I am a welder (at least thats the hat I need for this discussion) I dont think that what you have is metal fatigue but rather cross section rigidity working against you... Your swage body is big, your spring handle is small and a weld that is in the same plane as the spring action is a natural fulcrum and stress riser. Nothing you do to the weld is going to significantly increase the life span of the weld because its not the weld that is failing, it is the material next to the weld that is worked due to this stress riser... If you where to forge out a tang on the side of your swage body the same width as the handle and then overlap the handle and the swage and weld on both sides and NOT across the handle you would likely see a very big increase in life span... The other thing that might work is to start with handle material that is as wide as the swage body and then narrow it up for the spring and weld the swage inside the handle (so the die blocks where actually inside the handle.. Both of these are not as good as what Tim showed though... The main reason why his are better is the transitional mass from the die body to the handle that spreads the stress over a greater area, and that is the key to a tool that will will get a lot of twisting and hard use... there can not be a stress riser or it will fail regardless if its a rivet, a lap or a weld...

[Timothy Miller]

I should add that the spring on the first swage is hardened and tempered 4 hours @ 500 degrees in my kitchen oven and it is soft at the rivet joint.

[JNewman]

I will quantify this statement with the preface that I am not a metallurgist, I am a welder (at least thats the hat I need for this discussion) I dont think that what you have is metal fatigue but rather cross section rigidity working against you... Your swage body is big, your spring handle is small and a weld that is in the same plane as the spring action is a natural fulcrum and stress riser. Nothing you do to the weld is going to significantly increase the life span of the weld because its not the weld that is failing, it is the material next to the weld that is worked due to this stress riser... If you where to forge out a tang on the side of your swage body the same width as the handle and then overlap the handle and the swage and weld on both sides and NOT across the handle you would likely see a very big increase in life span... The other thing that might work is to start with handle material that is as wide as the swage body and then narrow it up for the spring and weld the swage inside the handle (so the die blocks where actually inside the handle.. Both of these are not as good as what Tim showed though... The main reason why his are better is the transitional mass from the die body to the handle that spreads the stress over a greater area, and that is the key to a tool that will will get a lot of twisting and hard use... there can not be a stress riser or it will fail regardless if its a rivet, a lap or a weld...

I think that you are right about the cross section rigidity, however I think that is what is causing the fatigue due to the concentration of the movement to a localized area.

[monstermetal]

Exactly what I mean, I think eliminating the weld across the spring somehow is the solution. Same with welding a post to a spring tool, you never weld across the handle, just in line with it and there will be 80% fewer failures

[Timothy Miller]

I can back this up with a visual aide. This is one spring tool I made. It is a double butcher for necking down stock to 3/4" round. I thought I should make the arms of spring extra heavy and move it away from the dies as it would last longer. It did last a bit longer but It broke right at the transition after about 500 pieces. I think this proves Larry right. So I guess you could weld on a spring with a gradual taper and that would work too. It would be perhaps less work than what I am doing. But I still like the idea of a tool that could be taken apart as it is more adaptable and can be put to other uses.

[pkrankow]

I can back this up with a visual aide. This is one spring tool I made. It is a double butcher for necking down stock to 3/4" round. I thought I should make the arms of spring extra heavy and move it away from the dies as it would last longer. It did last a bit longer but It broke right at the transition after about 500 pieces. I think this proves Larry right. So I guess you could weld on a spring with a gradual taper and that would work too. It would be perhaps less work than what I am doing. But I still like the idea of a tool that could be taken apart as it is more adaptable and can be put to other uses.

Any reason not to bolt a hinge in after the break is cut off and use a S spring like on a post vise?

Phil

[Nakedanvil - Grant Sarver]

Interesting design there Peter. I can see you've put a lot of thought into it. It appears in all designs the weak point is always right where the section changes. Pretty much exactly what Larry was talking about. Seems what we need to avoid is abrupt changes in section. Long tapers will spread the stress out over more length. I used to draw my springs out from the same piece as the swage. Still had the same problem at changes in section.

Stresses can be lowered some by making the top swage with a generous crown on top to avoid some of the "cocking" that happens when two flat surfaces come together at an angle.

[JNewman]

I think this tool was made worse by a handle I have on it to open it wide to fit the 1 1/8" into a 3/4" swage. The spring is doubled up about 1.5" from the swage. When I fix it I am going to run the opening handle right to the swage which should help to spread out the load.

[forgemaster]

Hey John

Swages breaking is a problem, that for us always seems to happen on the last job in a heat, I have no explanation for how this always happens I think it is some sort of VooDoo thing. the swages almost never break on the first job of a heat. We have found that modern steel is more prone to the breaking at the HAZ, then the older stuff. We used to have a grade called weldex which was a substitute for wrought applications, make handles out of this they would outlast anything, but they were I reckon closer to pure iron than modern steels that I feel are made to machine better. We have gone more to pin swages to get away from the handles problems. I think it shows us using pin swages on our Utube clip, (go, massey, 5cwt, blacksmith, chain anchors,). Normalising does not really have that much bearing on the outcome we have found, the welds running up the reins do make a difference though.

Hope this helps

Just checked, the clip shows Damo putting the swages in at about the 36 second mark. they are on a removable sqr peg which allows them to be turned around 90 deg when they are not in use rather than lifting them down off the table (that small table we call the coffee table)

Phil

[monstermetal]

Thanks Phil..

Here is the video he is talking about...

[Timothy Miller]

That's a great idea I love the pins but I think I would want more rigidity in how my swage was held on the bottom die. What I don't like is the tool moving around like that I have bruised my hands more than once with misaligned tooling under the hammer.
Tim

[JNewman]

Thanks Phil A lot of the tooling I got when I bought my Massey has pins and a handle about 3' long 5/8" or 3/4" dia handle on it for use with a helper. I have made some of my tooling with dowels as well especially stamps. The one problem I had with just cutting off the handle and welding on a shank was the top die bouncing off the pins, probably has to do with the weight of my foot, or maybe because my bottom die seems to be held a little more rigidly I don't know. I now weld a light spring on them and it works well, and helps keep them togetter for storage. I think the pins help with the dies not twisting as much and those springs last longer. Interesting to hear that you have come up with the beads along the spring as well.