Lin,

I would use the same process as you. Can you tell us what method Jim uses?

Hello Lin. I watched you do it and I do mine basically the same way, but with a couple more steps. I check the grain size after: no thermal cycling, one thermal cycling, two cycles, and then the final just below critical cycle. You can see the difference in grain size pretty well. I like the picture, especially the sparkly one <img src=' http://www.americanbladesmith.com/ipboard/public/style_emoticons//laugh.gi f' class='bbc_emoticon' alt=':lol:' /> , good shot.

Brion

They way that Jim does it is.........

Draw a taper with a fine point on one bar , thermal cycle it, come up to heat for the quench letting the fine tip area get a little too hot and then quench. He starts at the tip breaking off little sections.

The tip sections have big grain getting finer as each section is broken off. His way may be a more accurate immitation of an actual blade. I may start doing it this way.

Thanks Steve and Brion. I'd still like to hear more if you have a better way.

Good Stuff Lin!

I show that to each student who comes through the shop, and it's always a hoot to watch their jaw drop and eyes pop.

I do basically the same as you, sometimes with a taper, and sometimes with just a 1/4"x1" piece of bar stock. It's interesting to me just how many makers really do not know or understand thermal cycling, and just how important it is to the finished product.

Thanks Ed. This is one reason I brought this up. I think sometimes students dont really believe the principles of grain reduction till they see it for themselves. Of course, with time, they should have these things born out through more serious testing as in the JS test.

We have started a Central Arkansas Knife Meeting on the last Thursday of the month and since the meeting is only a few months along we have some very new makers and some new to forging and doing their own heat treatment. This subject is important to ones like this. I am encouraging them at every meeting to schedule a class at one of the ABS Schools. Some have already.

Lin,

I got to see Dan Petersen demonstrate grain refinement just like Brion described at the ABS school and it was as Ed said, truly eye opening.

It seems to me that Mr. Crowells method not only demonstrates the importance of refining the grain but also the importance of getting an even heat on your forged blade.

It was also VERY eye opening getting to see 9 students bend their blades to failure and compare the grain structure versus performance during the bending. I bring this up because my broken test blade was like Mr. Crowell's example with fine grain in the spine but coarse grain in bottom 1/4" . For demostration purposes I can see where his method of overheating the thin portion could be benificial in showing the importance of not only proper normalizing but also the effect of overheating the thin portion during the hardening cycle as it pertains to heat treating an actual blade.

Happy Holidays, Josh

Yes Josh, heating evenly and having full control of the temperature is key to controlling grain size. The tapered cross section demonstrates the problem a bladesmith encounters trying to juggle the temperature versus timing to achieve even heating.

Your test blade that had the course grain in the cutting edge is an example of what happens when one is either not in control or when one does not KNOW he must be in control. That is the value of the class. Now you KNOW.

There are several factors that might influence a maker's success or failure in this regard. For one, his forge needs to be set hot enough, but not too hot. Heat treating using the forge as the heat source takes practice. On the last heat before the quench I will sometimes turn of the gas and air and let the ambient heat from the forge do the heating.

Hey Everyone,

Great post / info. I was just trying to remember this type of test so I could be sure to get a great blade. Quick question: I saw someone using an iron pipe in a coal forge to help ensure getting an even heat on a blade before quenching. Would it be beneficial to use this technique in a propane forge as well? It seems to me it might help. What do you think?

Ed C.

Ed, by all means. Every forge is a little different with it's own peculiarities and if your forge has trouble burning at safe(to the steel) temps, just add the pipe.

Heating a blade in a forge evenly can be a sort of dance. I heat the rear of the blade first and turn it around and do the in and out motion to get it up to heat evenly. All of this might mean for your situation that you have to have longer tongs, heavy gloves, etc.

For that reason, I have two forges. One for welding and one small one with a lot of control of the heat. I can turn it down pretty low and hold it there if I want. I posted pictures of it on one of the threads. I call it the Little Johnny Forge after it's builder Johnny Jones.

Thanks Lin! I've got the gloves, got the tongs, have an NC Tools Knifemaker two burner forge and just built a coal forge (needs a blower). Haven't experimented much with turning down the gas yet but that's a great idea, as well as starting with the rear of the blade <img src=' http://www.americanbladesmith.com/ipboard/public/style_emoticons//wink.gi f' class='bbc_emoticon' alt=';)' />

I think it would be fun some time to do a variation of this test with different variables. Overheat the steel then treat it well for a while then quench properly. Overheat the steel then normalize once then quench properly. Overheat the steel then normalize multiple times then quench properly. (I'm sure someone has done it, but I would like to see it.)

The way I saw it at the school in Arkansas was a piece of flatstock was overheated and quenched. The a corner was broken off at a 45 degree angle, showing the coarse grain. Then it was normalized three times, heated to critical and quenched again. Then the other edge was broken at 45 degrees, making a point, and it had finer grain. This was done again with three cycles and broken off the backside making a little triangle piece of metal that showed the grain growth. You had three sides on one piece of metal with three different grain structures. I thought it was one of the best demonstrations I had seen on what overheating and cycling does to the steel. The down side is it took all day to do. We were shown a little bit a time, then got back to work. After a while when the heat cycles had been completed we were all called around to see the next step. It's not something that can be shown to everyone quickly.

|quoted:

The way I saw it at the school in Arkansas was a piece of flatstock was overheated and quenched. The a corner was broken off at a 45 degree angle, showing the coarse grain. Then it was normalized three times, heated to critical and quenched again. Then the other edge was broken at 45 degrees, making a point, and it had finer grain. This was done again with three cycles and broken off the backside making a little triangle piece of metal that showed the grain growth. You had three sides on one piece of metal with three different grain structures. I thought it was one of the best demonstrations I had seen on what overheating and cycling does to the steel. The down side is it took all day to do. We were shown a little bit a time, then got back to work. After a while when the heat cycles had been completed we were all called around to see the next step. It's not something that can be shown to everyone quickly.

That sounds like a great demo!

This is a great subject Lin and a picture is worth a thousand words.

I use methods as described here, but would like to share an experiment I tried many many years ago. I way overheated the bar and quenched. When I tried to break the end of the bar off over the edge of the anvil, the bar exploded into 7 pieces. Moral of the story -- be careful when breaking hardened bars of steel.

Another important point is to actually do this experiment with your equipment. This is something that is so important that you must know what is going on in your shop with your methods and your equipment. Because there are so many variables, you cannot assume that you are getting the same results as someone else -- you must actually prove that you are getting the expected results.

Brian