This is not a “complete” article about making a knife from scratch. The
most important task in making such a knife is correctly heat-treating it. Unless
it is correctly heat treated, the blade will be either too soft to hold an edge
or so brittle that it will break if dropped on concrete (yes, I have done that).
Unfortunately, I have heard and read a lot of half-truths and misinformation
about heat-treating even though it is really not that difficult. So,
heat-treating steel from start to finish is what this article is about.
I would say that virtually all gunowners use knives (other than in the
kitchen). Most gunowners have probably thought of making their own field knife
-- someday. A few have actually done so, with varying degrees of success.
Although ready made finished blades (without handles) are being sold now, most
of the people I have talked with who have actually made their own knife started
with a raw chunk of steel; a large file, heavy saw-blade, or leaf-spring as the
raw material. All three of these materials have enough carbon and other alloying
elements to make a good knife. Large, old, rusty files sold for very little at
garage sales are my personal favorite as a starting point.
I will not go into detail on shaping the blade, threading the tang/nut (or
riveting to a full-size tang), fitting a handle, guard, or a pommel. There are
plenty of other articles on that. This article will only describe how an
ordinary person with little experience can correctly heat-treat a blade. If you
start with the raw material mentioned above, there are three stages to
heat-treating it. All three are needed. They are as follows.
1. Anneal (soften or stress relieve) the raw material
first, before any shaping is done. The part must be heated to just above the
critical temperature (from 1,350 to 1,550 degrees Fahrenheit for most carbon
steels) and allowed to slowly cool. The critical temperature is where the
structure of the steel crystal changes. It is a little different when heating-up
vs cooling-down, but that is not really important here. The exact temperature
depends on the amount of carbon and other alloying elements, which is something
you don’t know. You also cannot see when the crystalline structure changes.
What can you do? Well, you can see color and that is roughly related to
temperature. This is not the best way to find the critical temperature even
though a lot of people use it that way.
The critical temperature is where the color looks dull-cherry-red to
cherry-red in dim light. In normal (not bright) indoor light, this same
temperature is dark-blood-red to dull-cherry-red (darker than in dim light). I
don’t believe that anyone can accurately estimate temperature by color in
bright sunlight. By the time you see something in bright sunlight, you have
already overheated the blade.
If you see bright-cherry-red, orange-red, orange, yellow, white, or if the piece
starts to form scale, you have overheated the steel. It is damaged by then.
Since you don’t know how much it is damaged, it is best to throw it out and
try again (don’t worry, old files are cheap). Unfortunately, even in low
light, accurately judging temperature by color takes a LOT of experience. More
experience than the occasional knife maker will ever have time to develop.
Remember what happened to the “low number” 03 Springfields. They were
hardened by color -- and hardened wrong.
Fortunately, there is a better way. A magnetic steel part will become
NON-magnetic at the critical temperature. This works for all carbon steels, but
not on some stainless steels or other exotic metals. You are on your own if you
don’t use carbon steel. Heat up the raw material with propane torches (two
torches -- one on each side of the blade -- work better than one) and test it
periodically with a magnet (the kind that does not melt) on a handle. Test it in
more than one place to make sure the entire piece is above the critical
temperature. Naturally, two sets of hands makes this a lot easier than trying to
do it alone. Once the magnet is no longer attracted to the steel, DO NOT
continue heating the part. Also, DO NOT “soak” the part in the heat at this
temperature any longer than necessary.
The maximum strength/hardness that can be obtained will be less the more the
part is heated above the critical temperature OR the longer it is soaked above
the critical temperature. Slowly cool the hot part over many hours until cold.
Placing a heated part in a charcoal grill, covering it with hot coals (not
ashes) and letting both cool is slow enough. Figure on at least overnight for
cooling. Obviously, starting the charcoal fire is the first step in this
process. When the hot coals are ready, heating the part to its critical
temperature takes very little time.
The part should be relatively soft and very ductile at this point. You can
easily cut, bend, grind, shape, thread, drill, or polish the part in this state.
This is when most of the work of forming the blade should be done. For me, most
of the shaping is done with a bench grinder. A belt sander is sometimes used.
Fine work is done with a hand file.
2. Quench (harden) the part after it is almost completely
formed. Heat the part to slightly above the critical temperature (see above) and
quickly cool it in a liquid “bath”. The faster it is cooled, the harder and
more brittle it will be. A chilled brine-water bath will give a harder blade
than a room temperature water bath, which will give a harder blade than a room
temperature oil bath, which will give a harder blade than a heated oil bath. If
brine is used, make it from non-iodized salt mixed in distilled water. The
higher the carbon content in the knife, the warmer the bath should be. A file
generally uses room-temperature oil. Cooking oil smells better than automotive
oil, but both will catch fire and both will form a black, burnt crust on the
blade. This is not a problem. You can clean it after tempering.
Submerge the hot knife blade vertically (cutting edge first) in the bath so that
it cools equally on each side. Hold the part still while cooling. If you swish
it, it will probably warp too badly to use. The spine (back) of a single-edge
blade can be left out of the bath (or barely covered) to increase its ductility
without sacrificing the cutting edge. If the part warps, anneal it and start
over. If it cracks, throw it out and start over. The next time you try
quenching, use a warmer bath. The part will be very hard and VERY brittle at
this point. It is hard throughout, not just a surface hardness. Do not work the
part (cut, grind, bend, or even drop it) in this state because it will break and
it is a B*TCH to work on anyway. And, whatever you do, DO NOT try to reduce any
warpage by bending it in this state. If the warpage is minor, keep going.
3. Temper (draw) the part next. THIS IS VERY IMPORTANT, but
it is often skipped by do-it-yourselfers. This is particularly important if you
start with a large file, heavy saw-blade, or a leaf-spring. They usually end up
more brittle after quenching than what is best for a knife blade.
Immediately after quenching the part (best results are when the part is still
warm), put it in a preheated oven (on “bake”, not “broil”) at 400-450
degrees F. for one to three hours. If you quenched it in motor oil, you may have
a problem with the wife (because of the smell -- another reason to use cooking
oil for quenching). The larger the part, the longer you heat it. Lightly wrap it
with aluminum foil to reduce hot or cold spots (relatively speaking) and put it
in the oven. Soak the part in the heat until it is hot throughout. Then shut off
the oven and let it cool slowly. The part should lose only a little of its
strength, but will greatly increase its ductility.
A longer soak time and/or higher temperature will decrease the strength even
more, but further increase the ductility. This is the end of the heat treating
process. Minor working (such as finishing, sharpening or polishing) can be done
in this state. Sometimes, minor warpage introduced above can be straightened
(particularly if it is in the tang area) at this time. Naturally, fitting the
handle is done after you are done heat-treating.
Miscellaneous Notes:
Strength and ductility are inversely proportional, but both are needed in a
knife. The goal is to get as much strength as possible, while giving up as
little of it as possible in order obtain enough ductility to keep the knife from
breaking (snapping) easily. You can vary the amount of strength AND ductility by
varying the quench and/or the tempering.
Very low carbon steel cannot be hardened regardless of what you do with heat. It
is soft and will remain soft. That is why it is best to start with a file, heavy
saw blade, or leaf-spring. We know from the start that they have enough carbon
(and other alloying agents) to be heat-treated. I would not bother with any low
carbon steel myself, but if you do, it can only be case hardened. Not very good
for a knife.
I am not sold on soldering the guard to a tang. Heating the blade enough to
solder it will change what you worked so hard to do -- correctly heat-treat the
knife blade for maximum strength and ductility. Cut the guard right the first
time or throw it out and try again. It is a good learning experience. You can
use epoxy to fill and seal small gaps, but the joint strength should come from
the fitting.
If the part is NOT heated to above its critical temperature, the
annealing/hardening reactions will NOT take place. If this happens, the part can
be reheated again to a higher temperature without damage. However, the part
should be annealed before attempting to reharden it.
If the part is overheated, even slightly, it will never reach the strength or
hardness of a correctly heated part. The more it is overheated, the less strong
it will be. If it is overheated enough to burn out the carbon, it can never be
hardened again and is worthless.
Normalizing is a word that is bandied about a lot. It is similar to annealing,
but not the same. A normalized part is air-cooled in still air after being
heated to the critical temperature. It is not as soft or ductile as a properly
annealed part, but it is close. And, it is much quicker. This is used in place
of annealing when production volume is important.
When checking for the critical temperature with a magnet, make sure that the
magnet is not attracted to the tongs holding the part.
Ok, now that you have read about it, try it. It is fun.
