Good way to compute the force of a hammer blow? What is a good and easy way to compute and/or measure the force of a hammer blow, not using any fancy or specialized equipment?
If the hammer is swung by hand through an arc, it is not obvious to me how to measure the speed the hammer will be when it strikes the metal.
Also, when the hammer strikes the metal, the heavier it is, the more force there will be persistent and the less rebound. Also, the smith may use what is called a "dead blow" hammer to reduce the rebound. Thus, measuring the force is not just a question of the instantaneous force of the hammer, but how much it "presses" after that first instant, its impetus so to speak.
Now, one idea I had was to use a teeter-totter. You could place a heavy weight on one end of the teeter-totter and then hit the other other with the hammer and see how far it moved. Of course, what will happen is that when the hammer hits the pad, the teeter totter will accelerate, reach a peak, then decelerate, and the profile of this curve of acceleration will be the measurement of the instantaneous force of the hammer over time. Perhaps this could be measured by an accelerometer, but it is hard to see how to make the measurement with no special instrument.
 A: Of course the force changes during the impact - so to get close to an answer, you need both the time of the impact and the magnitude of the momentum transfer.
As user77567 pointed out, a fairly simple way to measure momentum transfer is with a ballistic pendulum. This would be a heavy steel ball (much heavier than the hammer) hung from a long wire. When you strike the ball, the hammer will bounce back (since it is much lighter) and the ball will swing through an arc $\alpha$ before returning to the equilibrium position.
If ball has mass $M$ and hammer mass $m<<M$, then conservation of momentum tells us that
$$M\cdot v_{ball}= m\cdot \Delta v_{hammer}$$
For an elastic collision with $m<<M$, $\Delta v_{hammer} \approx 2 v_{initial}$
If the wire has length $\ell$ and moves through a distance $d$, so $\alpha = \tan^{-1} \frac{d}{\ell}$, conservation of energy tells us that for small deflections, the height $h$ that the ball rises after the impact is
$$h = \ell (1-\cos\alpha)\approx \frac{d^2}{2\ell}$$
Conservation of energy then tells us
$$M\cdot g\cdot h = \frac12 M v^2$$
and it follows that
$$v = d\sqrt{\frac{g}{\ell}}$$
Of course we could have got the same result directly from the equation of motion for a simple harmonic oscillator (pendulum).
The remaining interesting question is the impact time. This can be measured with simple electronic components. If you connect a resistor and a charge capacitor in parallel, with a "switch" formed by the contact between the hammer and the ball, then you can compute the impact time by observing the fraction of discharge of the capacitor due to the "closing of the switch" when the hammer hits the ball. Sufficiently thin and flexible wires should allow this measurement without disturbing the mechanics. Use a digital multimeter with sufficiently high impedance (at least 10 M). If the capacitor leaks slowly after you first charge it (say with a battery), you can observe the voltage dropping and hit the ball with the hammer just as the voltage hits a "round" value - this allows you to minimize the ffect of drift.
To make the measurement of impact time repeatable you could make the hammer part of a second pendulum that hits the ball from different heights: you can then plot the relationship between impact velocity and impact time, and this will allow you to get the time when you hit the ball really hard (when you might not get a good repeatable measurement of the time or velocity).
I hope this is enough to get you going...
A: I did some more research on this question and read some old books on the subject written back in the days when America still had blacksmiths.
What I found out is that there are two basic stages in understanding the problem. The first is to compute the energy of the hammer, which proportional to mass times velocity squared. The second is to compute distance over which the hammer acts on the item being hammered, so that the force can be computed.
For example, if a 5 pound hammer is swung at 50 feet per second, then the kinetic energy of the hammer is 5 * 50 * 50 / 64.32 = 194 foot pounds. If this then compresses a forging by 1/8 of an inch, then neglecting rebound, then it is about 194 * 96 or about 18,600 pounds of force on average. This is because 1/8" = 1/96th of a foot, and the average force is foot pounds / feet over which the force acts.
I found out that the way a hammer blow was tested in the old days is that two identical lead plugs would be made and one would be smashed with the hammer. Then the second would be put in a hydraulic (or screw) press and the press would be run until it crushed the second plug to the same degree. The force exerted by the press is then equal to the average force of the hammer blow. Since lead has very little rebound, that factor is taken out of the equation. The main problem with rebound is that it can conceal the amount of work truly done. For example, if a hammer bangs a piece of steel and it is crushed by 1/4 of an inch, but then springs back 1/8th of an inch, it is deceptive because it looks like the hammer only acted over 1/8th of an inch, when in fact it did work over the full 1/4. So, to compute things correctly, you have to figure out the full range of work and account for any of the metal that has sprung back into place.
So, basically the answer is: crush a lead plug.
A: 
... smith may use what is called a "dead blow" hammer to reduce the
  rebound. Thus, measuring the force is not just a question of the
  instantaneous force of the hammer, but how much it "presses" after
  that first instant, its impetus so to speak.
Now, one idea I had was to use a teeter-totter

They are not good solutions. The dead-blow hammer in particular does not help to solve the problem, but makes it worse, since the energy of the rebound is not visible, does not make damages , but is lost to heat etc through hysteresis, like in a sad ball.
The best solution is to use something like a ballistic pendulum: knowing the weight of the hammer and of the pendulum, gravity and the easy rules of collisions, you can get a very accurate measurement of the initial momentum measuring the displacement.
A: Perhaps the easiest way is to calculate torque and use the center of mass distance from the base (c_of_m_dist) in metres.  Torque =   c_of_m_dist x mass(N).  
An axe's  c_of_m_dist would be further out (below the steel head of the axe) and would have more torque. A knife's center of mass would be closer to the base, near the handle (less torque).
