Comment by ToValueFunfetti

No, I mean they had table saw rigs that would bring the saw up/down into the floor with an actuator as a 'ring hazard', ie, your robot could be subject to sawing at any moment if they happened to be there.

The question is, how fast does it need to be? Likely not that fast really, certainly not microseconds, and an actuator could easily yank the saw down without damaging it if it detected you were about to lose a finger.

There's also no reason you couldn't use the same actuator to do fancy things, like vary cut depth on the fly, or precisely set the cut depth in the first place. Can't do any of that with a soft aluminum pad that gets yeeted into the sawblade when it detects a problem.

Basically, SawStop exists to sell saws. Those saws happen to be safer, but that's a marketing point, it's not what ultimately makes them money. Look at the incentives, you'll find the truth.

An entry-level Dado blade can run about $100. The $10-12 sawblades can't make finish cuts that are worth a damn, because they chew through the work and tear splinters out rather than making precise nips at the front and back of each grain of the wood. For a saw blade an entry level blade that doesn't do this to your work can run you more like $60.

I know this because I've had to buy a table saw blade to replace a $10-12 one on my wife's table saw that someone threw on there because they were doing framing work.

How often do you use a saw? At $3500 a saw I care. I saw a lot of wood and inadvertently hit at least one staple/nail/screw per year. Over the last 20 years of using my saw that would be tens of thousands of dollars if even a portion of them damaged the saw. It would essentially price me out of doing woodworking.

SawStop works by detecting electrical conductance, and there are many reports of it misfiring when attempting to cut wood that isn't fully dry (i.e., there is moisture inside the wood, increasing its electrical conductance).

They are suggesting the blade retracted, broke the saw, in a situation in which there was no risk to the finger. Maybe there was a literally hotdog in the wood.

> If you're ripping a wet piece of wood, no thumb risk at all

It works on conduction and capacitance. It's not immune from false positives.

Frankly, the biggest problem is that this makes it impractical to test the brake. How do I know the brake even works, if testing it is not practical?

I do it so seldom and am so careful not to put my fingers within 3 inches of the blade that this is a non-issue for me. This is another one of those "let's put 6 extra buttons that all need to be pressed to start the saw!" kinda situations that doesn't do anything to improve safety because the stop is the first thing you disconnect if it throws a false positive.

If we're concerned about job site injuries then let's address the real problem, which is that a lot of people using these things do so as fast as humanly possible with little regard for set up, site safety, or body positioning because the amount of money they will lose by doing that eats so much margin out of their piecework that it's not worth it. As usual we don't want to solve the hard problem of reducing throughput to improve safety, but we're perfectly happy to throw a part that is as expensive as the sawblade on the unit just to say we're doing something.