SawStop Braking Mechanism

[Sideways]

This is an opportunist thread that I'm writing in case it's of interest to members.
It comes about because, purely by accident, @deema and myself happened to see the SawStop stand at a trade show a few days ago. We got to chat with the MD and we observed the "sausage" demo.

As part of the demo, SawStop Europe MD, Michael, swapped out the used braking cartridge and blade for new ones, and offered the "dead" items to anyone in the audience who wanted them. I jumped at the chance so that you can see some of what we saw.

The disclaimers that come with this are :
I don't have a sawstop saw. I probably never will as I use a 750kg 1980's SCM industrial tablesaw that I refurbished and modified myself. Sawstop don't make saws like this and I could never afford a comparable Felder, Altendorf, whatever to replace it even without the added safety tech.
I absolutely believe in training, good practice, push sticks, etc etc as first line of defense. Second line of defense is behavioural safety - training and self awareness.
BUT
For those who can afford the extra line of defence that active safety technologies can provide, I'd commend them to anyone.

Here are a whole bunch of photos and some observations around that blade and brake that I brought home with me from the demo.

First, here's the blade and brake as removed from the saw.

The blade's the standard supplied 24 tooth 10" one. It's a relatively cheap blade. Expansion slots are there but nothing sophisticated. Carbide teeth are small. 3mm kerf. Plate isn't tensioned. This is a site saw blade. You wouldn't bother resharpening this, you'd replace it when worn.

The brake cartridge is the blue plastic housing, electronics inside, and the sacrificial lump of aluminium.
I believe the replacement cost of the brake is £80 something. If you send an used brake cartridge back and SawStop judge it to have been used to prevent an actual injury, they say they'll replace it free of charge.

The activated brake had three blade teeth embedded in the aluminium block. They were well stuck, I couldn't separate them by hand. I did hold the blade and simply hammered on the end on the aluminium block in the "reverse" direction and after a few blows, the blade came free. Here are the three teeth seen from both sides. The tooth with the most obvious marks and one either side. The carbide teeth didn't seem that worse for wear and I would have kept that blade for further use if it were mine, at least for rough work.

Replacing the blade and cartridge need only the tools that come with the saw and took perhaps 3 or 4 minutes, though this "depends". If you were unlucky, it might take you a while longer.
The brake is held in place within the saw by what looks like a special pin. This has to be unlocked (by hand) and pulled out first.
The nut is removed from the saw arbor using a spanner as normal prior to removing the blade.
The blade and the brake then have to be slid off the saw arbor and a shaft that anchors the brake cartridge. With parts to slide off two parallel shafts, the fit was tight after the cartridge fired and some prying was needed. It took Michael a couple of minutes. Once it came free, fitting a new blade and cartridge takes under a minute.

If you look at the first photo, it appears that once the aluminium block engages the saw teeth, only two further teeth, 1/12 of a revolution enter the block before the blade is brought to a dead stop. It's a 4000 rpm saw, so that's 60 sec / (4000x12) or one and a quarter milliseconds. You'd have to add to this the time needed for the electronics to react and that big spring to launch the aluminium brake into the spinning blade, but it's still pretty quick.

During the demo, there was a substantial "bang" when the brake triggered.

The alloy block is obviously designed to deform as well as be cut in order to absorb the energy of the spinning blade.

The brake block is bent a little sideways too.

 

[Sideways]

How does it work

Words to come - it's past midnight here

This small spring steel tongue passes down the middle of the compressed spring and sits under tension. Something threaded through that rounded slot serves as the trigger. When the trigger releases, the spring can extend and it pushes the alloy block into the spinning blade.

The compressed spring sits inside the tube on the right below, with the spring steel tongue extending up the middle.

Note the black plastic piece with the fine wires bottom R of the photo above. This is a key part of the trigger mechanism. Before the brake fired, the fine wires formed an unbroken loop.

Note also the section of thick wire that the arrow points at in the photo below.
There are actually two of these thick wires a few mm apart and before the brake is fired, the fine wire loop crosses them both. You can see a groove moulded in the plastic that will keep the fine wire in position acros the thick ones.
Connecting the charged capacitor bank across the two thick wires means that the thin one crossing them suddenly turns into a fuse shorting them out, and the "fuse" wire melts and breaks. The electronic sensing and microprocessor control decide when to fire the brake and they connect the capacitors to the fuse using a high power electronic switch.

 

[Sideways]

Electronics

The brake cartridge contains the electronics that do the sensing and decide when to trigger the cartridge. The electronics in these cartridges have evolved over time. The current one has a copyright label on the pcb saying 2020. It's at least a double sided circuit board with a lot of components top and bottom, gold plating and a ST Micro implementation of the ARM Cortex M7 microcontroller plus a flash memory chip at the heart of the circuitry.
There are photos on the web of an earlier "legacy" model cartridge and the 2020 version is distinctly more up to date.
Here are some views

This model of ARM Cortex microcontroller contains 64k + 256k of memory for any electronics nerds. There's also a winbond W25Q80DV flash memory device on the board for added storage. The flash memory chip (designated U13) is usually used to store a computer program but I don't know if sawstop are using it to store the program or data collected by the cartridge.

I surmise that four capacitors store energy which is needed to fire the brake. Older verisons of the brake used a single 2000 microfarad capacitor. The current one has 4x 680 microfarad caps giving 2720 combined.

In the photo below, the arrow points to heat marks where a surge of current has been used to melt a thin wire stretched taught across the short thick silver wire that runs horizontally. There

The square device labelled Q10 is an N channel MOSFET (a device for switching larger currents).

U14 is a "gate driver", a type of chip which is designed as a middle man interfacing the digital digital circuitry to high power transistors like MOSFETs (Q10).

As @Spectric mentions in the comments below, the cartridge has an old fashioned 9 pin D type plug on the side. This will be familiar to older computer buffs as a serial communication connection that was fitted to almost every PC made for a decade and more. Industrial electronics need to be physically robust and that's probably it's advantage here. The D connector is used to connect the electronics in the cartridge to the electronics in the saw itself which provide power and deal with the start stop bypass switching, the saw's status lights and the self test that ensures the brake is functional before they allow the saw to start in normal use.

 

[Spectric]

First time I have seen the good old 9 way D type for a while, used to be so common on serial interfaces. It looks fairly simple, just a latched spring until the detection circuit releases it to stop the blade. I would expect there must be some continous monitoring circuit that would indicate a failure, be interesting to see what components are inside !

 

[Sideways]

Absolutely.
I didn't get to see an unfired cartridge. Just overlooked that. But the latch must be that piece of black plastic with the two fine strands of wire you can see in one of the photos.
Originally those two strands would have been a complete loop, under tension and stretched tight across a pair of much thicker wires.
Discharge the capacitors to provide a pulse of energy to the two thick wires and the much thinner one is suddenly behaving like fuse wire shorting out a power supply. It melts, the little black piece is allowed to move, releasing the strap down the middle of the spring and the spring is freed to expand and drive the aluminium upward into the blade.

 

[Sideways]

First time I have seen the good old 9 way D type for a while

Yes. You can tell it's USA origins, my metric allen keys won't fit those socket screws to release the connector and let me pull the pcb free of the blue pastic shell. I'll have to ferret out the imperial set in the daylight then add some better photos of the pcb.

 

[Cabinetman]

Thanks for that, the mechanics are simple enough even for me to understand.
We’ve had the discussion so I’m not saying any more.

 

[OrangeFive]

This is an opportunist thread that I'm writing in case it's of interest to members.
It comes about because, purely by accident, @deema and myself happened to see the SawStop stand at a trade show a few days ago. We got to chat with the MD and we observed the "sausage" demo.

As part of the demo, SawStop Europe MD, Michael, swapped out the used braking cartridge and blade for new ones, and offered the "dead" items to anyone in the audience who wanted them. I jumped at the chance so that you can see some of what we saw.

The disclaimers that come with this are :
I don't have a sawstop saw. I probably never will as I use a 750kg 1980's SCM industrial tablesaw that I refurbished and modified myself. Sawstop don't make saws like this and I could never afford a comparable Felder, Altendorf, whatever to replace it even without the added safety tech.
I absolutely believe in training, good practice, push sticks, etc etc as first line of defense. Second line of defense is behavioural safety - training and self awareness.
BUT
For those who can afford the extra line of defence that active safety technologies can provide, I'd commend them to anyone.

Here are a whole bunch of photos and some observations around that blade and brake that I brought home with me from the demo.

First, here's the blade and brake as removed from the saw.

View attachment 198920

The blade's the standard supplied 24 tooth 10" one. It's a relatively cheap blade. Expansion slots are there but nothing sophisticated. Carbide teeth are small. 3mm kerf. Plate isn't tensioned. This is a site saw blade. You wouldn't bother resharpening this, you'd replace it when worn.

The brake cartridge is the blue plastic housing, electronics inside, and the sacrificial lump of aluminium.
I believe the replacement cost of the brake is £80 something. If you send an used brake cartridge back and SawStop judge it to have been used to prevent an actual injury, they say they'll replace it free of charge.

The activated brake had three blade teeth embedded in the aluminium block. They were well stuck, I couldn't separate them by hand. I did hold the blade and simply hammered on the end on the aluminium block in the "reverse" direction and after a few blows, the blade came free. Here are the three teeth seen from both sides. The tooth with the most obvious marks and one either side. The carbide teeth didn't seem that worse for wear and I would have kept that blade for further use if it were mine, at least for rough work.

View attachment 198922


View attachment 198921

Replacing the blade and cartridge need only the tools that come with the saw and took perhaps 3 or 4 minutes, though this "depends". If you were unlucky, it might take you a while longer.
The brake is held in place within the saw by what looks like a special pin. This has to be unlocked (by hand) and pulled out first.
The nut is removed from the saw arbor using a spanner as normal prior to removing the blade.
The blade and the brake then have to be slid off the saw arbor and a shaft that anchors the brake cartridge. With parts to slide off two parallel shafts, the fit was tight after the cartridge fired and some prying was needed. It took Michael a couple of minutes. Once it came free, fitting a new blade and cartridge takes under a minute.

If you look at the first photo, it appears that once the aluminium block engages the saw teeth, only two further teeth, 1/12 of a revolution enter the block before the blade is brought to a dead stop. It's a 4000 rpm saw, so that's 60 sec / (4000x12) or one and a quarter milliseconds. You'd have to add to this the time needed for the electronics to react and that big spring to launch the aluminium brake into the spinning blade, but it's still pretty quick.

During the demo, there was a substantial "bang" when the brake triggered.

The alloy block is obviously designed to deform as well as be cut in order to absorb the energy of the spinning blade.
View attachment 198924

View attachment 198923

The brake block is bent a little sideways too.
View attachment 198925

I have the video from the demo at Bolton but for the life of me can’t figure how to upload it. I can send it to you if you PM me if you wish.

Mod edit> very many thanks to @OrangeFive for getting this over to me.
I'm happy to insert his own video here.

As you see, Michael was pushing the timber and the "finger" into the blade at a realistic rate. He wasn't holding back in any way.

 

[Jacob]

The main failing of this project is that you have to buy the machine, and the mechanism is not applicable to any of the other machines you may be using.
Whereas normal safety procedures, 2 push sticks with variations etc; work on every table saw, spindle moulder, planer, router table etc, all highly effectively at very little cost.

 

[Spectric]

I wonder why they have made the entire system a bolt on, it looks to be very self contained with the brake and electronics all in a single unit which suggest it might have originated as something that could have been aimed at a more more generic market.

That processor is a 32 bit ARM Cortex with 64 pins yet is only 10mm square but packs quiet a punch with both floating point capability and a DSP with many peripherals so rather like cracking a nut with a sledge hammer and it makes you wonder what features the designer wanted that made this decision.

What thickness is that wire, it has to be strong enough to resist the spring and so would need a fair pulse of current to guarantee to melt in a very short time frame. You can guarantee that they have protected the code within the micro which is a shame as it would be interesting to take a look at exactly how this algorithm works.

 

[JSW]

The main failing of this project is that you have to buy the machine, and the mechanism is not applicable to any of the other machines you may be using.
Whereas normal safety procedures, 2 push sticks with variations etc; work on every table saw, spindle moulder, planer, router table etc, all highly effectively at very little cost.

I take your point, but it could be argued that safety belts/airbags in cars are designed with one purpose in mind, protect the user, same as the SawStop tech.
Difference being there are no push sticks/riving knives/guards etc for cars, no first line of defence.
It's a poor analogy,but you get my drift.

 

[Eshmiel]

The main failing of this project is that you have to buy the machine, and the mechanism is not applicable to any of the other machines you may be using.
Whereas normal safety procedures, 2 push sticks with variations etc; work on every table saw, spindle moulder, planer, router table etc, all highly effectively at very little cost.

When first invented, the inventor attempted to both license the mechanism and, a little later, to persuade the US government to make his mech mandatory on all US table saws. Other manufacturers wouldn't play as the cost of redesigning and retooling to take the Sawstop mech was too much, in their eyes. I believe there's still an ongoing attempt to get the US government to make the mech mandatory on all saws.

As another poster wonders, why is the mech integrated to the degree that it is? It may be that this is the least costly way to make it but it does seem to make it more difficult to install on saws not designed for it. And when a mech is destroyed when fired, all those expensive electronics become waste along with the aluminium "brake pad", with the user having to buy a whole rather expensive replacement.

The very good report in this thread of the details of device indicates that a blade may not always be made unusable by a triggering event, which would reduce the per-firing cost; but those cartridges are still expensive. And according to various users on forums, over the years, a blade often does get damaged to the point where it would at least need a few new tooth replacements, an uneconomic procedure since it tends to cost something similar to a new blade. And many blades are expensive - rather more than a basic Sawstop blade.

**********************
It can't be argued against, that this device is effective at preventing dafties from sawing off their fingers. It can and does prevent many horrible "accidents". But it seems an unnecessarily expensive thing to use in terms of replacements and also those false positives, the rate of which aren't really known but do seem to come up regularly on forums.

****************
Its a mistake to make the device part of an either/or proposal about TS safety. Both would be better for safety purposes. Not all users grasp safe procedures as they're never taught them and have a thousand mass media examples in the WW media showing unsafe procedures. For them, Sawstop makes sense - although it isn't going to encourage them to learn more about other modes of staying safe with a table saw. Why would they if they think a Sawstop will do it all?

But how many can afford to ditch their present saw and to buy a Sawstop replacement, especially if Sawstop becomes de rigeure, by fashion or law, making the worth of an old machine not-much?

*******************
I'd be more positive about Sawstop if they redesigned their mech to de-integrate (let's not say disintegrate) its functions to make repalcements much less expensive; made the "brake shoe" in a way that can't damage blade teeth; and did more to prevent false positives. If they could also put in an Altendorf-style traffic light system to prevent events rather than just prevent consequences of events, that would be a good way to educate users too.

 

[recipio]

Will the European saws have the standard 30mm European arbour or the 5/8 "extended arbour for a dado blade ?

 

[Sideways]

.... normal safety procedures, 2 push sticks with variations etc; work on every table saw, spindle moulder, planer, router table etc, all highly effectively at very little cost.

@Jacob I agree with this 110%
All those good things should come first and yes they are the most cost effective.
But industrial safety has a concept called IPL. Independent Protection Layers.
This is that safety is built up in layers, each with a cost and a benefit.

I'd argue that nothing makes a bigger difference to keeping your fingers intact than education and training about the machine and process.

Your two push sticks probably second to safe usage. The training in layer one should ensure that you don't even attempt to do the job unless you have the appropriate push sticks. Clear space to work, good balance, no trip hazards, physical and mental fitness all belong in here too.

Third the riving knife and crown guard. They can't be used on every possible type of cut but they should be wherever they apply. Again, training should stop you from attempting all but exceptional cuts without the machines safety features in place, yes ?

But accidents still happen even with all of the above. So what other layers can we add ?

PPE. Absolutely not going to stop us making mistakes, but reducing the severity of injury if something does go wrong and we get hit in the face.

Jigs and workholding to improve safety on cuts that we do often enough to justify the cost of making and using them, or on cuts that are innately more risky. An example would be a fritz and franz jig for cutting small pieces. We're adding costs now but gaining a marginal safety benefit that may be worthwhile.

Behavioural Safety training raises awareness of the general causes of accidents - not just machine specific - tries to sensitise us to conditions like fatigue, loss of concentration, distractions, so that we recognise them in ourselves and others, and stop before we hurt ourselves. It teaches us to learn from near misses and to recognise line of fire and the energy inherent in machinery giving an added appreciation of eg kickback risk. Behavioural safety training isn't free. It costs at least time. My last employer spend millions ensuring every employee in the business received about two full days of training over multiple sessions, but it absolutely improves safety.

Sawstop / Felder / Altendorf type active safety technology is just another Independent Protection Layer like everything above. It's probably the most expensive and if all the other things were done right, it should be the least used. But it does provide another layer of protection that didn't exist 20 years ago.

You and I are free not to buy one, and I would not want to be forced into buying one. We may not have space or money for whichever one would otherwise do what we need of it, but this thread is to provide some insight for anyone who is considering buying that extra layer of safety.

Cheers.

 

[Sideways]

Will the European saws have the standard 30mm European arbour or the 5/8 "extended arbour for a dado blade ?

I can't and don't speak for Sawstop. I'd never even seen one a week ago.
But the saw is being sold into the UK and continental europe so it has to conform to our regulations.
The MD said that something like 400 modifications were made to the USA version of the saw for europe.

The blade photo shows that the saw at the demo had a 30mm arbor. I'm pretty sure but could be wrong, that the arbor on the saw I watched wasn't long enough to take any sort of dado stack.

 

[Amplidyne]

Be more effective for the US government to ban about 75% of the videos I've seen online of people using woodworking machinery.
Because that's about the amount I reckon I've seen that are asking to have instant amputations of digits.

Personally, I prefer to use knowledge of the rejection forces and dangers inherent to using these machines, guarding, and correct procedures to remain safe, rather than relying on something that works after the "accident" has happened.

 

[Spectric]

but it could be argued that safety belts/airbags in cars

But you do find these in all modern cars, only if you have an old classic might you find no seatbelts and definately no air bag. I think the point here is would you be more or less cautious if climbing without a rope ?

 

[Spectric]

Third the riving knife and crown guard.

All aspects of the machines condition and safety should be covered by PUWER.

Also the outcome of any risk assessment is to comply with ALARP but just one person getting injured would not push the demand for safety measures upto the levels that would be required if the incident could cause far more widespread problems such as a chemical release so the probability of any regulations that make sawstop mandatory are slim.

The question I always ask is why do the Americans have such a track record for saw injuries, what do they do differently or is it a behavioral thing, maybe it is just because of the vast number of Americans using saws and the percentage of incidents is not that much different to europe.

 

[JSW]

But you do find these in all modern cars, only if you have an old classic might you find no seatbelts and definately no air bag. I think the point here is would you be more or less cautious if climbing without a rope ?

Neither of those examples stack up.
I wouldn't drive a car without a seat belt or airbag.
I wouldn't climb with or without a rope, I ain't 20 years old dude!

Plus you snipped my post so it makes no sense, tsk tsk

the vast number of Americans using saws and the percentage of incidents is not that much different to europe.

See how that works?

The way I see it, the SawStop tech/mech is an additional level of safety, in addition to the usual blahblahblah discussed already, it's designed for the home user/small workshop, NOT industrial use. IIRC @Inspector said he's had one for the last 20 years? and is happy with it, that will do nicely as a starting point.

 

[Sideways]

But @Spectric None of this is mandatory for DIY and one / two man band users.
Most users of site and even cabinet saws need to take safety onboard because they are persuaded that it's a good thing.