Searching \ for '[TECH]:: Auto stop table saw - demo with real fing' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/mems.htm?key=table
Search entire site for: ':: Auto stop table saw - demo with real fing'.

Exact match. Not showing close matches.
PICList Thread
'[TECH]:: Auto stop table saw - demo with real fing'
2010\07\10@063357 by RussellMc

face picon face
Recommended viewing - even if you never use a table or other power saw.

Auto stop saw protection - blade is stopped in about 1 millisecond
after contact with a conductive object - typically a finger.

I've seen videos of this before but this is by far the best.
Shown it in action with a sausage and then in fast and slow motion
from various views with a real finger - attached to its inventor. He
said it tickled.

Also show the mechanism in action in slow motion so you can see how it
stops, how fast, what happens to the blade and the energy and more.

           http://www.youtube.com/watch_popup?v=E3mzhvMgrLE&NR=1

Approach speed doesn't want to be "too fast".
If stopping time is say 1 ms then a 1 m/s feed rate - probably at the
upper likely end of the speed range, would result in ~= 1 mm
penetration into the cutting zone after 1st contact. At say 1200 RPM =
20 rps a blade will rotate 20/1000 = 1/50th of a turn or about 7
degrees in 1/1000 second. Scale up for faster RPM. Blades have from
about 20 teeth up or from about 18 degrees per tooth or less, So
stopping COULD be under 1 tooth time but may be several teeth and
perhaps several mm into the "target" worst case (fast feed, fast
blade, many teeth). So you'd expect you MAY get a nasty nick worst
case, as long as the system works OK.

Blade and brake die at each actuation so you don't want to try this too often.


              Russell






                    Russell McMahon


__________________________

Reference from Iona.

2010\07\10@084555 by Carl Denk

flavicon
face
My fingers typically are not very conductive due to old thick skin, and
exposure to various solvents. I have measured the resistance fingers to
fingers on opposite sides recently, and it is very close to infinity. I
don't think I would want to trust it working. The stop is by jabbing a
piece of aluminum into the blade. There are parts including the blade to
be replaced after an event that as I understand, are not cheap. I still
have all my finger tips after 60+ years of use of similar equipment,
thanks to "Safety is Job ONE". Though at age 7, I was cutting stake
points on the circular saw that I still have today. One of the fall off
wedges hooked the back of the blade and threw it into my eye glasses. At
that time glasses were not the shatterproof, and the doctor had to pick
glass pieces out of my eye.

On 7/10/2010 6:33 AM, RussellMc wrote:
{Quote hidden}

2010\07\10@085906 by Michael Watterson

face picon face
 On 10/07/2010 11:33, RussellMc wrote:
> Recommended viewing - even if you never use a table or other power saw.
>
> Auto stop saw protection - blade is stopped in about 1 millisecond
> after contact with a conductive object - typically a finger.
>
>
So no use here where the wood could be more conductive. The rain today
is like something tropical. Been raining since sun.
May/June though was like some other countries Summer weather. I think
Poland got most of the rain we normally get in Ireland.

Still, an interesting idea.

2010\07\10@095138 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "RussellMc" <spam_OUTapptechnzTakeThisOuTspamgmail.com>
Sent: Saturday, July 10, 2010 11:33 AM
To: "Microcontroller discussion list - Public." <.....piclistKILLspamspam@spam@mit.edu>
Subject: [TECH]:: Auto stop table saw - demo with real finger

{Quote hidden}

Very impressive stuff. I am wondering whether it would be possible to save
the blade etc breaking by just retracting the blade very quickly instead.
Like Michael points out, it may not work well with wet wood, humid
environments etc and you wouldn't want to be replacing blades every five
minutes due to false alarms.





2010\07\10@115955 by RussellMc

face picon face
> Very impressive stuff. I am wondering whether it would be possible to save
> the blade etc breaking by just retracting the blade very quickly instead.
> Like Michael points out, it may not work well with wet wood, humid
> environments etc and you wouldn't want to be replacing blades every five
> minutes due to false alarms.

A good trick if you can manage it.
Somewhat akin to airbag action.
Blade needs to retract from immediate area at a rate well in advance
of feed rate.
I posited an upper feed rate of about 1 m/s or 1 mm/ms.
To bring the blade up to maximum feed rate  of 1 m/s in 1 ms requires
1000g acceleration.
Quicker or faster takes proportionally more g.
Very doable, but the forces and energy involved are significant and,
while you don't need to lock the teeth, you are still imparting
significant extra energy into the blade which then needs to be removed
again to bring the blade to rest.

The system may JUST kick the blade away from the user or away and then down.
There is a risk that the backwards motion may cause a hazard unless
adequate protection is provided.

Spring or explosive charge or compressed gas or a gas generator may all work.
Metal azide probably viable.

Blade kicked back and guard flicked over such that guard falls into
area where blade was "before finger can get there" should work.
(Grossly exceed maximum feed speed and you get whacked by the
descending (or rising) guard

You present a finger to the blade and suddenly there is a guard in
place a few mm back from the original contact point. Blade is behind
guard and decelerating fast under very heavy  traditional brakes.

Energy input rate is interesting [tm].
I keep getting answers re energy input rate that I appear to be
missing something.
Actual energy needed is not large as time to accelerate is small.


       Russell
.

2010\07\10@135414 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "RussellMc" <apptechnzspamKILLspamgmail.com>
Sent: Saturday, July 10, 2010 4:59 PM
To: "Microcontroller discussion list - Public." <.....piclistKILLspamspam.....mit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

> I posited an upper feed rate of about 1 m/s or 1 mm/ms.

I was wondering about these calculations. I don't know much about table saw
terminology but I assume you mean the blade edge travels 1 metre per second
or 1mm per millisecond?
The reason I ask is that in the video it says the blade is travelling at
5000RPM. By my calculations (which may be way out)
assuming a blade of 10in diameter = 31.4in circumference = 797mm. At 5000RPM
that's 83.3 turns per second which is 83.3 * 0.797 = 66.38m/s. So roughly
6.6cm in 1ms or about 30 degrees. Am I missing something here? is a standard
blade far smaller in diameter than 10in?
With those calculations it seems the saw would have to stop quite a bit
quicker than 1ms to avoid damage to finger.
He does say it stops quicker than 1ms in the video, but not how much
quicker.





2010\07\10@143425 by Olin Lathrop

face picon face
Oli Glaser wrote:
>> I posited an upper feed rate of about 1 m/s or 1 mm/ms.
>
> I was wondering about these calculations. I don't know much about
> table saw terminology but I assume you mean the blade edge travels 1
> metre per second or 1mm per millisecond?

No, feed rate refers to how fast the workpiece is moved with respect to the
machine.  In this case that would be how fast the wood and finger approach
the blade.  1 m/s is a really really high feed rate for a hand fed table
saw.  That would be quite reckless even if the saw could digest the wood
that fast.  I think Russell was trying to come up with a very pessimistic
upper bound.  I'd say typical feed rates are a few mm per second to a few
inches per second, certainly not 10s of inches per second, at least for a
home table saw the average basement woodworker will have.

A local lumber mill has some impressive circular saws that slice up whole
logs.  The feed rate might actually be around 1 m/s for those, but this is a
large industrial machine, the logs are automatically fed, and no human is
within 10 feet of the blades when they are cutting.

> The reason I ask is that in the video it says the blade is travelling
> at 5000RPM. By my calculations (which may be way out)
> assuming a blade of 10in diameter = 31.4in circumference = 797mm. At
> 5000RPM that's 83.3 turns per second which is 83.3 * 0.797 =
> 66.38m/s. So roughly
> 6.6cm in 1ms or about 30 degrees. Am I missing something here? is a
> standard blade far smaller in diameter than 10in?

No, that's about right.

> With those calculations it seems the saw would have to stop quite a
> bit quicker than 1ms to avoid damage to finger.

To avoid any damage, yes.  But the feed rate also matters.  Even if the saw
completely removes 1mm depth of flesh for the width of the blade, it's going
to hurt but your finger will heal and be OK.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\10@145333 by RussellMc

face picon face
Russell"
> >> I posited an upper feed rate of about 1 m/s or 1 mm/ms.

Oli:
> > I was wondering about these calculations. I don't know much about
> > table saw terminology but I assume you mean the blade edge travels 1
> > metre per second or 1mm per millisecond?

Olin:
> No, feed rate refers to how fast the workpiece is moved with respect to the
> machine.  In this case that would be how fast the wood and finger approach
> the blade.  1 m/s is a really really high feed rate for a hand fed table
> saw.  That would be quite reckless even if the saw could digest the wood
> that fast.  I think Russell was trying to come up with a very pessimistic
> upper bound.  I'd say typical feed rates are a few mm per second to a few
> inches per second, certainly not 10s of inches per second, at least for a
> home table saw the average basement woodworker will have.

Yes - I was referring to the rate that the finger may approach the saw at.
As Olin notes, 1 m/s is quite fast - certainly faster than you feed
anything except maybe light  wood with a firewood saw.
This sets an upper bound at which the mechanism needs to react at to
deal with almost any sensible situation.

With a 1m/s pop back rate you could ALMOST swing your hand against the
blade and have a reasonable chance of it saving you.
If you did this capacitively you *may* be able to trip it prior to contact.

I did post  some calculations re saw blade rotation and number of teeth etc.
I used 1200 RPM as a multiple of 60 seconds/minute and then noted that
you could scale this up as required.

Similar results to what Oli provided.

The "pop back" method seems liable to be workable - especially so if
you interpose a shield where the blade was so you don't have to go
back far or keep going back.


         Russell

2010\07\10@152255 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "Olin Lathrop" <EraseMEolin_piclistspam_OUTspamTakeThisOuTembedinc.com>
Sent: Saturday, July 10, 2010 7:34 PM
To: "Microcontroller discussion list - Public." <piclistspamspam_OUTmit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

> No, feed rate refers to how fast the workpiece is moved with respect to
> the
> machine.

I see, that makes more sense. I did consider that but to quote Russell:

>To bring the blade up to maximum feed rate  of 1 m/s in 1 ms requires
>1000g acceleration.

This confused me somewhat, and I assumed he must be talking of the blade
speed as it refers to the blade being brought up to "maximum feed rate",
rather than the material being cut.

Just saw Russells new post.
Capacitively sensing the finger sounds like a good method, harder to get
right but pretty ideal if done so. A 1m/s does seem like a very fast feed
rate (now that I know what it means :) ) so a pop back speed a bit above
this *should* provide adequate protection. In the video it looks like the
lock is placed so when the blade hits it its own momentum pulls it back
through the slot as well as topping it. I was thinking of something along
the lines of compressed gas to fire it back onto some shock absorbers.
What does is average blade made of? What do they weigh? Just wondering what
kind of forces may be required to achieve this.


2010\07\10@172109 by RussellMc

face picon face
> I was thinking of something along
> the lines of compressed gas to fire it back onto some shock absorbers.
> What does is average blade made of? What do they weigh? Just wondering what
> kind of forces may be required to achieve this.

F=mA

or you need the mass of the balde x the number of g acceleration.
Say the movable mass was 500 gram (~- 1 pound) and you wanted 1000 g you'd need
0.5 kg x 1000g = 500 kg or 5000 Newton.

I mentioned metal-azide "explosives" - as used in air bags.
Have the advantage of being fast. Which you need.

Airbag in car must travel from rest to chest in time it takes for
chest to travel say half distance from seat to wheel. Take it from
there and compare ...


             Russell



                     Russell

2010\07\11@021224 by Lucas Thompson

picon face
These saws have a spring loaded aluminum wedge shaped so that it gets
crushed when the safety is activated. Pretty sure that the "firing
mechanism" is simply a fuse-like wire that holds back the spring
tension. The firing cartridge has several capacitors. A friend of mine
has one.


On 7/10/10, RussellMc <@spam@apptechnzKILLspamspamgmail.com> wrote:
{Quote hidden}

> -

2010\07\11@031851 by RussellMc

face picon face
> These saws have a spring loaded aluminum wedge shaped so that it gets
> crushed when the safety is activated. Pretty sure that the "firing
> mechanism" is simply a fuse-like wire that holds back the spring
> tension. The firing cartridge has several capacitors. A friend of mine
> has one.

Yes.
Have you seen the video in my original post - extremely clear and
explanatory slow motion demonstration.

Oli and I were talking about a hypothetical system which moves the
whole blade "head" away from the drive to prevent damage. Acceleration
from rest would need to be in the 100 to 1000 g range depending on
assumptions made.

Adding to general ideas:
I can't yet quite 'put my finger on' how to meet all needs perfectly :-).
The triggering could be improved if the trip used an injected signal
passing between blade and person. TX/RX direction could be either way
with various pros and cons. A system will ideally be immune to eg wet
wood, not require any action by the user and be fail-safe (in the
formal sense), at least.  Quite effective system scan be envisaged
where user has to eg wear a transmitter unit or a receiver but in both
cases fail safing is problematic.

Capacitive coupling (as previously mentioned) and a very high
frequency oscillator may allow a "getting close" signal and "system
working" feedback.



             Russell




{Quote hidden}

> > --

2010\07\11@231548 by RussellMc

face picon face
> My fingers typically are not very conductive due to old thick skin, and
> exposure to various solvents. I have measured the resistance fingers to
> fingers on opposite sides recently, and it is very close to infinity. I
> don't think I would want to trust it working.

1. Partially capacitive.
2. Stick a needle into each finger and measure between them :-) -
resistance will be "far lower".

A "floating" body is enough to trip the device.

_______________

Material sent to a friend:

> Also surprised that nobody mentioned anything about the hand-soaking  - and that
> you didn't comment on my mentioning of it.

I had not heard anything re hand soaking and I've now read quite a lot
more and have found no mention of it. The SawStop literature suggests
that both conductive and capacitive sensing is used.

There is much on the web about the product. It sounds like it really
works well when needed and that the main issues are false trips
andcost - but that false trips are not usually substantial.

This is impressive:

SawStop informed me that the worst injury caused by a SawStop was when
someone lost their balance and in order to keep from falling on the
spinning blade had to push off the blade with their hand. This action
with normal table saws will produce an amputation, minimum. The
resulting injury, because the saw was a SawStop, required 5 stitches.

http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html



Home page

http://www.sawstop.com/

Detect waveform - successive teeth contacting finger can be seen in signal :-)
http://www.sawstop.com/howitworks/how_whitepaper.php

FAQ http://www.sawstop.com/howitworks/how_faq.php

Some real finger pictures (apparently)
http://www.sawstop.com/finger_saves.php

"White paper" - not overly detailed
http://www.sawstop.com/howitworks/sawstop_whitepaper.pdf

Discussion over licencing etc
http://medgadget.com/archives/2006/08/the_saga_of_the.html

Lumberjocks user discussion
http://lumberjocks.com/topics/4606

Woodweb OSHA discussion
http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html

Positive user review - he likes th saw's quality too
http://4renovators.com/blogs/pigeonpointproject/archive/2006/11/06/sawstop-part-2.aspx

Bosch sued for not incorporating SS type technology into a saw type
that SS themselves  don't cover.
http://www.protoolreviews.com/news/editorials/bosch-tools-sawstop-lawsuit


  R

2010\07\12@032221 by Vitaliy

face
flavicon
face
RussellMc wrote:
> http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html

I found this comment interesting:

"At the wood shop I am the safety manager at, we use the Brettguard system.
It is a fixed clear guard covering the entire blade that you crank up and
down, so you can get it to about 1/16 above the material being cut. It will
not raise up if you run your hand into it like some of the other guards. It
also helps hold the material down - works great with laminate. We had a
Sawstop as well, but sold it. It was an amazing product. However, OSHA does
not see the Sawstop as a guard, so you will still be fined for not having a
guard in place, since an injury has to take place for the system to work. "

OSHA's mission is to make workers safer, but in this case it looks like they
made them less safe because they followed the regulations, instead of common
sense.

Vitaliy

2010\07\12@063906 by RussellMc

face picon face
> RussellMc wrote:
>> http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html

Summary: OSHA not so dumb.

Challenge: Read the 3 "Which system would you rather ..." questions in
the text and answer them truthfully. Report back.

____

Vitaliy:
> I found this comment interesting

>> " > "At the wood shop I am the safety manager at, we use the
Brettguard system ...
>> ... However, OSHA does not see the Sawstop as a guard, so you will
>> still be fined for not having a guard in place"

Vitaliy:
> OSHA's mission is to make workers safer, but in this case it looks like they
> made them less safe because they followed the regulations, instead of common
> sense.

No.
Demonstrably not so.
It may seem that way at first glance, but it can be seen on a little
consideration that OSHA, in the instance in question, have a bit more
nouse than you are crediting them with.

Consider:

> "At the wood shop I am the safety manager at, we use the Brettguard system.
> It is a fixed clear guard covering the entire blade that you crank up and
> down, so you can get it to about 1/16 above the material being cut. It will
> not raise up if you run your hand into it like some of the other guards. It
> also helps hold the material down - works great with laminate.

So far so good.
He has a guard that works superbly, with which you CANNOT cut yourself
at all if you follow the shop rules (eg adjust to 1/16" above work
surface)(unless you have body parts < 1/16th" thick)(eg you could
probably cut long air with it but you'd probably have to try very very
very hard to do so)(tell me again, why WERE you lying on the saw
table?),

> We had a Sawstop as well, but sold it. It was an amazing product.

ie they concluded either that the Brettguard did everything that the
SawStop did that they needed OR they decided they were happy to be
less safe and to save some money. ie nobody made them sell the
SawStop. They could have used it as well as the BrettGuard. They chose
not to. Why we don't know, but there is good reason to think that the
Brettguard made it unnecessary.
Note that the BrettGuard is not cheap compared to alternatives. There
are many other guards that would meet OSHA specs, but they chose this
one. That suggests that they are serious about safety.

> However, OSHA does
> not see the Sawstop as a guard, so you will still be fined for not having a
> guard in place, since an injury has to take place for the system to work. "

OK. Two questions suggest themselves.

1. Are OSHA correct?
2. Are OSHA being reasonable?

Compare a Brettguard and a SawStop in use in a shop that tries to be
as safe as possible. Both systems are used according to manufacturers
suggestions and are adjusted correctly. Both have codes of practice in
place and these are enforced by a resident safety officer or similar.

Questions:

Which system would you rather offer a finger to?
Which system would you rather fall on with an outstretched hand?
Which system would you rather fall on if your body was going to be the
first point of impact just above the blade?

If you think that those are contrived and unfair questions - they are
and they aren't.
ie the situations apart from offering a finger are unusual. Nobody in
their right mind allows themselves to fall onto a saw blade in such a
way that they need to put their hand on the blade to save themselves -
let along fall on a bladed with their body? Do they?
Maybe not. But apparently there are people around who are NOT in their
right minds who use saws.
Quoting from somebody who says that they are quoting (favourably) a
statement by SawStop themselves:

    "SawStop informed me that the worst injury caused by a SawStop
was when someone lost their balance and in order to keep from falling
on the spinning blade had to push off the blade with their hand. This
action with normal table saws will produce an amputation, minimum. The
resulting injury, because the saw was a SawStop, required 5 stitches.

From:
http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html

Now, I consider that an  awesomely successful product. A rotary saw
blad that you can fall on with an outstretched hand, taking weight on
your arm to prevent body injury, and only needing 5 stitches in the
hand, is utterly awesome. It's a marvellous achievement of technology.
BUT, it is hard to imagine that if a Brettguard had been fitted that
the user would have needed ANY stitches at all. Nor needed an about
$100 brake and blade replacement (having read more that's more like
the cost). Nor needed an accident report, work stoppage, blade & brake
replacement (quick if you have both in stock), 5 stitches, emergency
room trip, OSHA intervention  ... . And, FWIW, the 5 stitches DO sound
amazing. But if that had been the users wrist and not hand that had
met the blade then the 5 stitch may have been a bit too late to help.

I've never seen a BrettGuard. In fact, I'd never heard of one until I
read that piece. But Amazon has:

    http://www.amazon.com/HTC-10A-L-Brett-Guard-Table-Guard/dp/B0001LQW7S

And here's what users think of them:

www.sawmillcreek.org/showthread.php?referrerid=5960&t=77274
Sounds good to me.

Note that a user suggests that most aftermarket "guards" are described
as "covers" by their makers.

But wait, there's more :-).

- The SawStop will sometimes mistrip -~ $100/time. Can be reduced by
proper setup and maintenance, but no guarantee of never happening.

- Point in favour: I have not seen anything suggesting that it ever
failed to trip as designed.

- There are some materials that it will not work with. You can check
these by putting it in bypass mode and seeing if it trips when offered
the material.If you don't mind the effort of disabling the system and
testing on every new piece of wood this MAY save your $100 events.

BUT the system CAN be bypassed and MUST be bypassed for some sorts of
use. A well designed guard system will be usable for a wide range of
work thicknesses and styles. There may be some conditions where it is
essentially necessary to convert to a "bare blade". BUT a SawStop
system that is bypassed ALWAYS converts to a fully unprotected
bar-blade system unless you eg add a guard as well. Then ... .

- Can't be used with non conductive blades or blades with non
conductive cutting inserts.

- Has a maximum blade size and mass that it will work with. Also a
minimum blade stiffness (for safety).

- MAY sometimes shed teeth on standard steel blades.

- Very unwise to use with tipped teeth (eg Tungsten carbide) even if
conductivity OK. Tungsten Carbide bullets may not be good for your
health. (A friend of mine managed to hit a fence wire with a TC blade
and he reports that it peeled the outer ring of teeth and  inserts off
the blade).

- SawStop reportedly are asking an
extortionate-in-the-extreme-by-any-usual-standards 8% royalties for
their system. That's fine as far as it goes in a
what-the-market-will-bear  open market system. BUT they reportedly are
seeking declaratory judgements (or some such) to have their maximally
safe system made mandatory on saws of this type.If both be true, and I
know not how true they be, then it seems an abuse of the WTMWB system
to use compulsory regulation to mandate your free market patented
device.

- Bosch are/were being sued by a user who hurt themselves with a Bosch
mitre saw because, they say, it did not have a SawStop type system
installed and could have had. But SawStop systems are not made for and
are not suitable for this sort of saw.

- Reportedly, the large majority of table saw injuries, some very bad
ones, are caused by the work being thrown back by the blade.
SawStop (presumably) does nothing to prevent this. I suspect that may
guard systems do or can prevent or reduce this happening.

That should do for now.


> OSHA's mission is to make workers safer

Does anyone think / still think that OSHA did the wrong thing by the
workers who they are mandated to act on behalf of?


                     Russell

2010\07\12@080842 by Jake Anderson

flavicon
face
On 12/07/10 20:38, RussellMc wrote:
>> RussellMc wrote:
>>      
>>> www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html
>>>        
> Summary: OSHA not so dumb.
>
> Challenge: Read the 3 "Which system would you rather ..." questions in
> the text and answer them truthfully. Report back.
>
> ____
>    
I think they clearly missed the point.
If you want to be safe use both systems.
The guard stops you from touching the blade (mostly), the sawstop stops
you from getting hurt when you touch the blade anyway.
"belt and suspenders"

2010\07\12@083551 by Carl Denk

flavicon
face
Concur. At my retired from employer, Ford (they were very big on
safety), there was a carpenter with a wood shop that included table saw,
joiner, radial arm saw, etc. To use some of that equipment in particular
instances with the full guards in place was next to impossible. When
OSHA went into effect, there was real battles between maintenance
(carpenter), and the plant safety people. But for the most part, OSHA
has helped prevent injuries, and when you read of injuries or
fatalities, it is usually as a result of not following the rules. The
common one is trench collapses during construction. OSHA says any
excavation deeper than 5 foot requires either protection (a trench box)
or bench (cut) back the sides in a specified manner. And several times a
year, locally, I hear of someone buried when a trench wall collapses.

On 7/12/2010 3:22 AM, Vitaliy wrote:
{Quote hidden}

2010\07\12@084936 by RussellMc

face picon face
>>>> http://www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html

>> Summary: OSHA not so dumb.

>> Challenge: Read the 3 "Which system would you rather ..." questions in
>> the text and answer them truthfully. Report back.

> I think they clearly missed the point.

The people using the saw MAY have.
OSHA didn't (IMHO,AE)

ie OSHA's mission is,  Vitaliy suggested, "to make workers safer".
I don't know what the formal wording is, but I suspect "safer" rather
than safe is a sensible word to use.
Nothing is "safe". Things can be "safe enough".
On the one hand we have pieces of "stuff" available from a wide range
of competing vendors in open market conditions.
The rule "If you have a $10 head buy a $10 helmet" almost applies BUT
OSHA set the lower safety limit. ie guards made from wet noodles don't
cut it (as the saw would).
Some minimum standard of bent steel will work.
Expensive useful super safe like the BrettGuard are available if you
want to afford them BUT they exceed OSHA's minimum standard. It would
be about impossible [tm] to but yourself with a BrettGuard in place
and used properly, and hard-enough [tm] with OSHA's minimum guard
which is no doubt much cheaper and somewhat  less effective.

WHEREAS the SawStop GUARANTEES that you will cut yourself on teh event
of an incident, it relies (very very probably) on electronics to work,
it's available from a single licenceable source, it costs much more
than an entry level OSHA acceptable piece of bent steel, and worst
case you could cut your wrist on the blade and die even when the
system is working completely properly.

The people who HAD to fit guards, and who fitted BrettGuards instead
of SawStop probably

- Made money in the deal
- Got a system which doesn't occasionally nuisance trip and cost them
$100, result in down time and/or , result in the need to carry spare
brakes and blades - in place of one which did all these things.
- Got a system which overall was safer - but which was generally
somewhat more annoying and occasionally much more annoying to use (a
no guard system is always easier than a guarded system).
- They MAY have a system which reduces job fling injuries compared to
a no guard system - and as most saw table injuries are from this
cause, they may have done better than expected.

Or not :-).

> If you want to be safe use both systems.
> The guard stops you from touching the blade (mostly), the sawstop stops you
> from getting hurt when you touch the blade anyway.
> "belt and suspenders"

Yes. If you want to afford it (see $10 helmet ...).
Note that in some cases a SawStop stop may be more dangerous or much
more dangerous. If using eg Tungsten Carbide tipped blades you may
wish to disable it. As long as you turn it back on subsequently ... .



Russell



,

2010\07\12@090056 by Alan B Pearce

face picon face
> >>>> www.woodweb.com/knowledge_base/Table_Saws_and_Safety.html
>
> >> Summary: OSHA not so dumb.
>
> >> Challenge: Read the 3 "Which system would you rather ..." questions
in
> >> the text and answer them truthfully. Report back.
>
> > I think they clearly missed the point.
>
> The people using the saw MAY have.
> OSHA didn't (IMHO,AE)

I would have to agree with Russell here. OSHAs remit is to avoid an
accident. Requiring a bandage around your finger (even without stitches)
around here is a notifiable accident (i.e. needs to go in the companies
accident log). So getting an injury from the saw stop is still an
accident, not an accident avoided. It is 'just' a potentially less
severe accident than may have otherwise happened. That is why the saw
guard will still be required.

Alan (who has a mangled hand due to incident with table saw).
--
Scanned by iCritical.

2010\07\12@093325 by Sean Breheny

face picon face
On Mon, Jul 12, 2010 at 9:01 AM,  <RemoveMEalan.b.pearceTakeThisOuTspamstfc.ac.uk> wrote:
> Alan (who has a mangled hand due to incident with table saw).

Wow, it hurts just to read that sentence. I guess that this was a long
time ago? Is it still fully usable? I hope so.

Sean


> --
> Scanned by iCritical.
>
> -

2010\07\12@102107 by RussellMc

face picon face
>> Alan (who has a mangled hand due to incident with table saw).

I had a friend (now dead) who was a joiner. He had 4 reasonably
functional fingertips with a diagonal line running across them at
increasing distance from the end of the fingers - where the saw blade
took them off and the surgeons successfully put them back on again. He
did it a long long time ago - I'm surprised how well the "repair"
worked. I'm sure he'd have been a SawStop customer if they'd been
available some decades earlier :-).

I think SawStop is a great idea - and one day (almost) all saws will
be built  SOMETHING like that. But not until they make one with  a
detection system which is truly preemptive and makes it truly injury
free in normal use AND costs zilch if it false trips.
I think I'd have a fair prospect of making something now which comes
close to all that. (Read prior posts for some of that).
SawStop are going to have to have extremely broad and insightful
patents if they want to keep asking 8% royalties.


              Russell

For following, all rights reserved.
:-) - what rights those are vary with administration. NZ = none. US =
much AFAIK.

Some mix of some of:

- Pop back blade to withdraw from user at a velocity which rapidly
exceeds some design feed rate.
- Cover/guard which shrouds blade - with blade having withdrawn far
enough that finger is not impacted by closing guard.
- Rapid but non damaging braking.
- Non contact proximity sensing. "Those skilled in the art" [tm]  will
easily envisage many methods by which this may be achieved.
 Examples only include capacitance, optical (& electromagnetic in
general)  (eg camera hand position tracking, beam blocking,
reflective, doppler, TOF ranging, ...) , acoustic (much of as for
light), etc.
- Adding a signal to machine or user to enhance operation both in
isolation and wrt conductive materials. eg em signal capacitively
coupled from user to blade system. Increasing frequency allows
decreasing capacity coupling requirement until substantial non contact
proximity viable.
- Glove / hand / finger piece or mark or ring etc fir highly positive
tracking. More applicable in professional systems allowing very great
freedom of movement relative to open blade etc in exchange for
discipline of utilising the system.
- Extension to many or any other systems - saws, rotating and linear
action machinery, presses, stampers, guillotines and anything that
moves. Non contact proximity vvv desirable in some of these  - eg
large press - ir pressing your fingertips is unwise AND a hand in the
air may be well inside "maw" of machine when closing action starts.
Active hand etc (limb, head, body ...)  tracking allows extremely good
control. Two or more systems can track and vote to give near failsafe
operation. This is getting well away from SS ideas but is even more
useful.






                        Russell

2010\07\12@105244 by Alan B Pearce

face picon face
> > Alan (who has a mangled hand due to incident with table saw).
>
> Wow, it hurts just to read that sentence. I guess that this was a long
> time ago? Is it still fully usable? I hope so.

Ah, yes, I was cutting some sheet aluminium on a table saw without the
riving knife or guard, and the piece of metal caught and kicked back.
All my own fault, but I was in a hurry to get something done!!

All five fingers on my left hand got 'touched' - the finger print shaved
off the thumb, index finger cut off at the base of the nail, and was
taken back to the joint in surgery, middle finger cut right through the
bone just on the outside of the top joint, but still had the inside
flesh connected, ring finger cut part way through the bone, little
finger had some saw tooth marks down the side of the nail. There was no
"look what is happening", it wasn't done then it was done ...

The middle finger had two pins through it to stabilise it while the bone
grew back, the ring finger had one pin. The ring finger will not
straighten on its own now, the middle finger will only bend the last
joint slightly when making a fist.

The piece cut off the index finger was found by the ambulance crew, and
kept in case it could be re-attached or the thumb needed a skin graft,
but in the end the skin graft wasn't required, and it was easier and
neater to take the ring finger back to the joint to get skin to wrap
around the end making a neat stub.

Luckily I am right handed, so it doesn't affect me too much, but
attempting to pick up items like electronic components or screws and
nuts off the bench with my left hand is very awkward as the index finger
doesn't have a nail and the thumb doesn't properly oppose the middle
finger because the middle finger doesn't bend enough.

The bit that always amused me was went I went back into work for a visit
while off on Accident Compensation (NZ has a wonderful compensation
system when you are an accident victim) and one of the salesman who I
always thought of as a pretty butch guy couldn't stand the side of my
hand with the pins sticking out of the end of my fingers. ;))



--
Scanned by iCritical.

2010\07\12@105718 by Alan B Pearce

face picon face
> I had a friend (now dead) who was a joiner. He had 4 reasonably
> functional fingertips with a diagonal line running across them at
> increasing distance from the end of the fingers - where the saw blade
> took them off and the surgeons successfully put them back on again. He
> did it a long long time ago - I'm surprised how well the "repair"
> worked. I'm sure he'd have been a SawStop customer if they'd been
> available some decades earlier :-).

Yeah, reminds me of a story told by an uncle of mine who owned a joiners
shop. When I was a small lad he had an employee who managed to cut a
finger off in a saw. One of his colleagues was nearby and asked 'how he
did that' ... 'oh like this' was the answer, to which injured employee
waved his hand at the saw again - and cut off another finger !!!

Moral of the story, turn the machine off after the first injury !!!
--
Scanned by iCritical.

2010\07\12@140429 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "RussellMc" <spamBeGoneapptechnzspamBeGonespamgmail.com>
Sent: Monday, July 12, 2010 3:20 PM
To: "Microcontroller discussion list - Public." <TakeThisOuTpiclistEraseMEspamspam_OUTmit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

> I think I'd have a fair prospect of making something now which comes
> close to all that. (Read prior posts for some of that).
> SawStop are going to have to have extremely broad and insightful
> patents if they want to keep asking 8% royalties.

Yes, I thinking the same :-)
That was why I asked about the standard weight of a blade etc, but you would
have to take the blade mounting into consideration too of course. Just
trying to do some rough calculations to work out what forces are needed for
decent acceleration/decceleration out of the way of finger in worst case
scenarios - does look like there will be a fair bit of energy required
quickly.

Also how to go about the capacitive sensing, ideally without user having to
wear anything. Or as you say below pretty thoroughly, maybe camera, optical
etc etc. Or some combination of. A system that is usable with other similar
machines (as you also mention below) would be great. Be interesting to make
a test model for proof of concept anyway, if I have time between various
projects I might have a little go at it, if for no other reason than it
interests me. Were you thinking of doing similar? Or just thinking out
loud..


{Quote hidden}

2010\07\12@142101 by Oli Glaser

flavicon
face
> and the piece of metal caught and kicked back.

It seems that most injuries are caused this way from what I have read. As
far as I know none of the safety systems currently available prevent this
from happening, although careful attention to detail and maintenance appears
to help. I wonder if there is a way to prevent this happening too, or
stopping injury when it does (I imagine it would be rather difficult given
the rather random nature of it, far harder than sensing fingers near the
blade) Some more research on the various factors involved in producing the
kickback would be needed. Maybe some way of detecting too much sudden
pressure on the blade would be an idea, but I wouldn't be too hopeful,
certainly without knowing a lot more about it all.


2010\07\12@143735 by RussellMc

face picon face
>Be interesting to make
> a test model for proof of concept anyway, if I have time between various
> projects I might have a little go at it, if for no other reason than it
> interests me. Were you thinking of doing similar? Or just thinking out
> loud..

I'm finding there are a limited number of hours in a lifetime and I
tend to fill too many of them wrongly already :-0>

It occurred to me that it was one more thing that would be fun to try
BUT it was unlikely I really would. So I described it as thoroughly as
reasonably possible off the cuff in a few minutes. In NZ that makes
anything novel unpatentable here immediately - and in the US, as I
understand it, gives you one year precedence to do something about it.
So, if there is anything usefully novel in what I said - and even if
there is there are numerous people on the list who could come up with
similar, then you (we :-) )  have 1 year to make something of them. A
year from now they become public domain if they aren't already old old
hat.

I rather like the pop back at > feed rate and optional shields up
addition. It would be as magic as SawStop is now - but better. You
reach for the blade and suddenly it is a few mm further away that it
has been and there is a guard between you and it ad the blade is
(perhaps optionally) decelerating at a saner rate. If you can get the
pull back and shield working well then you don't need to stop the saw
- just trip the shield release and continue.


                 Russell

2010\07\12@143907 by Olin Lathrop

face picon face
> For following, all rights reserved.
> :-) - what rights those are vary with administration. NZ = none. US =
> much AFAIK.

Probably not.  I don't remember signing a non-disclosure or intellectual
property agreement with you, so I believe I am free to use anything you say
to me without any obligation to you.  You also can't prove that I didn't
have some of those ideas independently before you listed them publicly,
which I did by the way.  I'm reasonably sure that your chance of
successfully getting a court to force me to give you compensation should I
develop these ideas is very slim.

>From the little I've been exposed to patent law by having been thru the
process a few times, I think this starts a 1 year clock ticking within which
anyone that wishes to file a patent must do so, else it can be rejected
easily if filed later.  I also think that many of your ideas are obvious, in
the sense that many engineers given the basic problem statement would come
up with them, although the patent office unfortunately seems unwilling to
apply the obviousness criterion much.

Fortunately, I have no current plans to develop a table saw safety device,
but find the exercise of thinking about one interesting.  Keep in mind that
ideas are cheap.  What's not cheap is the hard work to develop them into
real usable and producable products, and to effectively market and sell
those products.

One of the ideas I had I don't remember you mentioning (although I
admittedly didn't read much of your original message) is in part using the
blade's own momentum to move it out of the way.  Table saw blades always
spin so that the top of the blade is moving towards the front of the saw
(apposing the direction of feed).  Therefore a brake applied behind the
rotation axis would simultaneously slow the blade down while also converting
the blade's rotary momentum into downward translation.  That would require a
mechanism to allow the blade to move downward while the brake is applied at
the same time.  Harnessing the momentum or energy stored in the rotating
blade should require less external energy input, and probably be less
stressful on the various mechanisms.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\12@144205 by RussellMc

face picon face
> It seems that most injuries are caused this way from what I have read.

Far more than by blade contact they say.

{Quote hidden}

If the blade is well above the work the blade tends to pass more
vertically throug it and has less flinging vector - or so they say.

Having a guard which clamped the work near the table surface so it
could only fling horizontally may also help.

User education is probably a major factor - as in eg chainsaw kickback.


                 Russell


On 13 July 2010 06:20, Oli Glaser <RemoveMEoli.glaserspamTakeThisOuTtalktalk.net> wrote:
>> and the piece of metal caught and kicked back.
>

>
> -

2010\07\12@145152 by Olin Lathrop

face picon face
Oli Glaser wrote:
>> and the piece of metal caught and kicked back.
>
> It seems that most injuries are caused this way from what I have
> read. As far as I know none of the safety systems currently available
> prevent this from happening,

Not true.  My table saw has little spring loaded grabber thingies designed
not to let the wood slide backwards, for example.  It's just a ordinary off
the shelf table saw I got from Sears 25 years ago.  The wood has to get far
enough to catch on the grabber thingies, but they can help in a lot of
cases.

I am also concious of the plane of the spinning blade, and try not to stand
in that plane.  I often stand to one side enough so that if a small piece
broke off and got flung by the blade, I wouldn't be in its path.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\12@145334 by RussellMc

face picon face
>> For following, all rights reserved.
>> :-) - what rights those are vary with administration. NZ = none. US =
>> much AFAIK.

> Probably not.

If I had been super serious about it it wouldn't be on the list BUT I
suspect I'm closer to right than not. BUT that doesn't mean I'm right
about being right :-).

> I don't remember signing a non-disclosure or intellectual
> property agreement with you, so I believe I am free to use anything you say
> to me without any obligation to you.  You also can't prove that I didn't
> have some of those ideas independently before you listed them publicly,
> which I did by the way.  I'm reasonably sure that your chance of
> successfully getting a court to force me to give you compensation should I
> develop these ideas is very slim.

I THINK that the 3 above points are wrong - but I may well be wrong.
I understand that in the US I have one year to file a patent from
proven date of discovery or disclosure. This covers the whole lab book
record procedures which Intel and co take so seriously -and which are
unknown here.

So, AFAIK, if I were to file a patent I could use that post for
priority purposes.

As you note, there is little new under the sun in this sort of
technology but patent lawyers will probably try to tell you that
novelty can be demonstrated :-) - they may be right.

> I also think that many of your ideas are obvious, in
> the sense that many engineers given the basic problem statement would come
> up with them, although the patent office unfortunately seems unwilling to
> apply the obviousness criterion much.

Yes, more or less.
Seoul semiconductor recently got a Nichia patent overturned on the
grounds of lack of novelty (I think I got the 2 companies correct).
Much that is patented is not novel to an engineers brain. We seem to
live in an alternative reality :-).

> Fortunately, I have no current plans to develop a table saw safety device,
> but find the exercise of thinking about one interesting.  Keep in mind that
> ideas are cheap.  What's not cheap is the hard work to develop them into
> real usable and producable products, and to effectively market and sell
> those products.

That lesson I am extremely well aware of :-) :-(.

> One of the ideas I had I don't remember you mentioning (although I
> admittedly didn't read much of your original message) is in part using the
> blade's own momentum to move it out of the way.  Table saw blades always
> spin so that the top of the blade is moving towards the front of the saw
> (apposing the direction of feed).  Therefore a brake applied behind the
> rotation axis would simultaneously slow the blade down while also converting
> the blade's rotary momentum into downward translation.

The SawStop does this. I see the need for a limited rearwards motion
to counter the feed rate while a guard is interposed and / or  slower
braking takes place. It could be on an arc as long as the velocity
component away from the user was adequate..


 Russell

2010\07\12@145642 by Olin Lathrop

face picon face
RussellMc wrote:
> I rather like the pop back at > feed rate and optional shields up
> addition.

I like the pop back too, but am not so sure about the shield.  There is no
good direction to have the shield come from.  The shield would also have to
engage quickly, so you might just trade off having your finger sliced off
cleanly by the shield, as apposed to shreadded by the blade.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\12@145754 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "Olin Lathrop" <olin_piclistEraseMEspam.....embedinc.com>
Sent: Monday, July 12, 2010 7:39 PM
To: "Microcontroller discussion list - Public." <EraseMEpiclistspammit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

{Quote hidden}

I think (from watching the video Russell posted) that the Saw Stop does
utilise the blades momentum in the way you describe, so it effectively
"pulls" itself out of the way as well as stopping quickly. Not sure how this
idea easily translates to the "pop back without damaging blade" idea though,
unless you had some drive attachment that snapped into place (on a flywheel
or something) and used the momentum to do similar. Consistency would be an
issue here though, as it would depend on speed/weight etc, and unlike the
Saw Stop this is the main safety function so I suppose you would need it to
be as reliable as possible.


2010\07\12@150032 by Carl Denk

flavicon
face
Was said:
    If the blade is well above the work the blade tends to pass more
vertically
 There are available "finger boards" that can be bought or easily made
that push the work piece against a fence, and at the same time make a
ratchet of sorts to help prevent kick back. Also there are pawl like
teeth available that are on a bracket directly behind the blade that do
stop kick backs. These help reduce the need to have body parts near the
blade. And simple push sticks from scraps help a lot. There are many
tricks of the trade to help with safety, but just being aware is great.
I always chuckle when on a shop DIY TV show, they say "Read and
understand the safety in the manual"  Much the same when a newbie asks a
question on the list. Somewhere one needs to start, chicken or egg bit. :)

2010\07\12@151933 by Olin Lathrop

face picon face
Oli Glaser wrote:
> I think (from watching the video Russell posted) that the Saw Stop
> does utilise the blades momentum in the way you describe, so it
> effectively "pulls" itself out of the way as well as stopping
> quickly.

OK, I didn't watch the video.  I thought SawStop just stopped the blade
really quickly.

Another plan would be to do all this, but not as aggressively as SawStop.
Don't try to make it "safe", but make a cost effective backup system (to
something like a good guard that actually *prevents* injury) to reduce the
severity of injury should things get that far.  You might get a nasty cut,
but better than loosing a whole finger.  If the system is cheap enough and
the cost+hassle of resetting it is minimal, then it will be more widely
deployed and thereby save more injury than a fancy system nobody installs.
It also wouldn't give a false sense of security and invite sloppiness like
SawStop may in some cases.

So maybe the target should be to do what you can for a reasonable cost that
can be reset in one minute or less with no consumables to replace.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\12@153008 by RANDY ABERNATHY

flavicon
face
Olin:
 
Those are called "anti kickback fingers", go figure.  For the past 20+ years I have done industrial (and high end hobby) woodworking machinery.  It is true that table saws are toward the top of the heap as the most dangerous piece of machinery in a wood shop.

Randy Abernathy
CNC and Industrial Machinery
service, repair, installation and
design

4626 Old Stilesboro Rd NW
Acworth, GA 30101
Fax: 770-974-5295
Phone: 678-982-0235
E-mail:
RemoveMErandyabernathyEraseMEspamEraseMEbellsouth.net

--- On Mon, 7/12/10, Olin Lathrop <RemoveMEolin_piclistspam_OUTspamKILLspamembedinc.com> wrote:


From: Olin Lathrop <RemoveMEolin_piclistTakeThisOuTspamspamembedinc.com>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger
To: "Microcontroller discussion list - Public." <EraseMEpiclistspamspamspamBeGonemit.edu>
Date: Monday, July 12, 2010, 2:52 PM


Oli Glaser wrote:
>> and the piece of metal caught and kicked back.
>
> It seems that most injuries are caused this way from what I have
> read. As far as I know none of the safety systems currently available
> prevent this from happening,

Not true.  My table saw has little spring loaded grabber thingies designed
not to let the wood slide backwards, for example.  It's just a ordinary off
the shelf table saw I got from Sears 25 years ago.  The wood has to get far
enough to catch on the grabber thingies, but they can help in a lot of
cases.

I am also concious of the plane of the spinning blade, and try not to stand
in that plane.  I often stand to one side enough so that if a small piece
broke off and got flung by the blade, I wouldn't be in its path.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\12@160030 by Spehro Pefhany

picon face
At 03:30 PM 12/07/2010, you wrote:
>Olin:
>
>Those are called "anti kickback fingers", go figure.  For the past
>20+ years I have done industrial (and high end hobby) woodworking
>machinery.  It is true that table saws are toward the top of the
>heap as the most dangerous piece of machinery in a wood shop.

I asked a long-time woodworker about the anti-kickback fingers on my
table saw scratching the wood, and he said the first thing he did
when he got a new saw was to take them off. Go figure.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
RemoveMEspeffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



2010\07\12@161119 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "Olin Lathrop" <olin_piclistSTOPspamspamspam_OUTembedinc.com>
Sent: Monday, July 12, 2010 8:19 PM
To: "Microcontroller discussion list - Public." <spamBeGonepiclistSTOPspamspamEraseMEmit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

> If the system is cheap enough and
> the cost+hassle of resetting it is minimal, then it will be more widely
> deployed and thereby save more injury than a fancy system nobody installs.
> It also wouldn't give a false sense of security and invite sloppiness like
> SawStop may in some cases.
>
> So maybe the target should be to do what you can for a reasonable cost
> that
> can be reset in one minute or less with no consumables to replace.

Yes that sounds like a reasonable target, at least to start with, and if it
achieves better results then it's a bonus. The main reason it occurred to me
in the first place (as I'm sure it occurred to many others) was the cost of
resetting/replacing the SawStop system seems (to me) like it's biggest
problem, and possibly the main reason for people avoiding it (apart from the
fact that it cannot prevent injury completely as a guard if used correctly
will).
Something that works as a good backup system to a proper guard, is
reasonably cheap/simple to manufacture and install, resets quickly with no
extra cost, sounds quite attractive from a consumers point of view and as
you say would be less likely to invite sloppiness etc.
I think it makes sense that such a system would always have to have
something else in place that's non-reliant on a circuit performing it's duty
correctly, so backup rather than main safety feature sounds like a good
idea, especially for the times when people need to remove the guard for some
reason, or do something accidental or stupid.


2010\07\12@213501 by Jake Anderson

flavicon
face
On 13/07/10 06:11, Oli Glaser wrote:
{Quote hidden}

The field of stopping a rotating metal disk is a fairly thoroughly
explored field, they are called car brakes.
I'd wager with a hydraulic accumulator, shop air, and some ABS/brake
components you could stop the blade near arbitraily fast.
damage to the blade would be minimal, though you would want to check it
for warping if it got used often.


2010\07\12@221026 by Vitaliy

face
flavicon
face
Jake Anderson wrote:
> The field of stopping a rotating metal disk is a fairly thoroughly
> explored field, they are called car brakes.
> I'd wager with a hydraulic accumulator, shop air, and some ABS/brake
> components you could stop the blade near arbitraily fast.
> damage to the blade would be minimal, though you would want to check it
> for warping if it got used often.

I don't think that car brakes are designed to stop "arbitrarily fast",
compared to the time scales in question.

Vitaliy

2010\07\12@223140 by RussellMc

face picon face
> I asked a long-time woodworker about the anti-kickback fingers on my
> table saw scratching the wood, and he said the first thing he did
> when he got a new saw was to take them off. Go figure.

The Sith use[d] double ended Light Sabres. Seymore Cray, reputedly,
keyed in the boots strap loader for the first CDC mainframe 'out of
his head' using console switches (embellish that mis-re-collection as
desired) and designed Cray super computers in his head while digging
tunnels from his home to the nearby park while being advised by pixies
(or was it elves?) , real men program mainframes in assembler (or,
when in a real hurry, program in C). The rest of us would be advised
to retain the anti-kickback fingers.


          Russell

2010\07\12@224210 by RussellMc

face picon face
> Jake Anderson wrote:
>> The field of stopping a rotating metal disk is a fairly thoroughly
>> explored field, they are called car brakes.
>> I'd wager with a hydraulic accumulator, shop air, and some ABS/brake
>> components you could stop the blade near arbitraily fast.
>> damage to the blade would be minimal, though you would want to check it
>> for warping if it got used often.

> I don't think that car brakes are designed to stop "arbitrarily fast",
> compared to the time scales in question.

True, but the components may be useful.

A very significant part of Porches real world advantage on "real"
roads comes from the attention put into the braking system. A top
automotive braking system needs to have a horsepower rating similar to
that of its engine. F1 brake disks provide pretty images for
cameramen.

Automotive components are designed to handle instantaneous energy
transfer levels vastly in excess of that liable to be encountered  in
"very very fast" blade stopping. I suspect that even a very modest
automotive caliper brake would be a good enough foundation, provided
response times were adequate. Significant issues may be controlled
coupling to the blade surface rather than using the teeth. Servo
braking may be required to modulate the braking force to maintain a
desired braking profile. Electrically applied force  braking also
entirely feasible. *IF* blade can be adequately coupled to hub then
the hub could be braked allowing far more controlled design of brake
ie doesn't have to friction couple to a blade surface. Existing hubs
are usually friction afaik which almost certainly would not be good
enough.



                  Russell

2010\07\12@225126 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "Oli Glaser" <@spam@oli.glaser@spam@spamspam_OUTtalktalk.net>
Sent: Tuesday, July 13, 2010 3:23 AM
To: "Jake Anderson" <spamBeGonejakespamKILLspamvapourforge.com>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

{Quote hidden}

Yes indeed. However, we were talking of *moving* the entire metal disk
itself very quickly - to remove it from the immediate area to which a
finger/hand/random body part may be approaching. Actually stopping the disk
*rotating* quickly may not be necessary, in contrast with the SawStop
system
which does exactly that.
The discs in car brakes are a lot more sturdy and designed to withstand the
extreme forces applied during braking. I imagine it would be a far bigger
challenge to stop a disc blade quickly enough without damaging it due to
the
material being a lot thinner and the equivalent forces being higher - it
seems logical that if it was possible easily/cheaply the SawStop system
would probably do it that way to avoid the cost of replacement parts. I
find
it hard to believe the inventor did not try it this way first, but you
never
know.. Or maybe he considered it not to be certain enough that way, the
chances of failure to stop in time to be unacceptable - or something along
those lines.



2010\07\13@020319 by Sean Breheny

face picon face
On Mon, Jul 12, 2010 at 10:41 PM, RussellMc <spamBeGoneapptechnz@spam@spamspam_OUTgmail.com> wrote:
> desired braking profile. Electrically applied force  braking also
> entirely feasible. *IF* blade can be adequately coupled to hub then
> the hub could be braked allowing far more controlled design of brake
> ie doesn't have to friction couple to a blade surface. Existing hubs
> are usually friction afaik which almost certainly would not be good
> enough.


I once developed an electric-powered helicopter (in graduate school
for my thesis). It was one of the now-common four-rotor designs. The
rotors were all fixed pitch propellers - two of them off-the-shelf and
the other two custom made for us by APC to be pusher versions of the
first two. This allowed positive net thrust production with the props
counter-rotating in pairs. I digress...

The relevant part of this to the braking discussion is this: the motor
was about 1 HP, 30000+ RPM top speed (high performance expensive hobby
electric motor) and was coupled to the 18-inch diameter, about 200
gram propeller via a 6:1 reinforced timing belt drive reduction. Once
while it was spinning in a test stand at hover speed (4200 RPM), I
decided to see what happened when I instantaneously commanded 0 RPM.
The controller I had designed would not reverse the sign of the PWM
but it would short the motor. As soon as I hit the enter key on the
datalogging/command PC, the prop came to an instant complete stop,
accompanied by a loud thud/twang. No damage to motor, prop, timing
belt, or pulleys. The delay between command and full stop was totally
imperceptible to my brain. It was probably not as fast as this SawSafe
unit, but it couldn't have taken more than a few 10s of milliseconds
to stop the prop. It really is amazing what electric motors can do!

Sean


>
>
>
>                   Russell
>

2010\07\13@021349 by Jake Anderson

flavicon
face
On 13/07/10 12:51, Oli Glaser wrote:
{Quote hidden}

Moving all that requires accelerating alot of gear though, its not just
the blade, its all the rest of the gumf, some beefy bearing blocks,
pullys etc.
Also moving it back only covers a finger approaching from the direction
of travel, so you have to go down, and to below the table surface to
really be "safe" that could be in the 10's of cm range.

Most of the work in the car brake rotors is to absorb the heat energy of
2000kg's of car slowing down without warping.
If you clamp equally onto the blade it shouldn't warp significantly.

It would probably cost more initially than the sawstop but then if it
trips a few times the worst your out is a blade.

OHS here is pushing for everything to have brakes anyway, things aren't
allowed to spin on after you press the off button so you could perhaps
use the system for a (more gradual) stop in that case as well.

2010\07\13@024607 by Jake Anderson

flavicon
face
On 13/07/10 12:09, Vitaliy wrote:
> Jake Anderson wrote:
>    
>> The field of stopping a rotating metal disk is a fairly thoroughly
>> explored field, they are called car brakes.
>> I'd wager with a hydraulic accumulator, shop air, and some ABS/brake
>> components you could stop the blade near arbitraily fast.
>> damage to the blade would be minimal, though you would want to check it
>> for warping if it got used often.
>>      
> I don't think that car brakes are designed to stop "arbitrarily fast",
> compared to the time scales in question.
>
> Vitaliy
>
>    
brembo 6 piston caliper, 55 square centimeters worth of piston area
decent brake pad CF is ~.8
push the pressure to 3000PSI say.

a 10" blade is say .1" thick
weight is ~1kg, KE is 1500J

converting to a linear motion to make life simple
KE = .5MV^2
sqrt(KE /.5M) = V

V=54m/s
~=200km/h which seems to be ~ the tip speed of the blade as calculated
by others so it seems good.

clamping force on the disk is 8.5 square inches * 3000PSI = 25500 pounds
= 11590kg = 113590N
deceleration force = .8 * clamping = 90872N
acceleration = F/M = A = 90872M/s/s (~9000G)
v^2/2a = s
57^2/2*90872 = 0.017876794 meters
so about 20mm give or take.

seems fast enough for me?


2010\07\13@065930 by Dave Tweed

face
flavicon
face
Jake Anderson wrote:
{Quote hidden}

But that's only the very last step in the process. How long does it take
from when you receive the electrical trigger signal to when you actually
have the pads against the blade with 3000 PSI behind them?

-- Dave Tweed

2010\07\13@082006 by Dave Lagzdin

picon face
On 12 July 2010 08:49, RussellMc <TakeThisOuTapptechnzspamspamgmail.com> wrote:
> Note that in some cases a SawStop stop may be more dangerous or much
> more dangerous. If using eg Tungsten Carbide tipped blades you may
> wish to disable it. As long as you turn it back on subsequently ... .

That would make it useless.
I haven't used a non carbide blade in 20+ years. I doubt any
professional shop uses steel blades.

2010\07\13@082629 by Olin Lathrop

face picon face
Jake Anderson wrote:
> The field of stopping a rotating metal disk is a fairly thoroughly
> explored field, they are called car brakes.

Not really.

Car brakes have to dissipate a lot of energy and must be able to be "soft".
In other words, the control input is proportional, and the wheel must NOT be
frozen for less than full input.  That's rather different from stopping
something as fast as possible.

The energy dissipation issues are totally different too.  Car brakes have to
be able to take repeated application such that the heat must be transferred
away.  With the table saw blade, the whole event is over so fast that the
heat has no time to move anywhere before it stops being produced.  Then you
have lots of time to eventually dissipate it.

So no, people haven't been stupid (as your statement tries to imply in a
backhanded sort of way) for failing to suggest a car brake should be copied.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\13@083739 by Bob Ammerman

flavicon
face
>> brembo 6 piston caliper, 55 square centimeters worth of piston area
>> decent brake pad CF is ~.8
>> push the pressure to 3000PSI say.
>>
>> a 10" blade is say .1" thick
>> weight is ~1kg, KE is 1500J
>>
>> converting to a linear motion to make life simple
>> KE = .5MV^2
>> sqrt(KE /.5M) = V
>>
>>  <snip calculations>
>> so about 20mm give or take.
>>
>> seems fast enough for me?
>
> But that's only the very last step in the process. How long does it take
> from when you receive the electrical trigger signal to when you actually
> have the pads against the blade with 3000 PSI behind them?

But you also have to stop the motor, which has a lot more mass than the
blade.

-- Bob Ammerman
RAm Systems

2010\07\13@084016 by Olin Lathrop

face picon face
Jake Anderson wrote:
{Quote hidden}

But you're leaving out lots of stuff.  What mechanism is going go from 0 to
3000 PSI on 8.5 square inches of brake surface, and how long will that take?
The actual braking doesn't sound like the hard part, but actuating the brake
in a very short time is.  The larger you make the brake, that harder that
will be.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\13@084621 by Alan B Pearce

face picon face
> But you also have to stop the motor, which has a lot more mass than
the
> blade.

However if the blade is belt driven (which is quite likely to get a
suitable blade speed) you can always stall the blade, and let the belt
slip. Doesn't quite meet the full spec of no damage, but for what should
be a one-off event (at least very irregular) it does the job.
--
Scanned by iCritical.

2010\07\13@090729 by Jake Anderson

flavicon
face
On 13/07/10 22:26, Olin Lathrop wrote:
> Jake Anderson wrote:
>    
>> The field of stopping a rotating metal disk is a fairly thoroughly
>> explored field, they are called car brakes.
>>      
> Not really.
>
> Car brakes have to dissipate a lot of energy and must be able to be "soft".
> In other words, the control input is proportional, and the wheel must NOT be
> frozen for less than full input.  That's rather different from stopping
> something as fast as possible.
>    
The principle of operation however can be the same, and given the amount
of work done in the field it seems silly not to draw from it.
Not just car brakes, clutches, specifically dog clutches may also be useful.
> The energy dissipation issues are totally different too.  Car brakes have to
> be able to take repeated application such that the heat must be transferred
> away.  With the table saw blade, the whole event is over so fast that the
> heat has no time to move anywhere before it stops being produced.  Then you
> have lots of time to eventually dissipate it.
>    
Exactly, so car brake style engineering should be a superset of the
solution rather than a subset.
> So no, people haven't been stupid (as your statement tries to imply in a
> backhanded sort of way) for failing to suggest a car brake should be copied.
>    
You calling somebody for implying other people are stupid? pots n
kettles isn't it?
Or is it different when your on the other side?

Regardless, any high and mighty tone wasn't directed at Oli Glasser or
anybody else for that matter, I had been reading the thread and the idea
occurred to me, so I replied at that point. No offence was meant and if
any taken then I'm sorry.

2010\07\13@091753 by Jake Anderson

flavicon
face
On 13/07/10 20:59, Dave Tweed wrote:
{Quote hidden}

If you continue the car brake system analogy the pads should be sitting
microns off the disk, basically push them up to it (gently), spin the
saw up and let the wobble in the blade push them off. So application
time should be minimal, valve actuation time again can be almost
arbitrary direct injection diesel injectors operate in a similar problem
range although at 15,000 PSI. I don't have the numbers at hand.

Even so, if you doubled or tripled that stopping distance your still
better off, a deep cut (but you'll do worse in a kitchen) vs cutting
your finger off, and you'll use it because it wont kill the blades on a
false positive (also means you can make it more sensitive and perhaps
prevent a false negative)

2010\07\13@091908 by Jake Anderson

flavicon
face
On 13/07/10 22:37, Bob Ammerman wrote:
{Quote hidden}

The last one I pulled apart was belt driven (it was actually having
problems with belt slip) so just let the belt slip on purpose, if its an
issue perhaps a clutch in the string may help.

2010\07\13@092420 by Jake Anderson

flavicon
face
On 13/07/10 22:40, Olin Lathrop wrote:
{Quote hidden}

As I said in my original post a hydraulic acumulator connected to shop
air would provide you with a 3000PSI reservoir, then a valve analogous
to a fuel injector or ABS valve to apply that pressure to the cylinders,
the flow should be minimal so the pressure ramp time should be as well.
Besides you have complete control over these aspects of it so its "just
a matter of engineering". The basic principle seems sound, provided you
don't fling the carbide off the blade it seems a valid solution.
Any system that is going to pull the blade out of the way of fingers is
going to need to move at a similar pace so the carbides falling off is
going to be the make or "brake" part of the problem.

2010\07\13@093152 by Jake Anderson

flavicon
face
On 13/07/10 23:07, Jake Anderson wrote:
> Regardless, any high and mighty tone wasn't directed at Oli Glasser or
> anybody else for that matter, I had been reading the thread and the idea
> occurred to me, so I replied at that point. No offence was meant and if
> any taken then I'm sorry.
>    
Ahh I recall now where my ire was raised, Russell mentioned the patent
office, that probably contributed to the tone of the sentence.
An office which allowed patent 6360693 (a stick, cough sorry an animal
exercise device) is going to get my juices flowing.

As such citing "prior art" in the field of stopping spinning metal seems
not to have been out of line.

2010\07\13@094326 by Oli Glaser

flavicon
face


--------------------------------------------------
From: "Jake Anderson" <jakeEraseMEspamvapourforge.com>
Sent: Tuesday, July 13, 2010 2:24 PM
To: "Microcontroller discussion list - Public." <RemoveMEpiclistEraseMEspamspam_OUTmit.edu>
Subject: Re: [TECH]:: Auto stop table saw - demo with real finger

{Quote hidden}

I'm sure the brake system is achievable (although I do wonder why the
SawStop inventor decided against it), but the SawStop system *also* pulls
the blade out of the way using the blades own momentum. I think this
probably decreases the likelihood of injury a fair bit compared to just
stopping the blade.
That's why it occurred to me that if you were aiming to pull the blade out
of the way anyway, why not just concentrate on doing this as well as
possible and then there is no broken blade (as in the SawStop - not the car
brake of course), no need to stop it quickly, and if it works as intended,
no injury at all as there will be no contact with the blade whether spinning
or not. Plus I imagine there will be less wear and tear over time (to blade,
belt etc)



2010\07\13@094430 by Olin Lathrop

face picon face
Jake Anderson wrote:
> The last one I pulled apart was belt driven (it was actually having
> problems with belt slip) so just let the belt slip on purpose, if its
> an issue perhaps a clutch in the string may help.

If the blade is belt driven and the brake also swings the blade down as it
stops it, then this same swinging action can be arranged to loosen the belt.
That would give considerably more time (seconds as appossed to milliseconds)
to stop the motor and belt before either got damaged.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\07\13@114619 by RussellMc

face picon face
Summarising tech spec:
> ... motor was about 1 HP, 30000+ RPM top speed
> 18-inch diameter, about 200 gram propeller
> via a 6:1 reinforced timing belt drive reduction. Once
> while it was spinning in a test stand at hover speed (4200 RPM),
> instantaneously commanded 0 RPM.
> short[ed] the motor.

> As soon as I hit the enter key... the prop came to an instant complete stop,
> accompanied by a loud thud/twang. No damage to motor, prop, timing
> belt, or pulleys. The delay between command and full stop was totally
> imperceptible to my brain. It was probably not as fast as this SawSafe
> unit, but it couldn't have taken more than a few 10s of milliseconds
> to stop the prop. It really is amazing what electric motors can do!

E&OE

Decided a quick order of magnitude check worthwhile - motor/rotor
braking is of interest.
Gross assumption: mean mass radius = 6" and 0.5mV^2  = adequate energy model.

Calculated and got ~~~ 640 Watt/second which at one HP would take
about 1 second to stop.
Puzzled.
Then I noted you'd said 4200 RPM hover rev and I'd used full speed !!!
Recalc at v^2 = 50 times less and get about 13 Watt-second.
Maybe 20 mS at 1 HP and instantaneous braking under short could be
above rated power at least for part of profile.
Anything under say 40 mS is liable to be close to instantaneous to the eye.

At 20 mS full stop that's still about 2 full turns - so SawStop
technology would be marginal to useless there.
Energy level is not Too bad - nasty injury possible.
At full throttle you'd be "in trouble".
600+ Joule is well above 38 Magnum energy :-(.

E&OE, again

          Russell

2010\07\13@123636 by Sean Breheny

face picon face
Hi Russell,

Your final result sounds about right. Note that the 30000+ RPM I
mentioned is at the motor shaft. There was a 6:1 reduction to the prop
shaft. Our full-throttle prop speed was about 5700 RPM. Note that this
is actually 80% more thrust (because thrust goes as rpm^2) and 2.5
times the power (as power goes as rpm^3). Actually, I think that the
electrical input power to the motor was about 400W at hover, which
would mean 1kW at full power. Our batteries would not allow full power
for very long - it was only used for maneuvers. This machine overall
weighed about 6.5 kg and could accelerate upward at about 1 G. Prop
tip speed at hover was 100 m/sec. It was very scary to be around - we
used a 1 cm thick polycarbonate police riot shield whenever we needed
to get close while it was hovering.

I looked up the APC 18x6W prop we used and it only weighed 136 grams.
However, there was also the motor rotor moment of inertia, which,
although much less on an absolute scale, gets multiplied by the gear
ratio squared (6*6=36). So, a 200g at 6 inch radius estimate is
probably fair.

The 1HP rating for the motor is not a peak power limit but rather a
thermal limit for typical flight durations. The motor is perfectly
capable of MUCH more (maybe 10x) peak power for very short periods of
time. Shorting the motor at a prop shaft speed of 4200RPM (25200 RPM
at the motor shaft) will generate several kilowatts of electrical
power in the initial stages of braking. To give you an idea of the
current involved in shorting this motor, in an earlier test, shorting
it through a 10 cm long, 2mm wide PCB trace caused the trace to
explode (several cm of it were completely gone from the PCB surface).
I had to add external wires to the board as a quick hack to beef up
the current handling ability.

Sean





On Tue, Jul 13, 2010 at 11:45 AM, RussellMc <@spam@apptechnzRemoveMEspamEraseMEgmail.com> wrote:
{Quote hidden}

>

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...