Greetings all...

I have been constructing a new crosscut sled, and the thought came to me,
how close does this need to be to make nice joints...

Instead of using a square I cut the sides off an 8" square of birch
plywood, I then ripped a strip off the first side and measured the
difference with a dial caliper, and over 8" there is less than 1/128
difference, but that would be compounded over 4 cuts right?

So would you guys try to get it closer of just leave it as is?

"DCH" wrote in message
8...
Greetings all...

I have been constructing a new crosscut sled, and the thought came to me,
how close does this need to be to make nice joints...

Instead of using a square I cut the sides off an 8" square of birch
plywood, I then ripped a strip off the first side and measured the
difference with a dial caliper, and over 8" there is less than 1/128
difference, but that would be compounded over 4 cuts right?

So would you guys try to get it closer of just leave it as is?


I might try to get it closer but some times measurements can deceive or may
not be done accurately. Sawing technique can often render measurements
useless.

Use some scrap wood, make a frame and ask yourself, is that good enough?

Keep in mind that opposite parallel sides must be exactly the same length
also or even perfect 45's will not close properly. Use a stop to insure
same length "opposite" sides.

"Leon" wrote in
:


"DCH" wrote in message
8...
Greetings all...

I have been constructing a new crosscut sled, and the thought came to
me, how close does this need to be to make nice joints...

Instead of using a square I cut the sides off an 8" square of birch
plywood, I then ripped a strip off the first side and measured the
difference with a dial caliper, and over 8" there is less than 1/128
difference, but that would be compounded over 4 cuts right?

So would you guys try to get it closer of just leave it as is?


I might try to get it closer but some times measurements can deceive
or may not be done accurately. Sawing technique can often render
measurements useless.

Use some scrap wood, make a frame and ask yourself, is that good
enough?

Keep in mind that opposite parallel sides must be exactly the same
length also or even perfect 45's will not close properly. Use a stop
to insure same length "opposite" sides.



its kinda funny....but I was able to get it almost perfect, within a
needles width on the dial caliper, I would think that would be close
enough, but last night I had painstakingly straightend and squared the
bit that would become the main fence and after all that I managed to
install it with the wrong face to the saw....after switching things
around my test cuts came out much better....

Subject

Remember the old saying, "A flying Red Horse can't spot the difference
from 1,000 ft."

Still works.

Lew

On Apr 22, 12:56 pm, "Leon" wrote:

Use some scrap wood, make a frame and ask yourself, is that good enough?

I couldn't agree more with everything Leon said, not the least being
to try it out to see for yourself.

You are talking about making a wooden device that you think will hold
tolerances to within one 0.0078125th of an inch. Think about it; a
wooden jig that will hold completely true through humidity changes,
temperature changes, techniques differences, movement of the sled in
response to different weight, density and size of material, etc.,
etc. Not happenin'.

Don't cheat yourself For dead bang spot on, buy a calibrated miter
gauge.

Or... just use your miter saw. I for one have never understood the
folks that have to use their tablesaw for everything from precision
miter work for small pieces like building picture frames to making
raised panel doors.

Maybe its just me. I could see it on really large pieces I guess, but
even then... how often does one need a miter on a 12" board?

Robert

Garage_Woodworks wrote:

When calibrating tools 'eye-balling it' is NEVER good enough in my shop -
wooden or otherwise.

In lots of situations, the precision of decent eyeballs and finger tips
is actually very useful for checks.

Two examples:

Thickness planer parallelism - plane a board, cut it in half, put
opposite edges together, rub the fingers over the meeting point.

Table saw blade or miter gauge/sled 90 degree accuracy - Cut a board in
half, flip one board, place the cut edges together and check for gaps.

If they don't work out, the calibration tools make recalibration easier
and faster than more test cuts, but the cuts are good enough for
in-service spot checks.

"Garage_Woodworks" .@. wrote in message
...


"Leon" wrote in message
...
I might try to get it closer but some times measurements can deceive or
may not be done accurately. Sawing technique can often render
measurements useless.

So why not eliminate the 'sawing technique error from the equation when
calibrating the fence?? Hmmmm.

That would be my point. I am indicating that measuring will not always over
come techique and when you are talking a visible or invisible joint line the
technique problem may be so small that it could involve dozens of factors.
Including but not restricted to, is your table flat, is your stock perfectly
straight, are you working with soft or hard wood, is you blade "sharp", is
the surface of your table smooth, and the list goes on.



Use some scrap wood, make a frame and ask yourself, is that good enough?

This won't do it. 'Sawing technique error still there.

It certainly does it for me. Because we are not machines there will always
be some degree of imperfection with every cut. Measurements only get you so
close and if you were able to obtain the perfect setting your technique will
always add some degree of error from one cut to the next.

"Garage_Woodworks" .@. wrote in message
...

"Lew Hodgett" wrote in message
news:GCyPj.11330$pH4.4572@trnddc06...
Subject

Remember the old saying, "A flying Red Horse can't spot the difference
from 1,000 ft."

Still works.

Lew



When calibrating tools 'eye-balling it' is NEVER good enough in my shop -
wooden or otherwise.


Perhaps one day you will learn to make eye-balling work for you. It
certainly speeds up production.

"Garage_Woodworks" .@. wrote in message
...


So why not eliminate the error you have control over (the original
calibration)?? Why settle for a sloppy calibration because you might
experience wood movement?

Why correct error so small that it does not factor? Very often climate
changes totally negate all accurate measurements. One day your loose parts
to a drawer fit fine, the following day the fit is too lose or too thight.
With years of experience you learn to compensate and work with mother
nature. Precice measuring tools are fine to use for a start but the results
are often out of phase with climate changes and your technique.


Error Total = Error in calibration + error from wood movement over time +

Keep in mind that neither is a constant, one element is constantly changing
which pretty much over rides the results/effect of a particular setting.


The 'error in calibration' can be eliminated completely w/o making a
single test cut.

Calibration is important but results with any given calibration often change
with the climate and your technique. It's not a labratory invironment where
you can calibrate "everything" with materials that are a constant shape and
size.

Experience counts for a lot in the results you get.








--
Brian
www.garagewoodworks.com

"In the land of the blind, the one-eyed man is king."
Desiderius Erasmus

On Apr 23, 7:22 am, "Garage_Woodworks" .@. wrote:


You don't even need to settle for the 'error from wood movement'. Just
check the dam thing before you use it and fix it.

LMAO.... pretty hot issue, eh? I like it when someone replies to
their own posting.

You missed my point. You should always get your jigs, measuring
devices, etc. as close to perfect as you can.

We were talking about a WOODEN jig. In my experience, I have
never seen a usable wooden jig hold to perfect tolerance under all
conditions. Close, but not perfect. Obviously you feel differently.
To me, if I am to put a lot of time and effort into making something,
it needs to a design that can perform to the standards I want. Wood
can be a satisfactory fabrication medium for certain things, but not
repeatability of tolerances within a couple of thousands over a period
of use.

But since DCH didn't post what size board he was using, how do we know
what the table saw is the best tool for the job? What if he is
cutting 4" or 6" wide material? Should he be using a table saw and a
homemade sled?

Certainly if that is the case, I would put any of my three miter saws
against his table saw and shop built jig. And for compound cuts, a
table saw is not even a consideration if I can get it under my miter
saw.

The milled aluminum and steel bed and fence calibrated with my
machinest's square are much more comfortable for me to rely on than
plywood, mdf, white glue,. hardwood, etc.

Robert

"Garage_Woodworks" .@. wrote in message
...


Perhaps one day you will learn that precision is better and faster when
you learn the tricks.


That is not what I was talking about, I'd be willing to bet that you would
find nothing wrong with my precision and or joints. It's that with years of
experience you learn how to achieve that with out having to use high
precision measuring tools with every adjustment.

Case in point, you bought a new Powermatic 2000 TS. If your measuring and
precision set ups on your old Delta saw were good why change saws? I'm just
saying that a precision measurement is not always the answer to a problem.

"Garage_Woodworks" .@. wrote in message
...

"Leon" wrote in message
...


Then eliminate the 'factors' you have control over. Like a precise
calibration.

That is fine but if the precice calibration only corrects 10% of the problem
you need to learn how to make compensations to remedy the problem.

AND going back to my original points to the OP, dial calipers are good for
measuring and setting up equipment. The results of using that equipment is
not guaranteed to give good results when using wood as the material to
construct with.





Including but not restricted to, is your table flat, is your stock
perfectly straight, are you working with soft or hard wood, is you blade
"sharp", is the surface of your table smooth, and the list goes on.

All variables that effect final result. Which of those do you have
control over?

All of them if you can compensate, but for instance a precicely calibrated
saw will not eleminate tear out on the back side of a cut. There are other
factors to consider to minimise the tear out. Excessive tear out can ruine
a joint and technique can certainly minimize tear out. For instance when
cutting a 45 degree miter on your TS if you have the miter gauge clocked to
45 in a particular direction it prioduces a better/cleaner cut than if you
clock the gauge to 45 degreesin in the opposite direction. This all falls
into technique and no amout of precision will make both setting equal in
results on a consistant basis.

"Garage_Woodworks" .@. wrote in message
...

"Leon" wrote in message
...

Error Total = Error in calibration + error from wood movement over time
+

Keep in mind that neither is a constant, one element is constantly
changing which pretty much over rides the results/effect of a particular
setting.

Did you see any constants in my equation? That's why they are called
variables NOT constants. (See equation).

Well hopefuly your equipment is more of a constant and can be depended on
for repeated settings with out using an externam measuring device.

On Apr 23, 11:24 am, "Garage_Woodworks" .@. wrote:
wrote in message

You missed my point. You should always get your jigs, measuring
devices, etc. as close to perfect as you can.

I missed it because you never made this point. Where was it posted?

Sorry... you are right. I simply took it for granted that one wold
assume that if you take the time to build a jig, you would build one
to the best of your abilities OR requirements. To me, a jig should
work to tolerances that satisfy one's own needs. To me that was a
given for any experienced craft person, but you have made a good point
about taking things for granted. So to simplify, I believe you should
build a jig to tolerances that do the job they are designed to do as
good as possible, but with the thought of repeatability foremost in
mind.


You stated that you don't understand folks that like to use their table saw
for precision miter work.

It was a global statement that was obviously another avenue of
confusion. I should have said, "I don't understand why someone would
use a table saw to perform an operation that is better performed by a
purpose designed and built machine that is task specific for one
operation".

Having tried to cut 21/2" crown on a table saw, it sent me running
back to my miter saw. I needed stain grade work, and I was unable to
see how to cut one degree, or a half degree off with the table saw to
close a hair line crack in a joint. OK, some clarification he it
wasn't on a shop built cabinet. It was in a house, where every
ceiling corner is a square as the framers framed it, and the tape and
float guys finished it. Each small piece of a corner may have to be
cut several times to get the right angle to compliment the out of
square corner.

You are obviously a proud defender of the table saw, and looking at
your site (good work, BTW) it is easy to see how important that tool
is to you. BUT FOR ME.... if there is a better tool for the job, I am
all over it. My carpentry jobs rely on speed and accuracy. I am to
start a crown molding job in a house in a couple of weeks. I won't be
taking my table saw to do the cuts. Sadly, I have a tendency to go
with the tool that does the job the most accurately with the least
fuss.

If you are comfortable with you saw, wooden jigs and calibration
equipment, why not?

But since DCH didn't post what size board he was using, how do we know
what the table saw is the best tool for the job? What if he is
cutting 4" or 6" wide material? Should he be using a table saw and a
homemade sled?

Certainly if that is the case, I would put any of my three miter saws
against his table saw and shop built jig.

I would take that challenge.

I see where this is headed. And if you believe that a wood jig can
take the daily rigors of use as well as a purpose built metal jig, all
I can say is "good for you".

Since I rely on my tools for my living, I like metal guides, rails,
beds, ways, and metal on metal adjustments. I am tasked with working
on site 99% of the time, and my tools are loaded and unloaded day
after day. Just the movement knocks them out of ajdustment
sometimes. If I worked in the closed environment of a shop and had
all manner of tools on hand like the TSA Jr, dial gauges, extended
reach calipers, etc., at my finger tips, I just might feel
differently.

As with me, you are certainly welcome to your opinion.

YM obviously varies...

Robert

Garage_Woodworks wrote:
"Leon" wrote in message
...

So why not eliminate the 'sawing technique error from the equation
when calibrating the fence?? Hmmmm.

That would be my point. I am indicating that measuring will not
always over come techique and when you are talking a visible or
invisible joint line the technique problem may be so small that it
could involve dozens of factors.

Then eliminate the 'factors' you have control over. Like a precise
calibration.

Including but not restricted to, is your table flat, is your stock
perfectly straight, are you working with soft or hard wood, is you
blade "sharp", is the surface of your table smooth, and the list
goes on.

All variables that effect final result. Which of those do you have
control over?

Use some scrap wood, make a frame and ask yourself, is that good
enough?

This won't do it. 'Sawing technique error still there.

It certainly does it for me. Because we are not machines there
will
always be some degree of imperfection with every cut.

You are missing my point. See above.

Measurements only get you so close and if you were able to obtain
the perfect setting your technique will always add some degree of
error from one cut to the next.

See above.

Geez, even optical flats have tolerances. Figure out what your
tolerances need to be and then set up your system to maintain them.

--
--
--John
to email, dial "usenet" and validate
(was jclarke at eye bee em dot net)

I wrote:

Remember the old saying, "A flying Red Horse can't spot the
difference from 1,000 ft."

It sparked a spirited discussion; however, it clearly defines a few
things.

In the data acquisition business, if you truly want an accurate
measurement, you make a differential measurement, not a single ended
one.

Single ended measurements introduce instrument as well as people
errors into the measurement, something differential measurements
eliminate because they cancel out.

Using ANY measuring instrument to make a measurement is a single ended
measurement.

Ask 5 people to make a measurement with a micrometer of vernier
caliper, and chances are pretty good you will get 5 different answers.

Set up a table saw, cut a piece, then break the set up.

Now reset the fence to the same dimension, cut a 2nd piece and compare
it with the first.

They will be close, but they will be different.

I submit your fingers are the best instrument for this measurement.

Biggest reason I know for "sizing" all the material for a job using a
single setting for the tool (Planer, table saw, etc) as the first step
in processing the wood.

BTW, you also do it all at the same time.

Weather conditions tomorrow will be different than today.

Lessons learned the hard way AKA: Expensive.

Lew

" wrote in
:

On Apr 23, 11:24 am, "Garage_Woodworks" .@. wrote:
wrote in message

You missed my point. You should always get your jigs, measuring
devices, etc. as close to perfect as you can.

I missed it because you never made this point. Where was it posted?

Sorry... you are right. I simply took it for granted that one wold
assume that if you take the time to build a jig, you would build one
to the best of your abilities OR requirements. To me, a jig should
work to tolerances that satisfy one's own needs. To me that was a
given for any experienced craft person, but you have made a good point
about taking things for granted. So to simplify, I believe you should
build a jig to tolerances that do the job they are designed to do as
good as possible, but with the thought of repeatability foremost in
mind.


You stated that you don't understand folks that like to use their
table saw for precision miter work.

It was a global statement that was obviously another avenue of
confusion. I should have said, "I don't understand why someone would
use a table saw to perform an operation that is better performed by a
purpose designed and built machine that is task specific for one
operation".

Having tried to cut 21/2" crown on a table saw, it sent me running
back to my miter saw. I needed stain grade work, and I was unable to
see how to cut one degree, or a half degree off with the table saw to
close a hair line crack in a joint. OK, some clarification he it
wasn't on a shop built cabinet. It was in a house, where every
ceiling corner is a square as the framers framed it, and the tape and
float guys finished it. Each small piece of a corner may have to be
cut several times to get the right angle to compliment the out of
square corner.

You are obviously a proud defender of the table saw, and looking at
your site (good work, BTW) it is easy to see how important that tool
is to you. BUT FOR ME.... if there is a better tool for the job, I am
all over it. My carpentry jobs rely on speed and accuracy. I am to
start a crown molding job in a house in a couple of weeks. I won't be
taking my table saw to do the cuts. Sadly, I have a tendency to go
with the tool that does the job the most accurately with the least
fuss.

If you are comfortable with you saw, wooden jigs and calibration
equipment, why not?

But since DCH didn't post what size board he was using, how do we
know what the table saw is the best tool for the job? What if he
is cutting 4" or 6" wide material? Should he be using a table saw
and a homemade sled?

Certainly if that is the case, I would put any of my three miter
saws against his table saw and shop built jig.

I would take that challenge.

I see where this is headed. And if you believe that a wood jig can
take the daily rigors of use as well as a purpose built metal jig, all
I can say is "good for you".

Since I rely on my tools for my living, I like metal guides, rails,
beds, ways, and metal on metal adjustments. I am tasked with working
on site 99% of the time, and my tools are loaded and unloaded day
after day. Just the movement knocks them out of ajdustment
sometimes. If I worked in the closed environment of a shop and had
all manner of tools on hand like the TSA Jr, dial gauges, extended
reach calipers, etc., at my finger tips, I just might feel
differently.

As with me, you are certainly welcome to your opinion.

YM obviously varies...

Robert


soooooo....do you guys think a variation of way less than 1\128
compounded over five eight inch cuts would be close enought for a dude
in his garage trying to make some nifty things for his house and
freinds?

This sure is better than all the spam we been gettin as of late...

"DCH" wrote:

soooooo....do you guys think a variation of way less than 1\128
compounded over five eight inch cuts would be close enought for a
dude
in his garage trying to make some nifty things for his house and
freinds?

Let The Flying Red Horse be your guide.G

Lew

On Apr 24, 9:37 pm, "Lew Hodgett" wrote:

Let The Flying Red Horse be your guide.G

LMAO...

No kidding!

Robert

On Apr 24, 9:21 pm, DCH wrote:

soooooo....do you guys think a variation of way less than 1\128
compounded over five eight inch cuts would be close enought for a dude
in his garage trying to make some nifty things for his house and
freinds?

This sure is better than all the spam we been gettin as of late...

Probably not if you are making segmented pieces or pentagons.
Although, if they were compound cuts and all oriented the same at the
time they were cut, and assembled with the same orientation I'd bet no
one would notice.

Seriously, it just depends on what you are making. I would suggest
you just try you jig out, and if you like the results, all the rest is
just baloney.

Robert

DCH wrote in news:Xns9A8AE3AAA90E3hayniedcgmailcom@
216.77.188.18:


soooooo....do you guys think a variation of way less than 1\128
compounded over five eight inch cuts would be close enought for a dude
in his garage trying to make some nifty things for his house and
freinds?

This sure is better than all the spam we been gettin as of late...

Going back to physics measurements, if your uncertainty (variation in the
cut in this case) is 1/128, then 5 cuts would give you a total
uncertainty of 5/128, or just over 1/32". That's +/- 5/128, too, so some
cuts could be more and some could be less.

One hidden truth is that some cuts will be a little less and some will be
a little more... You'll wind up somewhere in the middle of your
uncertainty range, not at the edge cases.

Puckdropper
--
You can only do so much with caulk, cardboard, and duct tape.

To email me directly, send a message to puckdropper (at) fastmail.fm

On Apr 25, 2:56 am, Puckdropper puckdropper(at)yahoo(dot)com wrote:

Going back to physics measurements, if your uncertainty (variation in the
cut in this case) is 1/128, then 5 cuts would give you a total
uncertainty of 5/128, or just over 1/32". That's +/- 5/128, too, so some
cuts could be more and some could be less.

One hidden truth is that some cuts will be a little less and some will be
a little more... You'll wind up somewhere in the middle of your
uncertainty range, not at the edge cases.

Exactly. And a great explanation, too. It illustrates well when it
is time to do something rather than to contunue to fiddle over the
last RCH.

And of course the optimum results would be that the cuts would somehow
be arranged (or cut) in a way that compliment each other making the
difference almost non existent.

Robert

" wrote in
:

On Apr 25, 2:56 am, Puckdropper puckdropper(at)yahoo(dot)com wrote:

Going back to physics measurements, if your uncertainty (variation in
the cut in this case) is 1/128, then 5 cuts would give you a total
uncertainty of 5/128, or just over 1/32". That's +/- 5/128, too, so
some cuts could be more and some could be less.

One hidden truth is that some cuts will be a little less and some
will be a little more... You'll wind up somewhere in the middle of
your uncertainty range, not at the edge cases.

Exactly. And a great explanation, too. It illustrates well when it
is time to do something rather than to contunue to fiddle over the
last RCH.

And of course the optimum results would be that the cuts would somehow
be arranged (or cut) in a way that compliment each other making the
difference almost non existent.

Robert


Are ya'll sure thats not backwards....the variance over 5 cuts is less than
1/128 not 1/128 over one cut....

On Apr 26, 9:19 am, DCH wrote:


Are ya'll sure thats not backwards....the variance over 5 cuts is less than
1/128 not 1/128 over one cut....

Puckdropper is right on. Try it yourself. Unless you try to orient
your cut pieces so that the variances align as complimentary angles,
they will compound themselves.

But just as importantly, the other issues that are raised here come
into play. That is technique, repeatability of the underlying
equipment, materials, etc.

Robert