LumberJocks

[another] Accuracy in Woodworking.

  • Advertise with us
Blog entry by antmjr posted 1385 days ago 5685 reads 1 time favorited 7 comments Add to Favorites Watch

Hi all. Well, I was in the blocklist of GarageWoodworks, so the only way to explain my point of view was to start a twin discussion , hope not to bother anyone of you.

The terms are accuracy and tolerance (which btw are often mistaken for each other). I guess there is the same difference in English as there is in Italian: accuracy refers to the precision of execution, to our ability to build something according to plan, while tolerance refers to the max error one can tolerate, according to the realistic accuracy one can expect from himself. Each of us must set his/her own tolerance.

Here the etymo of accurate, from which accuracy comes (http://www.etymonline.com/):

accurate: 1610s, “done with care,” from L. accuratus “prepared with care, exact, elaborate,” pp. of accurare “take care of,” from ad- “to” + curare “take care of”. The notion of doing something carefully led to that of being exact (1650s)

———————————————-
The term that intrigues me is tolerance (in its technical meaning). Accuracy apart, which is – realistically – the tolerance one must set in his/her project? is it always right to set the smallest tolerance, as the video suggests? Can one set a tolerance for a joint independently from the whole, I mean, without considering the rest of the building?

Here a somewhat extreme example that shows the problem of setting the right tolerance.

These fingers are pretty exact: the basic module is 6 mm (0,2362’), the finger is 5,8 mm thick (0,2283’), while the groove (I mean, the empty space between the fingers, I don’t know the exact term) is 6,2 mm wide (0,2441’).

Apparently there is a great tolerance: for each finger, the tolerance is 6,2-5,8= 0,4mm i.e. 0,0157’.
Unfortunately something went wrong though: in fact I had to hammer the two pieces to make the fingers enter the grooves (luckily the wood is black locust, hard enough to support my savage hammering).

Had I made larger grooves, everything would have worked without the need of forcing the fingers by means of a hammer. Now I know that I had to set the right tolerance for the basic module, and to decide the width of the groove accordingly. Here what happened:

Since the total error for the basic module was greater then the tolerance for each single finger, that total error could not be compensated by the groove between the fingers.

This example shows that it is silly to set a tolerance for a single finger joint without thinking at the whole first. In my example, if my smallest realistic tolerance for the basic module (with regard to my ability, tools, etc) is, say, +/-0,1 mm (0,004’), and I think it may be possible to do the same error for, say, 5 modules, the total error would be 0,5mm (0,02’), that is the tolerance a single finger joint must be able to handle in my case. A smaller tolerance for the single finger joint would simply mean that I have understood nothing about wood and woodworking.

So with regard to the advertising of that hightech-kerfmaker, “0,002’ is unacceptable”. Yes, 0,002’ is unacceptable, 0,02’ or better 0,025’ would be acceptable.

—-
(The other problem with joints so tight as that of the video is the glue: if one uses water based glue, the wood swells, and the joint may either break or not be assemblable, as most of us know)

-- Antonio



7 comments so far

View mafe's profile

mafe

9486 posts in 1716 days


#1 posted 1385 days ago

Interesting.
And yes I agree, allways set the tolerance from center to center, this is how we also layout houses as architects.
Best thoughts,
Mads
And then I can post also here:
I think there is always a mix of reality in these discussions.
A master craftsman spends year’s even generations to make the perfection, sometimes by making the same item for a life time. This is where you can make hand cutting with an accuracy that is the same as a kerfmaker or so.
Then there is craftsman ship what is when a craftsman work on a variation of wood products, he becomes highly trained, but not specialized.
Then there is production line, and here it’s really interesting to be accurate to the highest possible level, because this is what you can promise the customer.
And finally there are all the garage amateurs, where we do a project once in a while, for the challenge, need or fun.
Nothing are more right or wrong, but when you take this discussion, try to look at the projects you make, do they need the accuracy of 0.001? ‘What is it good for’?
Finally there are your skills, if your skills are high, you can adjust your table saw extremely precise down to the .00’s, are you a armature that wants’ perfection without making a test cut, then perhaps a dial indicator are wonderful.
Also I think that if it is precision that are so important, why not go for CNC, I think to spend so much time for perfection not seen, can be waste of time, and I would also never remove glue from a unseen place.
A good example from real life are also the house painter, when he paints an inside door he never paints the top because no one sees it – the amateur he will take the door of to make sure every corner are fully covered.
So my conclusion is:
There is no conclusion, nothing right or wrong.
Me I adjust my accuracy after the project, and would never use more accuracy than needed.
Others would not be able to sleep at night if the cut was 0.001 of on their new garden fence.

-- Mad F, the fanatical rhykenologist and vintage architect. Democraticwoodworking.

View antmjr's profile

antmjr

262 posts in 1811 days


#2 posted 1385 days ago

Thank you Mads, you are the kind person we know.
—-
Brian, I like discussing with the others, in fact my block list is and will always be empty.

-- Antonio

View ChuckV's profile

ChuckV

2399 posts in 2154 days


#3 posted 1385 days ago

I hope that I do not sound like I am being picky. There is a difference between precision and accuracy. Precision has to do with repeatability or consistency. Accuracy takes this another step:
accuracy = precision + correctness

You can be precise but wrong. Imagine that you are shooting arrows at a target. If you end up with a tight cluster of arrows in the barn 20 feet from the target, you are precise, but not accurate. Or in woodworking, you might be building a table with four 30” legs. If all four of them end up being withing .001” of 20”, youir cutting was precise, but not very accurate.

-- “That it will never come again / Is what makes life so sweet. ” ― Emily Dickinson

View Dark_Lightning's profile

Dark_Lightning

1691 posts in 1736 days


#4 posted 1384 days ago

Right, ChuckV. A simple way I used to teach the difference between accuracy and precision was to cut the first inch off some yardsticks. Then, I had the students measure the height, width and length of the lab (I taught physics), and calculate the volume. Some had the shorter yardsticks, some had good ones. Of course, they had different answers. When I explained that some of the yardsticks had one inch removed, I got a lot of complaints about “not fair!”. But the point was made, and they didn’t get marked off or a lower grade for their work. I bet they still talk about that class! If people get pulled out of what they expect in a science class, they don’t forget the lesson. Their measurements were precise to 1/8th inch marks on the sticks, but not accurate.

View stefang's profile (online now)

stefang

12875 posts in 1961 days


#5 posted 1384 days ago

I’m probably the last person who should give an opinion on this subject, especially when architects are involved in the debate, but I just can’t resist. My approach is more practical and philosophic than technical.

I have always felt that the appropriate precision, accuracy and tolerances for woodworking are pretty much related to what you are building and also very much based on it’s size as well. Smaller items require greater precision as even tiny inaccuracies tend to show up on small objects. To my way of thinking there are three important criteria. The first is the quality. Will it hold together? The second is the appearance. Does it satisfy the eye? and lastly (or maybe this should be first) does it satisfy the maker?

Having said the above, I will admit that when I started out in woodworking I didn’t fully appreciate the need for precision or accuracy. Over the years I have developed an appreciation of the advantages to be had by working accurately and my work has improved accordingly. However, any skill can be driven to the level of obsession. I think people like us who like to make things tend to be a little that way, and perhaps that is a positive thing that leads to improvement for folks like me and to greater things by those of you who are more gifted.

-- Mike, an American living in Norway.

View antmjr's profile

antmjr

262 posts in 1811 days


#6 posted 1383 days ago

Thank you all again (and thank you Mike, your opinion is always one of the most welcomed)

-- Antonio

View William's profile

William

8976 posts in 1469 days


#7 posted 1383 days ago

I am not very good at these technical discussions, but I had to try and throw my two cents into this one because of accuracy (precision, blind luck, or whatever else you wish to call it) problem concerning finger joints, which is what you cut here. From my experience, the problem with finger joints goes much deeper than just the accuracy of one single joint.
Let’s take for example one huge finger joint, just for the sake of argument so that I can easily explain my problem with making sure there is a very minute tolerance level for finger joints.
Let’s say I’m cutting one inch finger joints. I cut one board. Then somehow my jog gets knocked off mark before I cut my second board (yes this has happened to me). Let’s say it gets knocked off by 1/8” of and inch by the time I cut the first finger. Then, without knowing this has happened, I continue cutting the second board at 1/8” off. The first finger then is 1/8” off. Then the second is 1/4” off, then 3/8”, 1/2” and so forth. Now, of course, on a one inch finger joint, I would notice this. I give this extreme example to demonstrate the need for very accurate cuts on a finger joint because even minute differences can create huge problems when making multiple cuts like with a finger joint.
I hope I haven’t confused anyone with my wandering here. I know I’m a little confused.
If this adds to the discussion, then I am proud of myself, because I can’t seem to put into words what I’m trying to say. If it does not, please disregard my statement and I will leave this discussion to the guys who have much better understanding of these tolerances than I.

-- http://wddsrfinewoodworks.blogspot.com/

Have your say...

You must be signed in to post the comments.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics

HomeRefurbers.com

Latest Projects | Latest Blog Entries | Latest Forum Topics

GardenTenders.com :: gardening showcase