I think repeatability is more important than accuracy in most cases. if one rail of a door is significantly different than the other rail, yeah it will be out of square if you perfectly align corners. Usually, it wont matter that much if the rails are 24.000" or 24.015", as long as both are the same.

Inset doors need to be way more accurate than overlay. With overlay, you may not notice even if they are a bit out of of square! So it really depends on the application.

Dovetails, box joints, dadoes, etc, I try to get as accurate as possible. Missing by small amounts can look really bad. The flip side is... get them too exact and assembly becomes an issue.

So, I think it really depends on the specific project.

That said, I cut several pieces of 3/4 plywood to 3 inch lengths the other day in response to a thread on this forum. Using the Incra jig, I was able to get 5 pieces cut at +- .003". Pretty good for most tablesaws. However, I can feel the difference in them when stacked together. Close enough?? Your mileage, application, and opinion may very. Doesn't make either answer right or wrong!

 

Hi Bill.
Have to agree that wood "moves".
That's why I prefer to use plywood for larger areas.
Also interested in your comment on wood measurement technique, in that, while some measure and scribe their wood with a magnifying glass and knife edge, I am from the old school where my eyes and a pencil width are adequate for almost all of my wood measurements.
Generally more concerned in the setup of my saws where a true right angle is important to make everything fit.
Mark

 

I am a machinist by trade, woodworking with machines about 4 years.
In the metal working, one important thing, is not to use super precision where its not needed.
An example- two plates held together with four bolts.
Semi precison layout "scribing lines" should get the bolt holes at least .005" from each other, alittle inacuracy using a drill press added- the plates will still bolt together just fine, using a drill 1/32" oversize of the bolt diameter.

The other side of the coin,
An end cap of a power transfer case, lets say a hydraulic pump body.
Along with bolts, there are four dowl pins that have a slip fit for the cap of .001", and a press fit in the body .001".
That becomes a serious matter, where as, not only the locations of the dowl pins becomes important, but also the hole sizes for the press and slip fits for the dowl pins.
There, a more specialized machine is needed, that better not be out very much more then .0001"- one hundred millionths.

Yes, I have seen folks say you dont need better then 1/64" accuracy in woodworking.
But, they never say just for what- a chicken coup, or a fine jewlery box?

At any rate, I do what I can adjusting my woodworking machines for the best I can get. But not fret over the chicken coup projects.

 

One thing you didn't mention, Bill, is the traditional woodworking way to overcome many fitting issues: instead of spending hours tuning-up cheap tools to make them more accurate it is often just simpler to get out a hand plane and use hand and eye to achieve the result. You can't do that with steel. As you say there is a time and a place for super accuracy - but there is also a time and a place to use hand tools. The pity is that there are a lot of people who call themselves woodworkers and yet never master the true basics

Regards

Phil

 

Some members of this forum had almost talked me out of the importance and/or value of what Bill has brought up in this thread. Certainly the issue of application is important. In my case, my knowing that I can to close tolerances whether I need them or not in each project im.portant. I put emphasis on "in my case". I for one, like the time I spend on fiddling with the machinces to see how well I can make them perform. To be clear here, I do not mean to imply that I have the skill or knoe how to actually modify the machines, I am referring to learning and/or acquiring the skills to get the most out of them.

It's a personal thing of course. Those of us that do like to deal with getting tolerances as close as we can are not correct and others are wrong and vise a versa. Both schools are right, it's a just a matter of "what rings your bell I suppose". I for one have really enjoyed this thread and am looking forward to reading more posts to it.

Jerry

 

Accuracy also depends (IMHO) on the wood you are using. 1/32 in soft pine is not bad but in purple heart or such it is unacceptable. Pine "moves" a lot more than hardwood.
Just me, I guess.

 

Real-world example, a typical cabinet door 12" wide, 30" tall in a 24" wide cabinet: Let's say my crosscut on the rails is 1/128"(.008") out of square in 2-1/2 inches of width. Now that 1/128" is negligible because wood moves, right? I clamp that rail tight against the 30" stile, that .008" deviation from square is now 12 times that deviation at the end of the stile(30/2.5=12) so now one side of my door is .096"(3/32") out of square. But wait, there's more! The other side of the door is out an equal amount so I now have a door that's racked 3/16". Now I have the task of trimming the doors on each side from nothing to 3/32" just to square them up. So I finally figure out how to do that but now my perfectly sized doors are 3/8"(3/16"x2) short of the cabinet width they're supposed to fit into.

But that's OK. I'll just tell my customer "wood moves". Right.....

Your thoughts?

Bill

Uhhh, unless the 2 rails are not the same length, that should not be. You may have a joint that doesn't meet across the width of the rail. But unless clamped and glued out of square, the door should be square! This points out the need for consistency and craftsmanship as opposed to accuracy. There is a difference between precision and accuracy. 12.00003 is quite precise, but may not be a very accurate at all!

The door example calls for a dry fit and check before glueup, The error should be found before there is a need to shave a finished door! As Phil properly points out, at that point a hand plane can easily and quickly fix the "error".

This is one of the problems common in woodworking today. We expect our machines to eliminate the need for craftsmanship. That is only possible to a certain degree. Even if cut on a CNC, if parts are not assembly properly, errors can still occur!

 

Amen, Phil and Duane!
Take a critical look at any good quality, old furniture. No digital or carbide anything used in their production, just a lot of skill and craftsmanship.

 

i agree.in japan when i meet traditional shrine woodworker who are specialize to repair hundreds of year old shrine.they have really very great skill.you can see the bird mouth joint or other japanese joint.
but again their skill is pass generation to generation.there is alot difference between hobby and professional woodworker.

look at this video they are planing wood to thousands of inch
https://www.youtube.com/watch?v=v3Ad6tBdLbM

 

Accuracy - tough subject for many wood workers including me who are hobbiests and who do the best they can with the tools they have. If for instance, am doing something that will be a decorative item in the home, try, if making multiple pieces, no more then a thumb feel difference which is in the low thousandths - doubtful if its as much as that .008 or 1/64th. If it's a piece of molding to case a door frame or window to be painted, close enough could be good enough as there is always painters caulk:fie: - and no, this is NOT how I want to do that job - and mine is a lot better then the 'pro' who originally molded the house in 1970. Then there is that kitchen that I finished out with stained oak where close enough couldn't work.

My thinking, such as it is, there are places where the tolerance can be as wide as that proverbial barn door, then there is that area where there is virtually none. Only the craftsman can make the determination as to what 'works' where.

 

I worked on a job once upon a time, where there was a carpenter who was well past the first flush of his youth, as indeed most of us here seem to be. Whenever he signed out a pound of six-inch nails in the morning, the wags joked that he was doing carpentry that day. When he took out four-inch nails, it was considered to be joinery - much finer work ;-)

I have learnt a few things here and elsewhere, including from Norm Abrams of course, about setting out work, in the never-ending quest to improve the quality of fit and finish. First, I don't depend on the scales fitted by manufacturers to saws etc, for example setting a tablesaw or bandsaw to be absolutely square to the table. I use a steel engineer's square always. Next - use a fine pencil - 0.5mm leads or smaller. Next - use a vernier caliper, either digital or traditional pure vernier. Next - use the same ruler as far as possible when doing a job.
Develop a work sequence where each step negates any error in the previous step, and if possible sequence the steps so that redundant setting-out marks are machined out as you go. Mark everything with a very simple system as if the next step is to be completed by an idiot, most of the time it is.
Routing/thicknessing/sawing - 'sneak up' on the finished dimension. Don't try to get there in one go.

jc

 

The more I work wood, the more I find its like metal.
Like this, very soft wood leaves a somewhat fuzzy cut edge, very hard woods has more of a smooth cut edge. The same with steel, vrs soft steel, and harder alloy steel.
The rate of feeding the work has the same effect also.

Another observation, with cutting tool material.
Most machinist will agree, its possible to get a high speed steel cutter sharper then a carbide cutter.
Though carbide has much longer life, at times steel cutters work better.

I noticed this with some chisels I have.
I can get a hundred yr old plain carbon steel chisel edge, just abit sharper, then a chrome alloy steel chisel. Though the alloy steel edge holds up a little longer in use.

There are still woodworkers using carbon steel power saw blades for those reasons.

 

There are still woodworkers using carbon steel power saw blades for those reasons.

That's a biy like saying there are woodworkers who don't use hand tools but work direct from the saw or planer (jointer) without any further planing, sanding or scraping.....

I try to do my cuts and measurement to the millimetre whenever (comfortably) possible. I try to avoid tape measures because of the wobble on the hook end.

Then, Mark, this will fry your noodle: I was taught not to measure excessively on projects at all. So whether you are making a sash window or a frame and panel door you only measure stuff at the very beginning and transfer them to a rod (story pole). Thereafter dimensions are transferred either by direct marking off the rod or by use of wing dividers. Same goes for taking measurements off at site of stuff like kitchens - the tape should only be a sanity check. In any case buy a box of ten mid market tapes, like the Stanley Fat Max or Lufkins, etc, and compare them side by side - you'll only get two or maybe three which are identical end to end. So if you depend on a tape for measurements use just the one tape for the entire job from start to end and you'll have less problems

Regards

Phil

 

Its probably more important to make parts that go together that should be same length cut with stop blocks.

 

Don, I am also a machinist. I totally agree with you about, tooling, cutting metal, and cutting wood. So many times I hear, "you have to know what you can get away with."
Either a machinist, or woodworker it does not matter, they are saying to me, I just do
not want to take the time to do the best that I can. That drives me nuts, but I still strive to always do the best.
Wood moves, we all know that. That should not be the the excuse to be sloppy in our
set-up or tools. To me, get it right so when it comes to assembly, it will fit. Clamps are
for holding wood together until the glue dries. They are not meant to "pull" the joint together. For many years I have said, "I like a challenge, but not a fight." Just my opinion.
Ellery Becnel

 

Clamps are for holding wood together until the glue dries. They are not meant to "pull" the joint together. For many years I have said, "I like a challenge, but not a fight." Just my opinion.
Ellery Becnel

True, but if you clamp it out of square, it will be out of square after the glue dries. That "doing your best" bit applies to all aspects of the project, not just machining! If not, you can have the most accurate machined parts ever made and still produce junk!

 

The expression is that there are several ways to "skin a cat", but nobody I know of skins actual cats. The expression actually refers to skinning catfish and there are numerous ways to [skin cats] do most projects.

For almost anyone, if we have multiple similar projects to construct, each project will take progressively less time. This is because we learn from trial and error. Many methods to get a project completed to the customer's satisfaction are usually acceptable, but we need to all be willing to learn from the experiences shared herein by fellow members.

Accuracy is important, but experience teaches us where to focus tightly and where that focus is of little importance. Never will woodwork be able to be done with the precision of metal working. I've never heard of cylinders, pistons and piston rings made from wood, for many different reasons. Absolutely beautiful wood projects were built many years prior to "super-accuracy measuring devices", yet it would be close to impossible to find fault in their fit or function.
Otis Guillebeau from Auburn, Georgia

 

Its funny about wood and metal in these ways.
Metal also has grain, they way it is flowed in process.
Metal also warps and twist from the cutting stress, and working harding from most machining processes.
Metal most always has built in stress, from its forming, only to be relieved by machining, causing twist and warp.
Most problems can be reduced by machining the metal, perhaps from all sides during rough cutting operations, if possible let it sit for some time, bang on it with a hammer, and at times heat it up, and cool it down before finish cutting.
Seems to me I have some of the same problems with wood, usually to a worse degree.

A More experienced woodworker gave me this tip,
When working a cupped board on the jointer, cut the convex side first. This is opposite of conventional approaches.
The reason being, this will help break the built in stresses in the board. I found this does work for me.
To help with that, I use a hand plane to start a flat on the convex side, to better guide it through the jointer.
Woodworkers may not know this, some of the most precision surfaces on cast iron, is created by hand with a scraping tool, by very skilled people.
I do think it is the same with a skilled woodworker, using hand planes and scrapers.

 

Artist vs. Artisan

I recently had a discussion with myself about this very question of accuracy and tolerances while working on a new little wooden automaton. [Of course I talk to myself, who else is such an excellent raconteur? :happy:]

I was trying to get both sides of the wings for this little bee to exactly match. Then it dawned on me that there is a time to be an artisan and a time to be an artist. When I'm building cabinets and such I need to be an artisan and make everything as perfect as possible. But when I'm building little arty projects, I need to be an artist and just let the creative urge flow. Little differences and imperfections add to the charm of art projects but detract from the serious ones. Life is simpler since myself and I have agreed on this point.

 

Yet woodworkers continue to buy rulers that are not marked correctly and squares that are not square

Bought a combination square a few weeks ago...wasn't square. Made in China....aaaarggh. In my moment of fury I tossed the handle bit out and kept the ruler part but looking back now I may have been able to return it to the store for an exchange or refund. The handle bit had a slight upslope on the last cm of it which, if pressed against a surface a certain way, gave measurements that were off. Oh well.

I try to do my cuts and measurement to the millimetre whenever (comfortably) possible. I try to avoid tape measures because of the wobble on the hook end.

 

i don't think you can do better than this. ZEN man ZEN.

Kezuroukai, Planing Competition Finals 2012 - YouTube

jack
English machines

 

"That's a biy like saying there are woodworkers who don't use hand tools but work direct from the saw or planer (jointer) without any further planing, sanding or scraping....."

I certainly never implied anything of the sort.
However, in my experience, the better the machine cut, the less time/amount of hand work needed in finishing.
It also could be said, many run blades and cutters way past time for sharpening, making matters worse.

You know its just mind boggling that the English just couldnt match metal working machines with the US.
This Monarch 10EE being only one example.

Yes they tried to copy it, but just couldnt do it, tsk tsk.

 

"That's a bit like saying there are woodworkers who don't use hand tools but work direct from the saw or planer (jointer) without any further planing, sanding or scraping....."

I certainly never implied anything of the sort.

You know its just mind boggling that the English just couldnt match metal working machines with the US.

Really? What relevancce has that to this discussion, or are you trying to justifty your mildly xenophobic rant? For that matter I'm BRITISH - not that I think you'd understand the difference.

I've rarely come across anyone doing serious work who'd look even at a steel saw blade today simply because they just didn't last anywhere near as long and the surface quality of cut was often much poorer than a carbide tipped blade. Often people didn't know how to or didn't want to spend the time side-dressing the blades to get an even set - a process which requires the operator to stone the blade whilst it is running and something I'm glad I no longer have to do. I started at a time when we were just moving from such blades to carbide tipped. I don't know of a single professional woodworker who'd ever go back to them because TBH they weren't that wonderful to use and had a very low life. Most woodworking requires further processing after sawing to get to a finished surface sio in reality the quality of surface finish of a saw cut isn't the most important factor

Regards

Phil

 

Wood moves, plywood panels move, and even metal moves. So how do you deal with this movement in the materials?

Try to cut all of your like parts at the same time. Cut them to width at the same time. Cut them to length at the same time. Cut all your miters at the same time. Store all of the parts together while you build the project so they are subjected to the same environmental factors.

If 2 like boards are cut the same width and the same length at the same time and are subjected to the same environmental factors when stored, then when 1 board changes length or width because of environmental factors the length or width of the like board will change also.

There are things that do make differences in the movement of wood like knot holes, straight grain versus cathedral grained areas, sap content ...

I feel the most important thing is to be as accurate as possible with measurements and machining but don't think you need wood to be with in 0.0001". Make sure your machines are aligned and checked now and then to make sure they are still in alignment.

 

Not much new under the Sun...
Parks Woodworking Machine Co. - History | VintageMachinery.org

 

Wow! Tough subject. Guns come to mind here. I believe its the Ak-47 that is the preferred wepon over the M-16. It's Sloppy tolerances allow it to be shot when dirty, while the M-16 jams. This rivets on a mig fighter jet are eyed up. The plane flies fine. As a North country, back woods carpenter, I am expected to do reasonably good work in a reasonable amount of time at reasonable cost. This requires compromise. When my wife asks for a book shelf, she doesn't want me out in the shop forguY the next 6 months. Woodwork for many is a means to an end. Not an end. Of course accuracy is extremely important, but for those who need to make a living and with many other things to do, we can't work to far right of the decimal point.

 

Really? What relevancce has that to this discussion, or are you trying to justifty your mildly xenophobic rant? For that matter I'm BRITISH - not that I think you'd understand the difference.

Gee phil!
I was only pointing out that the "British" made crappy metal working machines.
But, it is remarkable with such sub-standard machine tools they made good saws.
Perhaps they used American machines.
But, you would have zero understanding there.

Your rants on "British" saws are so great and US saw are so bad, are irratating, as at least another pointed out, but you certainly are a "Professional Irratator", though, I would not expect you to understand that.

Perhaps it would be best to back off a little.

On the carbon steel saw blades, dont use them myself, on another forum, others do use them for the reason I pointed out in that post. And thats it. I do not know if they are "Professional", most likely hobby.
Perhaps,get off your "Professional" pedestal while your at it.

 

Gee phil!
I was only pointing out that the "British" made crappy metal working machines.
But, it is remarkable with such sub-standard machine tools they made good saws.
Perhaps they used American machines.

Don you really need to get out more.:big_boss: I had heard the US could make some stuff.


Not like boeing is getting there jet engines from china now are they. and let face it there not made in the USA.:lol:

BTW it was me who said Mercan saw are sub par not Phil. but they sure are cheep money wise that is.

time for a :lazy2:




jack

English machines

 

Well if nothing else I've actually seen a word - xenophobic - used by a real person, not an author filling space for word count and royalties.

 

Perhaps, you should re read some of the bashing. Its irratating. How dagerous the American equipment is and such

Just as I pointed out American vrs British Machine tools, didnt like that I would suppose.

Now things are made with CNC equipment, by lower skilled, low paid people in Asia.

There are CNC saws now, the blade adjust, the fence moves without human hands, wood can be fed in to them automatically.

So, most everything is made that way now, less demand for a skilled person running a Wadkin saw, or a Monarch EE lathe. But, still a little out there to do.