Tool accuracy

[nathandavies]

A joiner says 'That's near enough'.

A carpenter says 'That will fit'

A furniture maker says 'That's spot on'

Runs for cover ..............

or as a carpenter I worked with once used to say "it'll look alright from my house"

nathan

 

[GK1]

I thought than in woodworking, relative accuracy is more important than absolute accuracy. It's important for a tenon to fit a mortice but usually irrelevant if it's a third of the width or a third plus a millimetre. Like the slightly differing clock cases mentioned.

I've never bought a new woodworking machine so I don't have a rejection option. I was surprised to discover that my 50-years old "top dog" dimension saw has a non user-adjustable fence that's a bit out. Maybe the thickness of cartridge paper along 16". I doubt you'd spot it when cross cutting a simple stile. There's absolutely no play whatsoever when it's attached but I wouldn't use it to square a critical 600mm panel. Instead, I attach a long fence which can be adjusted and tweaked it until I can get it diagonally perfect - within the limits of measurement when you place a steel tape across the corners of a 4 x 4, say a millimetre. But then, I may take the fence off, and replace it, is it repeatable? The limits are in getting the work accurately placed, for example, a piece of timber which might be imperceptibly bowed will present differently on the mitre fence depending upon where you hold it. The piece that is bowed might have come off a good planer but maybe you varied the pressure a bit as you fed it.

I regularly place an engineer's square on my fences and make sure there's no daylight visible, they're easy to adjust. The planer blades are more tricky, I have to use the short block method and adjust until I get the same travel when rotated on the blade edge, I usually call it a day when the variation in travel across the length of the block is within 2mm. I don't know what that translates to in actual cutting circle but it gets me an acceptably flat top plane. My thicknesser table is lifted by a lead screw each side of the table and it's a bit out of level with the block. Maybe as much as a millimetre across 19". I think about dismantling the machine and seeing if I can turn one lead screw a bit to level it but that could be two or three days work, I live with it.

The thing that I've never understood is when I regularly see people saying, square your cross cut/mitre fence against the blade. That's just plain wrong. The table may not travel perfectly parallel with the blade (by design) and how many blades run perfectly true?

I just wish the finished outcome of my woodworking was somewhere near my machine tolerances when I've finished obsessing over trivia.

 

[Mar_mite]

When you come to assemble a job you kinda know if your machines are accurate. If something's amiss you get a feel for it. If I think something's out then I have to deal with it, other than that if it ain't broke don't fix it. Why would I bother to try and put a number on say how square my table saw cuts?

 

[blackrodd]

When you come to assemble a job you kinda know if your machines are accurate. If something's amiss you get a feel for it. If I think something's out then I have to deal with it, other than that if it ain't broke don't fix it. Why would I bother to try and put a number on say how square my table saw cuts?

Zackerly! A running or dimension saw will usually have the fence which runs the complete bed length.
The fence will be set parallel to the saw blade, having a fine tooth, and small set.
The ripsaw fence will have a slight "lead" into the saw blade and the fence should stop parallel to the "root" of the gullet. This is to help keep the set dimension between the blade and fence constant and not cut under the required size
when we have a "hobby" type of saw it is designed as a little of both uses, slight very "lead" into the blade, not enough for the blade to cut on the"up" stroke at the rear. The fence should be set just beside the riving knife.
when the saw cuts true then the fence is the point to square from.
Please do not ask for a plus or minus on accuracy, how big is you're digital sguare costin less than a couple of pints.
The planer cutters are normally set at a slight angle from the parallel to give a "shear" cut, helping the cutting action, which can usually be seen, its how it's s'posed to be!
Should you're set-up be different and working to youre satisfaction then disregard my spouting. As the man so rightly said, If it aint broke don't fix it!. regards Rodders

 

[Rhossydd]

NB you don't seem to have answered your question yourself - we are all waiting, with bated breath!

That's an unnecessary tone in your last sentence.
All I did was ask a question of others, I don't have a question to answer myself.
It's what forums are about, sharing information and comparing experiences. I'm not setting exam questions or have any hidden agenda here, just curious to know what others do.
I've been surprised so few others take a similar interest, but that's not a judgemental statement. It's obvious that precision modern measuring equipment isn't needed to make fine work, but it just makes it easier for me to work to definite numbers.

I'll explain why I've this interest now. I'm planning a couple of big jobs next year that will involve making a lot of drawers that will need to look good. I want then to have dovetail construction and for that I've bought a Leigh jig. For the Leigh to work well it need precisely accurate timber or the joints won't be as tidy as I'd like, or may not work at all, so everything else in the chain needs to be working as well as possible.

To start with I've just fitted new blades to my planer thicknesser. The first set I bought (cheap off eBay) didn't look great when I took them out of their packing, but I continued to install them as I needed a new set. Using the 'moving stick' method of set up, I got the edges set correctly, moving the sticks about 4mm. However the centre stick hardly moved at all. Knowing my out feed table to have no major hollow I removed the blades and when placed edge to edge there was a significant gap in the centre between the two blades, measuring them I found the error to be around .25mm. I considered this to be unacceptable and returned the blades for a refund, although the seller claimed they were within the manufacturer's 0.3mm tolerance. I don't think that's acceptable myself.
I bought another set from APT and the blades met perfectly in the middle, no visible gap and as true as I could hope for. They've installed well and I can now be sure that the lack of precision with the PT's output will be down to my own mistakes to install them perfectly plus any inherent error in the machine itself.
On investigation it seems the thicknesser bed is out by 0.05mm, probably not a problem to me, but the blades are out by a further 0.15mm which increases the error. My next task will be to re-fit the blades until they're as square as possible to the block, rather than the out feed table to see if there's a problem with the out feed table not being square to the block, as I'd thought I'd managed to fit them pretty accurately.

This experience has piqued my interest in what sort of tolerance other people think is acceptable.

Similarly I've just bought my first table saw. Most of the set up is straight forward and easy, but trying to adjust the blade's alignment with the table becomes somewhat random; loosen carriage/tap with hammer/tighten/measure/repeat until satisfied. When should I be satisfied ?

Having now spent a little time working on this after my initial posting here, I've got the blade to within +/- 0.02mm of the mitre slot, which is at the limits of my measurement capability. Coupled with some improvements to the mitre bar and sliding table mechanism, I'm getting cuts as square as my Moore & Wright engineers square will measure. No problems there any more.

So there's my numbers. Do you have anything definitive to add to this and help me ?

 

[Jacob]

..... Do you have anything definitive to add to this and help me ?

Most people would do a run and look closely at the product before crawling all over with dial gauges and straight edges. If you start by looking for faults in the machine you will certainly find them.
NB don't buy cheap planer blades from ebay - or if not well finished, then take them to a saw doctor for sharpening

Also - it's a basic feature of woodwork that there will be errors and there are a whole set of techniques for reducing them, or contriving to have them out of sight. A good joiner covers his tracks. You will still get errors even with perfect machines, due to a host of other factors. This is why you need to run a process so you can see where/what goes wrong and decide how to deal with it.

 

[Rhossydd]

If you start by looking for faults in the machine you will certainly find them.

Yes and the smart thing is to correct them.
One afternoon in total to turn an eBay bargain into a superbly performing table saw is effort very well spent in my book.
Similarly making sure an old machine is delivering top quality results throughout it's life is really no hardship either.

Or you could just struggle on with poorly performing kit wasting time and money trying to make good second rate output.

 

[Jacob]

If you start by looking for faults in the machine you will certainly find them.

Yes and the smart thing is to correct them.

Not necessarily. It depends on whether they show on the product

One afternoon in total to turn an eBay bargain into a superbly performing table saw is effort very well spent in my book.
Similarly making sure an old machine is delivering top quality results throughout it's life is really no hardship either.

We all do that as a matter of course. But not necessarily by "quantifying" anything.
In fact there is a persistent "quantifying" theme in a of woodwork toolery - think of all that bevel stuff and "flatness" issues :roll: . It may seem attractive to engineering types but can be a will o the wisp as far as the end product is concerned - where the only thing that counts is visible error. You can get by with all sorts of variations as long as they aren't seen.

 

[Rhossydd]

You can get by with all sorts of variations as long as they aren't seen.

Sure you take that approach if you want to, but for me I don't want the making process to be unnecessarily tedious, wasteful or time consuming.
There's no pleasure or profit in having to continually try and work round problems that could be sorted once and for all earlier in the production process.

 

[nanscombe]

No wonder some people like the accuracy and repeatability of power tools over hand tools.


For me, who's not selling anything, it's usually close enough for government work :mrgreen:

(idiomatic, humorous, disparaging, slang) not worth investing additional time on perfecting this thing

 

[Jacob]

You can get by with all sorts of variations as long as they aren't seen.

Sure you take that approach if you want to, but for me I don't want the making process to be unnecessarily tedious, wasteful or time consuming.
There's no pleasure or profit in having to continually try and work round problems that could be sorted once and for all earlier in the production process.

"once and for all" is the weakest part of the argument! Machines need continuous monitoring and maintenance. A craft skill in its own right.
PS obviously you set it up as best you can to start with, but there's nothing to say it'll stay that way.

 

[Rhossydd]

"once and for all" is the weakest part of the argument!

Making mechanical modifications to the machinery to improve performance IS a one off fix.

Machines need continuous monitoring and maintenance.

So now you're saying that people do need to keep measuring and assessing the performance of machinery ?
Which is where I started; asking people if they use modern accurate measuring tools to do that and if so what tolerances they worked to and achieved.

Still nothing to actually add to the facts then jacob ? I'm beginning to see why you have a certain reputation here.

 

[Jacob]

"once and for all" is the weakest part of the argument!

Making mechanical modifications to the machinery to improve performance IS a one off fix.

Machines need continuous monitoring and maintenance.

So now you're saying that people do need to keep measuring and assessing the performance of machinery ?
Which is where I started; asking people if they use modern accurate measuring tools to do that and if so what tolerances they worked to and achieved.

Still nothing to actually add to the facts then jacob ? I'm beginning to see why you have a certain reputation here.

If you bothered to read the replies you would see that people have replied to your questions. You don't seem to understand them however and have taken it upon yourself to be critical and sarcastic.

The main thing you don't understand is that assessing the performance of a woodwork machine (etc) means looking at the product for visible errors (and occasionally measuring them too).
You seem to be fussing on about your machinery without having run anything through as a test. You don't know what the consequences of your cheap planer blades and inaccurate thicknesser settings would be, as you haven't tried them. You've been measuring the machinery instead, which is a bit, er, stupid!

 

[Mr T]

Hi

Unusually I find myself agreeing with Jacob here. As I posted about three pages back, one of the skills of woodwork is knowing when you need to be accurate and when it's not so important. Part of this is achieved by judging by results. Machines have a habit of unsetting themselves, my Felder saw spindle has a particularly unruly rip fence. When setting up a machine to take a batch of cuts I'll do a trial run first to check performance, if it's not up to the accuracy I need I'll set about adjusting it.

Having lots of super accurate measuring devices and setting ones machines up to super accuracy may appeal to those with a slightly obsessive engineering background but the "judging by results approach" also produces good work.

There's always more than one way to skin a cat!

Chris

 

[Rhossydd]

You've been measuring the machinery instead, which is a bit, er, stupid!

No. I've been measuring the wood that comes out of the machines as well as the machines themselves.

 

[Mar_mite]

I enjoy working with wood, that's why I'm a member of a forum like this. But my machines are there for one purpose , to pay my mortgage and put food on my table. I can't get bogged down with measuring this or quantifying that. If a job goes together square and looked right that is good enough.

 

[bugbear]

You've been measuring the machinery instead, which is a bit, er, stupid!

No. I've been measuring the wood that comes out of the machines as well as the machines themselves.

Jacob does tend to castigate people for crimes that he's just made up on a whim, with no evidence.

BugBear

 

[Jacob]

You've been measuring the machinery instead, which is a bit, er, stupid!

No. I've been measuring the wood that comes out of the machines as well as the machines themselves.

Jacob does tend to castigate people for crimes that he's just made up on a whim, with no evidence.

BugBear

Bugbear does tend to troll around after me making pointless comments because he hasn't got anything interesting to say, ever, but doesn't like being left out.

 

[DTR]

Jacob does tend to castigate people for crimes that he's just made up on a whim, with no evidence.

BugBear

Bugbear does tend to troll around after me making pointless comments because he hasn't got anything interesting to say, ever, but doesn't like being left out.

And you both tend to derail threads by baiting each other.

Sigh.

 

[Lons]

Strangely, I'm sort of in agreement with Jacob as well :?

My machines are set up as and when I need them to produce and I monitor them by the results I obtain when using them. I'd love to have incredibly accurate machinery but not at the expence (if even possible) of my precious little time available for my hobby.
Perfect accuracy isn't at all necessary for me as I would always machine slightly oversize and finish off by hand to the best of my ability. Straight from the machine production would put me in the factory / Ikea category imo.

All I'm interested in, despite having the measuring equipment available, is that my blades are sharp, fences are true and the heavy work only is comfortably produced by the machines - the finish being down to my skill or lack of!

Now - the fact you act and think differently is up to you and surely part of the satisfaction you gain from your activities which makes it right for you but not for others. No need for anyone to criticise anyone else on the matter and perfectly aceptable for you to pose the question to guage other members views.

Bob