How accurate is accurate enough?

[ColG]

Ah now SWMBO!!!!!

Don't you just love it when you've toiled for ages on a job and SWMBO says "what about that bit there" instead of looking at the 99% you've done well and commenting accordingly, she picks up on the 1% you've not done so well!!

Am I alone or are others blessed with the same beady eyed woman (or man)?

 

[Alf]

Sounds like you're being too productive, Col. TPsTB here are just pleased to see a completed project at all... :lol:

Cheers, Alf

 

[Anonymous]

You must enter a message when posting

 

[engineer one]

or even the cheaper ones they often sell in lidl, or aldi et al.

obviously you should always work from a reference edge. however when you have a board with a special surface, and you are worried about break out, then it becomes more difficult, but still marking the surfaces is a good lesson.

however, what do you use when working with faced board, or light woods?
pencil often goes too deep into the grain, and i find that masking tape leaves a tell tale when removed.
in the old days, when you could get it, we were taught to use chalk, but now???

about the rod, what is a good length?

paul :wink:

 

[mr]

in the old days, when you could get it, we were taught to use chalk, but now???

paul :wink:

Axminster sells chalk lines if you wanted to use chalk, as does Dieter Schmit Fine Tools.
Cheers Mike

 

[Scrit]

However when you have a board with a special surface, and you are worried about break out, then it becomes more difficult, but still marking the surfaces is a good lesson.

However, what do you use when working with faced board, or light woods?

Here is where a panel saw with a sliding table or a radial arm saw with stops on the fence pays dividends. Transfer the sizes directly to the stops from the rod rather than the work piece. Handle spelching using scrap as breakout blocks. An old patternmakers trick for reproducing lengths without setting a stop - crosscut the workpiece with a scrap piece at the end where the blade goes (i.e. cut the scrap piece), then swap round so the scrap block goes at the stop end - every cut after than will be identical in length to the original piece (or the rod). Works in RASs and chop saws, too.

Similar procedure when chopping mortices - set the first one up then use the stops on the machine, so no need to mark work. If needs be make a rest cut into scrap to check the positions/sizes.

About the rod, what is a good length?

Depends on the job. For laying out a kitchen or bedroom an 8ft strip off the edge of a sheet works well, but obviously too big for making a bureau.....

The trick is always to mark the work as little as possible - less to clean up then.

Scrit

 

[mr]

A rod to me has always meant a full size drawing done on (white) hardboard or similar though I take it rod in this sense is what I have always called a "stick" ie a stick with length markings on rather than a drawing or have I misunderstood something?

 

[Anonymous]

You must enter a message when posting

 

[engineer one]

actually jacob i have used them before, but different people call them
different things. i think the idea of a 6in melamine panel is a good one.
nice background, not too much weight, and as you say easily re-usable.
often i have used a piece of 2x1 hardwood, but these days that seems such
a waste of good wood.

i guess i don't go into childrens toy shops often enough to know that people
still sell chalk

scrit like the idea about using the waste cuts to set the stops makes
perfect sense.

and yes jacob you are right too many people want metalworking tolerances in wood, but if you start with panel goods, then you become
used to being able to work more precisely, but not as you say, more
accurately and a combination of both is the thing to aim for.

paul :wink:

 

[Anonymous]

You must enter a message when posting

 

[Anonymous]

Whilst I found this an interesting thread, I cannot really fully agree with many sentiments expressed here.

For instance, in my Houndstooth DT box in the 'How-to' section of the forum, if I had worked to 1/16" it would have looked absolutely awful and ended up in the scrap bin.

So I say, yes, 0.5 or 1mm out in large dimensions are not noticed by anyone (if you're lucky) but sometimes engineering level precision is the key, particularly when cutting visible joints (DTs, finger joints, through tenons, mitres) or making small components.

Finally, I feel much better about something I made accurately and with better , more accurate construction than available in a shop and I am willing to spend a little extra time practicing my skills before tackling the actual job

 

[Chris Knight]

When he was at Parnham, Makepeace was talking about engineering tolerances (plus/minus a thou) in his woodwork. When you look at some of his pieces it's easy to believe he did work to these tolerances.

(And yes, I do know wood moves - and no, I can't explain it)

 

[engineer one]

another two cents worth. actually tim i agree about accuracy on joints,
but i think that is different from the kind of talk throughout the
thread.

all joints should be as accurate as possible, but then that is surely a
part of using rods or sticks, and then transferring the data.

obviously dovetails more than any other joints should be made against
each other to ensure the accuracy, but as i have said before i think a lot
of it depends upon sharp tools. both measuring and cutting, and also carefully marking the waste, and then ensuring your faces and edges stay in the same relationship.

i wonder how many concerns are actually about chip out etc which occur during the cutting and fettling operations. one important consideration is to
mark out all areas with a knife to break the top surface of the wood,
and then working back toward the line rather than start at the line and work forward.

obviously the defects in the wood have an impact on your work, but you have to work around them.

paul :wink:

 

[Anonymous]

tim????