Anyone use the 5 cut method to square up a crosscut sled?

[Rorton]

Im making a new sled, this time using 1/2 Birch Ply as a base, and laminated some 1/2 to make the fences - last one was 18mm and heavy!

Ive watched William NGs video on YouTube about his 5 cut method, and im struggling to translate the acceptable tolerance into metric.

Found an online calculator here to just put the results in

https://www.windridgewoodcrafts.com/topics/five_cut/
William does all his calculations in inches, and gets to a point where he is so many thousands of an inch off square and he is happy and makes no further adjustments,

How does that translate to metric, and and what point should I stop chasing perfection and be happy with the result? Am I chasing down to 0.x of a mm, or 0.0x, or even 0.00x

 

[marcros]

40 thousands to a mm if that helps.

 

[alan895]

I was tempted to try that years ago when it first came out but decided against it. Whilst its a good test for accuracy, the sled I was using at the time was made using a reliable square so I didn't see the point in getting it to within fractions when it was already acceptable for my use.

 

[BillK]

I did it on mine though not via that vid, still ended up adjusting slightly by trialling. That said my saw is not the greatest probably not helping things.
My sled is also 1/2" base, the fence is laminated PAR oak with ply lams for the ends. The slider rails are plastic, something that I had in sheet form. Thought it'd help not having them shrink & swell and move.

Regards accuracy it depends what you need really but i do need as accurate as I can get. The adjustments were small fractions of a mil, but with feeler gauges and a pointed stop/reference block (dunno what to call the thing) that didn't make it any harder to do.

 

[Rorton]

I did it on mine though not via that vid, still ended up adjusting slightly by trialling. That said my saw is not the greatest probably not helping things.
My sled is also 1/2" base, the fence is laminated PAR oak with ply lams for the ends. The slider rails are plastic, something that I had in sheet form. Thought it'd help not having them shrink & swell and move.

Regards accuracy it depends what you need really but i do need as accurate as I can get. The adjustments were small fractions of a mil, but with feeler gauges and a pointed stop/reference block (dunno what to call the thing) that didn't make it any harder to do.

Ive done similar with the runners, although made a mess the first time around as I pre drilled holes and used countersink screws, forgetting they would push out the plastic in the slots.

Revisited with flange head screws now and that works better.

I'll screw the fence on tomorrow and see what happens.

From reading online, people seem to be happy if their tolerance is 0.005inch, which translates to 0.127mm, so I guess if I aim for around there, that should be good enough

 

[Rorton]

Managed to get this done today, my fence error tolerance is 0.14mm based on the 5 cut method, converted to inches is around 0.0055inchs which looks good enough for me

 

[XH558]

Been there done that and got to 0.002". Brilliant system and as long as the piece you are cutting is on the same as the pivot point, then Bob IS your uncle.

 

[XH558]

Im making a new sled, this time using 1/2 Birch Ply as a base, and laminated some 1/2 to make the fences - last one was 18mm and heavy!

Ive watched William NGs video on YouTube about his 5 cut method, and im struggling to translate the acceptable tolerance into metric.

Found an online calculator here to just put the results in

https://www.windridgewoodcrafts.com/topics/five_cut/
William does all his calculations in inches, and gets to a point where he is so many thousands of an inch off square and he is happy and makes no further adjustments,

How does that translate to metric, and and what point should I stop chasing perfection and be happy with the result? Am I chasing down to 0.x of a mm, or 0.0x, or even 0.00x

The formula works just as well in metric.

 

[Rorton]

yeah, it just getting my head around is 0.14mm enough, or should it be one more decimal place, but converting what my result was to inches, and then comparing to other websites and other peoples results, it looks ok

 

[XH558]

yeah, it just getting my head around is 0.14mm enough, or should it be one more decimal place, but converting what my result was to inches, and then comparing to other websites and other peoples results, it looks ok

I bought a set of feeler gauges from Halfords for about a fiver. They helped hugely in dialling the thing in.

 

[transatlantic]

I've seen quite a few videos showing the 5 cut method where people spend a lot of time narrowing down the error to a few thousands of an inch. Yet they completely miss the fact that in practise, the fence they're referencing will have far more error. As well as several other variables.

I'd like to see someone make the same cut with different sized pieces and maintain the same error. or make the same cut a week later when things have moved slightly. I think it's a waste of time personally. Get your tools fettled accurately, but within reason, and don't chase silly numbers that will be different from one cut/day to the next.

Same with tracksaws, square guides, MFTs etc. The jigs are milled to silly numbers (4 Microns in one case), but then it's being attached to a piece of aluminium that will have no where near the same level of accuracy.

 

[XH558]

Bearing in mind, that most of us are not working in a temperature and humidity stable environment, I can do nothing but agree with you. It is not a huge amount of time to do the 5 cut once you understand what you are doing BUT unless you are using something like Accoya, it's going to move. We just need to stop seeing who can piss highest up the wall and just get on with it.

 

[Sgian Dubh]

I was tempted to try that years ago when it first came out ...

Ha, ha. You must be, er, extremely old with a vampire like life/death span. Slainte.

 

[moosepig]

Surely the error - the "unsquareness" - has to be measured in degrees, not mm or thou? Without knowing the length of the test cuts, being 0.x mm out is meaningless.

 

[RobCP]

The problem with the 5 cut method is you’re measuring the off cut not the cut.
The blade needs to be perfectly aligned with the mitre slot.
Any toe in and the rear teeth will contact some of the off cut and no adjustment will bring it in.
Any toe out (and you really shouldn’t) will let you get a perfect off cut but all your cuts will be off.

 

[Bod]

25.4 mm to the inch.
First convert fractions to decimal inchs, then the maths is simple.

Bod

 

[harryc]

Surely the error - the "unsquareness" - has to be measured in degrees, not mm or thou? Without knowing the length of the test cuts, being 0.x mm out is meaningless.

That’s exactly what the 5 cut method does you have a test piece and it gives you how much you are out of square over the length of your test cut.

 

[Rorton]

as I understand it, the final number is the distance you need to move your fence about the pivot point to make it closer to square, in my example, I need to move my fence 0.14mm,

The general idea is:

take a board, make a cut
put that cut on the fence, make a second cut

Keep going round till you are back at the start, and the 5th cut, make it about an inch or so (doesn't matter) and keep the offcut. Mark the end of the offcut away from the fence 'A' and the bit nearer the fence B

Measure the 2 ends of the offcuts, and then subtract B from A - note if this is a positive or negative number

Divide the result by 4 (4 cuts, 4 angles) - this provides the error ratio, then divide this number by the length of the offcuts provides the error per inch

Finally measure from the pivot screw on the fence to a point at the other end of the fence where you will make the correction to the angle of the fence

Multiply the number that was the error per inch by the distance of the pivot screw to the adjustment point, and this final number will show you how far forwards or backwards you need to move your fence.

Agree you can get caught up in chasing numbers when if it looks square it should be square, and ive yet to repeat it and get exactly the same result!!

 

[moosepig]

That’s exactly what the 5 cut method does you have a test piece and it gives you how much you are out of square over the length of your test cut.

You're right of course, I'm just being a picky scientist

You can't express the error in an angular measurement solely in terms of distance; the two are not dimensionally equivalent.

So when the OP asks "what point should I stop chasing perfection and be happy with the result? Am I chasing down to 0.x of a mm, or 0.0x, or even 0.00x?" The answer is "it depends on the cut length", which the OP has now explained himself, I've just noticed, in his excellent description of the process

 

[Steve_Scott]

You can't express the error in an angular measurement solely in terms of distance; the two are not dimensionally equivalent.

You are correct that angles are not dimensionally equivalent to lengths (angles are dimensionless), angles are defined as the ratio of radius to arc length, both measured length. Although I’m not overly familiar with the five cut method, it seems this is measuring two lengths (running length and width) and looking at the ratio (m/m) which is dimensionally identical to the definition of angle (radius and arc, m/m).