Checking Cabinet Squareness

[George Bokros]

How close is acceptable when checking across corners of a cabinet? Example on a current project across one pair of corners it measures 45 1/4 diagonally and across the other pair of corners it measures 45 11/32 diagonally or 3/32 more, is that good enough?

Still learning so bear with me.

George

[Shawn Pixley]

I don't build many cabinets, but build furniture. That wouldn't be square enough for me. With a diagonal dimension of ~45" it is a pretty big cabinet. Assuming the stock / sides are all flat and square, you might be able to make it work. However, I expect you will find you will constantly be fighting the elements out of square when installing slides, kickers or whatever.

Is it glued up yet or is this dry fit?

[johnny means]

I would put that at the outside limits of acceptable, but good enough for most applications. Of course, there are times when that much error would be a problem. Things like inset or flipper doors, for instance, can be really hard to fit if your carcass isn't dead square.

[Jerry Miner]

I'm with johnny on this one---you are at the outer limits of "close enough". For me, I'm not happy unless the diagonals are within 1/16"---1/32" is better. (And I stop fussing at 1/32)

But "close enough" is largely up to you, and can vary depending on the project. Overlay doors are more forgiving than inset, etc. ...

[ian maybury]

Seems like the variation in diagonals can pop up in different ways. As a result of the sides not being of equal lengths, or when they are but the assembly is not square, or as a combination of both.

The effects can be quite different - if for example two of the four corners are accurately 90 deg then the resulting rectangle will have three sides square to each other, and so shelves, door widths etc will all end up the same length - and only one side will be out of square. (but more so for a given difference in diagonals)

If on the other hand the variation is the result of all four sides being marginally out, then nothing much will be parallel, and pretty much everything would require custom fitting - and could look bad if the error results in two lines close to each other ending up out of parallel. A European style cabinet with its need to exactly align doors with cabinet sides will highlight errors unless the door is adjusted to suit, while face frames are likely to be more tolerant.

As an example a triangle with sides in the ratio of 3:4:5 is right angled. So in that particular example the diagonal is 5. Whatever ever the difference in the diagonal is the error in the longer side will be roughly 4/5 of that, and the error in the short side 3/5. Meaning that the variations are not trivial.

It's not my space, but a local kitchen guy would say that the biggest labour saving and quality improvement he ever got in his business was as a result of his spending a lot of money on a big CNC router to cut out his parts. With everything spot on accurate the whole thing goes together like a Lego set - everything fits, no forcing to close gaps, no adjusting stuff.

My personal instinct (based on limited experience) is that time spent in making sure that the parts are accurately to size and the assembly as near to perfectly square as is feasible is always a good investment - one that pays off later in the job on labour, and in quality of output. Ditto the use of squares and the like to check corners for squareness....

ian