Perpendicularity and flatness of hand planes
For a hand plane to work precisely, it should have a flat, straight sole. Moreover, the soles of planes that are also used for joining should be at right angles to the vertical sides of the plane body. "Joining" in the context of planing refers to machining the edges of a board with the aim of obtaining a right angle to the face of the board. This step is indispensable for precise, neat bonding of wooden surfaces.
People who buy our planes quite often complain about the flatness of the soles and their perpendicularity to the sides of the planes. When we come to check these planes we find that in most cases the complaints are unfounded because the planes are perfectly alright. How do these discrepancies arise?
1. There is a popular view that the vertical angle or the sole must be absolutely accurate. This is only theoretically possible: depending on the resolution of the measuring equipment you’ll always get a deviation. More important is that for practical purposes the deviation is negligible. What matters are tolerances specified by the manufacturer that must be adhered to in production.
2. The precision requirements of metal processing are transferred unthinkingly to woodworking. But of course, wood is an organic material that is quite capable of changing its form even as you’re busy working on it. If you cut a straight board along its length you may well end up with two bent halves, simply because of the stress relief that occurs when you cut the wood. The same happens with metal, but the change is nowhere near as noticeable as with wood.
3. Inadequate measuring methods – for example using a ruler with a bevelled edge –often give spurious results or increase the apparent deviation. Occasionally we are sent a photo of a plane with its sole placed on a bevelled straight edge showing a distinct light gap. The sender suggests that this showed a substantial deviation – however without sending precise information about the width of the light gap. A ruler with a bevelled edge or a straight edge amplifies deviations of even a few thousands of a millimetre as can be seen in the photo below. This is just not realistic.
When we performed our measurements we found that the deviation from true flatness was less than 1/25 mm. Would you have thought that in view of the picture? Our measuring method is described further down.
4. Another inadequate measuring method involves the use of an inaccurate try square on a surface that is far too small so that only half the stock is in contact with the reference surface. The photo suggests a sole that is far away from forming a right angles with either side of the plane body.
A check at our premises (see below) showed that regardless of the test set-up, the 0.04 mm thick feeler of a feeler gauge could not be inserted anywhere between the reference surface and the sole of the plane. 0.04 mm is equivalent to 1/25 mm!
This is how we check the perpendicularity of the sides to the sole of the plane and flatness of the sole:
1. A calibrated, highly accurate granite slab
2. A calibrated high-precision square (accuracy GG 0)
3. A feeler gauge
The set-up is the same as recommended and used by leading plane manufacturers. The important thing is that the sole itself must not function as reference surface; instead a neutral base must be used; you could call it an auxiliary reference element, that in fact needs to be extremely accurate. Another consideration is dirt; everything needs to be squeaky clean to avoid the results being influenced by dirt or dust. Then a calibrated high-precision square (not one with a bevelled edge) is placed on the granite slab and moved up against a side of the plane. If you can’t see a light gap, the accuracy is spot on. If you can see a light gap, you need to measure its width at the open end with feeler gauge. A gap smaller than a 0.002 inch (0.05 mm) feeler implies an accuracy within the tolerance range specified by manufacturers such as Veritas. If even greater accuracy is desired – not necessarily practical in woodworking – it is up to the user to provide it himself by the usual lapping methods. For Juuma planes the light gap at the open end may not exceed 0.15 mm.
To measure just the flatness of the sole, you put the plane on the granite slab and then attempt to slide a feeler between the plane and the slab at various places. The tolerances are as above (these apply to planes from Veritas, on cheaper planes higher tolerances will need to be accepted).
Now for the measurement results for this plane:
Left side from the front: no light shining through, just perfect!
Right side from the front: no light shining through, again perfect!
Flatness of the sole: a 0.04 mm feeler (equivalent to 1/25 mm) fails to slip in under the sole anywhere along the plane.
Conclusion: there’s nothing wrong with this plane.
We generally recommend that you carefully check your measuring equipment and eliminate any error sources. You probably haven’t the same resources to hand as we do for performing these checks. But quite simple methods should be sufficient to approximate the test set-up described above. Use a glass plate instead of the granite slab. At least one good, verified high-precision square should be part of your toolkit. You’ll need a feeler gauge – but this is not an expensive item.