Plane sole flattening
joel
>In the past I have been an advocate of scraping a sole of a plane to surface plane but frankly it's time consuming and too finiky for me.
Lapping the entire surface of a plane always runs the risk of getting a convex surface, expceailly from side to side and I have seen that problem more often than not.
My heavy glass surface plate got broken when moved.
anyway for these reasons I have been experimenting with
"the silly way of flattening a plane"
and in the two planes I have done it too it seems to work rediculously easy with no special tools or jigs.
basically here's the method.
1) retract blade a little.
2) wipe off plane sole
3) wipe of straightedge
4) hold plane up to light and note gaps. Check serval places, also from left to right in several places. get a sense of what needs to be done.
5) cradle plane in lap.
6) with small squares of sandpaper - sometimes using a backing block, sometimes not, sand away the high spots. around the mouth use a backing block to make sure you don't round out the leading edge of the mouth (very important)
I've been using 220 grit 3X paper because I have some - I might try 180 then follow up with 220
7) repeat steps 2-6 until the plane is flat enough for you.
clean plane properly - metal grit with get in the mouth and you want to shake it out so it doesn't contaminate the straightedge readings and also when you try the plane after you are done you don't want metal filings on the work.
Elapsed time to do a pretty decent #4 to very ver decent - 15 mintes including setup time (tearing sandpaper to 2" squares).
Anyway - I haven't done a long plane like this yet. I don't know if I can get a plane surface grinder perfect but it seems to be very easy to do, imporvie performance a lot, and best of all not require any real effot to built jigs or setups. If I didn't have an 18: straightedge I Would use a starrett rule from a combination square (check for burrs first).
Anyway - it's very similar to bugbear's method, only with sandpaper, althought files would work faster with a more difficult case. A straightedge seems to work fine over a surface plate and it's easier and more accessible. Snadpaper is very forgiving.
I haven't tried this on a plane that is way way out - maybe .005" but I thought I'd toss out the method because it seems to work and with somefeed back I am sure we can arrive at a conclusion on how the idea rates and if anyone can suggest improvements of fatal flaws.
One of the main appeals for me is that it doesn't require special equipment and it's easy to teach. I think a straightedge will give a perfectly descent measurement of flatness provided you check in a bunch of places on the sole. but of course I haven't tried a long plane yet so I don't know what heppens with that.