Thread: Hand Plane fettlers should read this

Originally Posted by Michael G

Just dropping in from the metal work forum because this is an interesting discussion (to me anyway)

To pick up on a point that Greg Q suggested in post 22, the plane sole will heat up from use but if the work being planed is narrower than the blade of the plane then the heat will be applied in a strip along the sole. As the sole is not a uniform section but varies with ribs, bosses and even the frog being bolted on, the plane will distort anyway.
Even if the work is wider than the plan so that the whole plane is in contact with the work with a 'uniform' friction heating effect, the varying section thickness will ensure non-uniform distortion. Of course, as soon as that happens there will be more pressure in some areas so the frictional heating becomes non-uniform...

From an engineering point of view it's sounding like a thick uniform section is needed with a temperature control mechanism built in (eg chilled water circulation) if you want to get (and stay) flat to a micron level.

Michael

Michael,
Welcome to the Dark Side. The temperature variations i'm sure effect the flatness and when we get the the cable for the automatic compensator (pressure and temperature) for the interferometer I will have ago at some actually measuring flatness and straightness after planing and see if I can reliably measure the distortion by heating effect as it applies to the planes. (I think that might get tricky :S)

Ray and I have been talking about the results we are seeing with the laser I can only conclude so far that with a Stanley style plane: how you plane can have a significant effect on the shape of the sole; and that is only looking at twist not deflection as well.

We do not have any equipment for measuring torque applied while actually in use (if any one can put a number on that it would be great)

The Lie Nielsen #4 in bronze so far has the best resistance to twist yet seen, about twice as good as the Bailey.

Thanks for the welcome Josh
20 years ago I used to push wood planes but currently I do more with metal, so there is not as much work for them in my shed. Maybe in another 20...
If I had an FEA package handy I could easily show the distortion due to heating but I think that you will be struggling to show that with the interferometer because of the small size of the planes.
Figures for torque will vary depending on the person, grip, action and even working position. The best way to measure this would be with an instrumented set of handles (strain gauges etc). You might be able to infer a torque if you mounted a dial gauge from the back of the plain indicating to a point on the front (assume the rigid mount does not move therefore any movement is due to twist). Trouble is you start getting observer effects changing numbers. More thought required.

Michael

Originally Posted by Michael G

Trouble is you start getting observer effects changing numbers. More thought required.

Michael

I suspected this might happen when we started looking for ultra small observations. What we've bumped up against is the planing mechanics problem of Brobdingnagian's plane, which plane is both twisted and not, until the observation is made.

Originally Posted by mic-d

which plane is both twisted and not, until the observation is made.

We had a cat around here with a similar problem of mortality ...

I'll hold back on the conclusions until we have had a chance to think about it a bit more, but the thing which is unexpected ( to my mind ) is the amount of twist that you could get just from the way you hold and push the plane...

The infill is much more rigid but difficult to measure because we can't easily clamp it in the setup... Also with an infill the way the plane is gripped is quite different which probably leads to less twisting..

Originally Posted by Michael G

If I had an FEA package handy I could easily show the distortion due to heating

I suspect it would be possible to model the whole twisting versus different shapes and materials.. Just need a faster computer...

Regards
Ray

I will need some more planes and a way to constrain the coffin body infill that Ray made for a definitive study but as it stands this will have to be it.
I have also included the experimental error in these charts. When I do this again with more planes I will repeat these planes measured here but also measure twist in both directions as well as bow along the axis of the sole.

Did someone say FEA?

Originally Posted by Brobdingnagian

Did someone say FEA?

Is that analysis using your real displacement data?

The rendering is from an FEA package, I don't have all the data to model the absolute displacements (yet).

How accurate is the model though - have you included the bosses and the frog?
From the uniform colour gradient across the sole I suspect not and with the lack of slot I don't know whether you will be able to match up to your experimental results very closely.

Michael
(Apologies - I used to do this for a living, so can get a bit nit picky)

The FEA model was more detailed but the software crashed so I switch back to linux to save my sanity, the one I posted was from a earlier simulation and it is a simple U cross section no frog or cross bracing.

Seriously..... you guys need to get out more

If any one wanted to do the modeling of the plane bodies I would be happy to do the FEA, or even technical drawings would help to make the modeling quicker. That is if people are interested in going that deep into analysis of plane dynamics, although we maybe getting a little off topic.

I can do a technical sketch or two of a Stanley No. 4 (no modelling software to speak of at the moment) if that is of use.

I personally don't see this as OT. The original post was on the distortion of a plane body, illustrated through the scraping process. Interferometry has shown that it does indeed do this; an accurate FEA model can then be used to expand on this, showing both mechanical and perhaps thermally generated distortions. This is the sort of thing that I have done at a couple of my former employers to aid in understanding why an item has failed the way it did. Seeing the results obtained so far I would expect (as the original title suggests) that some useful information could be developed so that people don't go overboard in fettling plane bodies (especially less solid versions) and recognise that even if they have a plane body flat to microns, the way they use it could negate any advantage that brings them.
(It could also mean that knowing this information, someone could design a plane more tolerant of user generated distortions)

Michael
(Josh, PM me with an email address if you are interested in those sketches)