Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Super Resolution Thermal Imaging Camera

how flat is flat enough

Posts

how flat is flat enough

#1

how flat is flat enough

RobRing

>I posted this originally in the general message board, but the concensus was that the topic might stir up a bit more traffic (if not agreement) over here.

I have a stanley #7 plane whose sole I need to flatten. I acquired it sight-unseen. I wouldn't have paid for it if I had a chance to check the flatness - buyer beware. But now I have it so use it I must! The length of the plane far exceeds my usual flattening surface, so I thought I'd use a spare cast iron table saw wing I have. Of course, it's not perfectly flat either. The wing is 27" X 10". Across the width from top to bottom it's as flat as it needs to be (within 0.001"). Along the length it ranges from a .002" dish in the center of the wing to a .004" dish near each edge. diagonally it's dished .003" along one diagonal and .004" along the other. I try to avoid being so obsessive about applying machining-type tolerances to my woodworking, but in this case it's technically metal working. Is this flat enough to use as a flattening surface for the plane? I can have the wing precision surface ground at a local shop for $40, but I'm not sure if it's worth the trouble.

Thanks,

Rob

Re: how flat is flat enough

#2

Re: how flat is flat enough

Todd O. Cronkhite Maine Native, Presently Away

>Rob,

What makes you feel that the sole needs to be flattened?

If this thing is twisted to the point that you can actually see the "out of flatness with the naked eye, then I'd say strip it for parts and toss the sole as you'd never get it flat again to the point that you'd want and still have metal left on the sole.

If you can't see with the naked eye then it's flat enough.

Have you used it yet?

Are you new to hand planes? I ask because alot of newbies to them get all obsessive over NASA type tolerances of the soles, and that is absolutely foolish. Sharpen and use it. Use it to do some edge jointing, then make your determination. If you use for surface flattening your more then likely going to follow behind it with a smoothing plane so even if it .001 out it isn't going to matter.

Don't make Todd H. call you a Machine Man. I sure wish that thread could be found as I feel that it is very enlightening and will set you free of obsessive foolishness.

T.O.

Re: how flat is flat enough

#3

Re: how flat is flat enough

Derek Cohen (in Perth, Australia)

>Hi Rob

I'll begin by echoing the question of Todd, "Have you tried the plane and decided that the performance is not up to snuff?"

The point is that looks may be deceiving. Any and all handplanes do not need to be flat over their full length. They only need to be co-planar over the toe-mouth (especially the front but also at the back)-heel. The question is 'how do you determine this if you do not have a long-enough straight edge'?

Sooner or later everything moves, whether it is wood or cast iron. So I keep a couple of long lengths of 3/8" thick glass plate around. These are used for both lapping and maintenance (if needed). One is 1m long for long planes, and the other is 300mm for shorter planes. Shorter is better for stiffness (glass flexes easily), and so the glass is attached to three layers of 3/4" thick MDF. MDF is flat, but it also flexes, so the whole lot is used on a flat surface, such as a bench top, table saw, or the floor (as below - note that this is an old picture. The glass is now attached with 3M sprayed contact glue).



I glue on beltsander belts. These wear better and are easier to attach and remove than paper.

Use a fine grit and run the plane sole over it to identify the wear pattern. You will immediately see whether the important areas are co-planner or not.

If you do need to lap the sole, start with a coarse grit, say 80, then roughly "double jump" ... 120/150 .. 240 ...400 That should be more than enough. The 80-level is the hardest work, but it goes quickly once you are working with flat surfaces since you are smoothing not removing metal.

Regards from Perth

Derek

Re: how flat is flat enough

#4

Re: how flat is flat enough

Warren in Lancaster, PA

>Rob, you don't say how bad the plane is, or whether it is concave or convex or what. My guess is that it could be improved to a usable state by lapping on your cast iron, maybe even on your 9 x 13 granite. I am not saying this would be the optimum condition, but definitely usable. I do think some flatten the sole to finer tolerances than they can actually appreciate in use.

I suggest first seeing how it performs in its present state. Then true it as well as you can, checking progress with a straight edge, and then check performance again. Checking performance before and after will give you a feel for what the truing does for you and some judgment about how carefully it must be done.

Re: how flat is flat enough

#5

Re: how flat is flat enough

Alan DuBoff

>** Caution - entering SARCASM zone **

I would go out and buy the best machinist straight edge, just to make sure your work is at least that flat until it continues to acclimate.:-/

** You are now exiting the SARCASM zone **

Re: how flat is flat enough

#6

Re: flattening

Moses Yoder in White Pigeon, MI

>When you read an article or a book about tuning a plane they all talk about flattening the sole of the plane. The truth is the sole has to be flat. The question is, what does that mean? Articles in magazines are usually limited to a certain size. Books are also; they are all calculated to be a certain size to sell at certain price points to the target market. The author's time is also limited, plus he/she needs to keep the interest of the reader. So I think in a lot of cases "flattening the sole" could be explained better. There are a number of steps involved in tuning a plane. Several steps are more important than flattening the sole. 1) Making sure the frog is bedding correctly. 2) Making sure the iron is bedding correctly on the frog, at least at the front edge of the iron. If the frog has a concave face it can cause the iron to flex downward under pressure, preventing a fine cut. You will adjust the iron till it just bites into the wood only to find it digging deeper into the wood and causing chatter. The same thing will happen if the frog is not bedding correctly at the mouth of the plane. So those two items need to be addressed before you flatten the sole.

I think this may have been done already, but I plan on doing it anyway. I'll set up a dial indicator and see how much flex the sole of a #7 stanley plane has. If supported ont he two ends and pushed downward, how much will it bend? This of course would vary between planes, and would also vary considerably between the old stanley planes and the new LN or LV planes. But I'm pretty sure the sole will flex on such a long plane. A few areas are critical. When I talk about flattening a plane I refer to "high" as those areas that hit the board first. A "low" spot on the plane would not make contact on the board because of the high spots. 1) The area right behind the mouth cannot be protrude out past the area in front of the mouth. It actually works fine if the area behind the mouth is a low spot, but preferably the area right infornt of and right behind the mouth are on the same plane. 2) The toe and heel of the plane should be on the same plane as the mouth. Here the question of "flat" arises. How flat is flat enough? I think the best method to determine this measurement has already been discussed in previous posts; actual use of the plane. I would not waste my time checking a #8 stanley bench plane with a machinist straight edge. I think it would be an excersize in frustration. If the iron is bedding correctly and the area around the mouth is flat then simply using the plane will determine if further flattening is necessary. A good method to determine how flat the sole is has been shown with the plate glass and sandpaper; a few strokes across the sandpaper will reveal the high spots. I think your saw table is plenty flat enough to accomplish the sole flattening, but I like having a 1/2" plate glass set up for a number of things such as flattening my water stones. I don't care much for the idea of getting spray glue and metal filings and stone dust around my saw. The 1/2" plate glass can normally be purchased pretty inexpensively and is a handy thing to have around the shop.

Re: how flat is flat enough

#7

Chris Scholz

Re: how flat is flat enough

Chris Scholz

>Not as flat as you might think, Rob. Here's why:

If I am not comletely off than all you do with a plane is to convolve two surfaces with each other.

For argument's sake let's say the profile of the surfaces of your board is f(x) in x and g(x) is the profile of the surface of your plane. For simplicity let's say both surfaces perfectly flat in the y direction and the z direction is in the dirction of the surface normal.

The resulting surface h(x) is simply


h(x) = ∫0L f(ζ) g(x-ζ) dζ,

where L is the length of the board.

Just decide how flat your final board has to be, how many passes you want to take, measure how flat to board is to start out with and solve for g(x). That you should get pretty close to an answer.

Chris

---

Chris Scholz

Atlanta, GA

Galoot-Tools

Re: how flat is flat enough

#8

I am in trouble, not good with math

Stephen Shepherd in the State of Deseret

>

Re: how flat is flat enough

#9

Been awhile ;-)

Dan Donaldson

>Gee, it has been a few (well, maybe a lot ;-)) of years since I did any integration. Let's see, does this mean that everyone will need to take calculus if they want to do accurate woodworking? ;-) Maybe that is why I am in trouble ;-)

Re: how flat is flat enough

#10

Re: how flat is flat enough

RobRing

>There's my problem right there! I was using "W" (width) as the exponent, not "L." I always make that mistake.

Re: how flat is flat enough

#11

Re: how flat is flat enough

RobRing

>Wow, you hand tool guys take the fun out of everything! Next thing you know you'll tell me that wood moves so I should stop measuring my tenons with a micrometer ;).

Just kidding, you're actually confirming my inclination on the flatness question. The plane does not appear to have any twist. It's concave, as in the mouth sits higher than the toe and heel - if it was convex I wouldn't be as concerned. I've flattened a few other (shorter) planes, and my "flat enough" guideline is once the scratch pattern looks uniform. I'm sure it's not to machine tolerances, but I can notice the difference in planing performance. Too be fair I haven't tried using the 7 yet - I will for kicks and than try again to see the difference after I've satisfied my flattening OCD with my flat enough table saw wing (it's a spare, so I'm not as concerned about getting adhesive / metal filings / etc on it).

By the way that wasn't my real question, it was just a teaser. I really wanted to know at what bevel angle I should sharpen the iron. I usually set it at 31.5 degrees, with a 0.75 degree microbevel. I've been toying around with 31.75 and even 32.125 for really knarly grain. But I'll probably skew this plan a little less when I use it, so I could probably get away with something in the 30.8 to 31.2 degree range, do you think? Of course I could just change the microbevel by an eigth-degree or two instead. And I suspect the answer is different if I use a bezel instead of a bevel, right?

Re: how flat is flat enough

#12

Re: how flat is flat enough

Wiley Horne -- So. Calif.

>Rob,

Concave is by far the simplest case to deal with, because it means that you can simply lap the sole rather than using more complex methods.

The plane will not work as well with the mouth in the air, as it will if the front of the mouth is sitting down flush on the work. What you want to do is follow Moses Yoder's method, with the blade in the plane but retracted. Lap the sole until you have the front of the mouth, the heel, and the toe showing the same scratches--at that point they are coplanar. Use the entire surface of the abrasive, and reorient the plane occasionally front for back, to randomize the scratching.

When the front of mouth, the heel, the toe, and the periphery are coplanar, the plane is self-jigging for flat. Never mind any hollow places in the interior--such hollow places are OK, and they're your guarantee that you haven't convexed the sole yet.

Wiley

Re: how flat is flat enough

#13

is that really how you think?

Tom Sontag - St. Louis

>Your method may be dead on right as far as I know, but except for a fully employed engineer using college level math daily, it is of little use to a woodworker. Just because mathematics is capable of expressing a problem does not mean it is the best or fastest way to solve it. I am not even sure what the number you solve for gives us toward a final solution to flattening a sole.

How about this method? Apply tool at hand (imperfect cast iron surface) to problem at hand (unflat sole) with ~80 grit abrasive in between after tuning other already-mentioned more important plane tuning aspects. When significant results are noted, attempt to use the plane on wood. Note results and adapt further efforts accordingly.

Re: how flat is flat enough

#15

Re: how flat is flat enough

Dan Donaldson

>You said:



By the way that wasn't my real question, it was just a teaser. I really wanted to know at what bevel angle I should sharpen the iron. I usually set it at 31.5 degrees, with a 0.75 degree microbevel. I've been toying around with 31.75 and even 32.125 for really knarly grain. But I'll probably skew this plan a little less when I use it, so I could probably get away with something in the 30.8 to 31.2 degree range, do you think? Of course I could just change the microbevel by an eigth-degree or two instead. And I suspect the answer is different if I use a bezel instead of a bevel, right?


Those numbers are probably all right, but if you really want to get accurate, you need to take into account the phase of the moon when the iron was made, as well as the orientation in the Earth's magnetic field. The first initial of the person running the furnace that smelted the iron has some effect also, but this can normally be ignored as long as you only need shop accuracy.

Re: how flat is flat enough

#16

Chris Scholz

I don't disagree...

Chris Scholz

>...but it appears this question (how flat is flat enough?) has been discussed a gazillion times, maybe it is time to try a new approach?

Re: how flat is flat enough

#17

Re: how flat is flat enough

Marv....... Northern California Lake Co.

>Rob,

When lapping for the purpose of flattening a surface, if it is possible, a figure 8 motion is the best and easiest way to do it.

Marv

Re: how flat is flat enough

#18

a question for successful hand lappers

Bill Tindall, E.TN

>Apparently some of you have accomplished something I have been unable to do successfully. When you push metal across an abrasive sheet, this plane for example, how do you keep the leading edge from cutting more than the trailing edge?

This tendency, in my hands, for the leading edge to cut more aggressively, will upon turning the plane end for end, soon produce a concave sole. Others I know reproduced this result, suggesting that it is easy to do, if not normal. So how do you not do it?

Re: how flat is flat enough

#19

Re: a question for successful hand lappers

Denis Ch�nard, Orl�ans, Ont.

>Did you mean convex sole? Or did I miss the boat? :-)

One piece of advice I read from Mike Dunbar is that when flattening a sole, one should not use the "standard planing stroke", i.e. press down on the toe at the start of the stroke, shifting the weight to the heel as the plane passes over the abrasive. That is a good way to get a convex sole.

Instead, one should just "slide" the plane back and forth over the abrasive without putting extra pressure on any point of the sole. So far it has worked for me up to jack size. I tried doing a #8 but it didn't work because my reference surface wasn't long enough. I do have one now, but no planes soles to flatten...

I didn't read the whole thread so I might be missing something...

DC

Re: how flat is flat enough

#20

Bill, did you mean to say 'concave sole'

Wiley Horne -- So. Calif.

>or 'convex sole'? To me, concave means hollow, while convex means bellied outward.

Your analysis seems to imply that a convex sole will be produced, but you said 'concave'. I may be dense here--wouldn't be the first time.

I have a couple of remarks, once we get this sorted.

Wiley

Re: how flat is flat enough

#21

Re: a question for successful hand lappers

Joel

>I discovered the hard way that if you want to save odles of time - don't try lapping the entire sole at one. Figure out the high points and with a hardwood sanding block and wet or dry paper locally sand the high spots. be carefull around the mouth.

Re: how flat is flat enough

#22

Re: how flat is flat enough

paul womack

>I won't post my link again about HOW to flatten a plane sole.

But I will link to a highly useful pice of maths/mechanics from the OLDTOOLS list, way back when.

It tells you the flexibility of a plane.

It is obvious (I hope) that precision beyond the rigidity of an object is pointless.

And this page concludes that a #8 under 10lbs of downforce may well deflect by .020" (circa 1/64").

ACTUAL measured deflections (later in the thread) are around 0.002 - 0.007.

maths!

BugBear

Re: how flat is flat enough

#23

Re: a question for successful hand lappers

paul womack

>I discovered the hard way that if you want to save odles of time - don't try lapping the entire sole at one. Figure out the high points and with a hardwood sanding block and wet or dry paper locally sand the high spots. be carefull around the mouth.

Yeah - one of the reasons "normal" lapping gets slow is that as the flatness increases, the area in contact with the abrasive goes up, hence the pressure (and cutting action) slows down.

BugBear

Re: how flat is flat enough

#24

error!

Bill Tindall, E.TN

>change my concave to convex. Always had trouble with those words.

Re: how flat is flat enough

#25

Re: error!

Marv....... Northern California Lake Co.

>Bill,

Me too...."innie" and "outie" works for me.

Marv