How Flat is Flat Enough

[Andrew Kertesz]

I'm hoping to pull on everybody's wisdom here. I am in the process of flattening the backs of some vintage Stanley 750's. My question is; What is flat enough? After 220 grit sand paper they have an almost mirror like finish, I was planning on going to 400 and stopping. When I lay the ruler from my Starrett 6" square they are not flat and there is light showing under the ruler. Is this something I need to be concerned about or not? The base I am using is a piece of certified marble for machinist work I acquired. Any thoughts or guidance?

[Jake Helmboldt]

Andrew, I'm far from an authority, but a super keen edge is the result of two very flat, highly polished edges meeting. So you need to go higher than 400. I go to 1200 and often an emery polishing paper or a 8000 grit water stone on the backs and the bevel. You only need to do the back once, but it is worth the extra time and effort to get the backs flat and polished (IMHO).

[Mel Fulks]

Flat is good. But I think a lot of the emphasis on flatness comes from the modern mfgs. Seen a lot of old good chisels that are not perfectly flat. When I flatten one it is usually a narrow one that actually rocks a bit . And most of the "machining"is done on the side of the grinding wheel. Do some work with it ,and see if the chisel is working. I have no doubt that for sale somewhere there is something to check if the soup has enough salt, but tasting it works too.

[Terry Beadle]

The flat side of a chisel must be flat for two main purposes. First flat enough to index correctly to the working surface. That is in the case of a tenon shoulder less than 2 thou. Anything more than that, the separation with the mortise surfaces draw the eye. Second, it must be flat enough to meet what Jake Helmboldt stated for the purpose of creating a keen edge. That usually means substantially less than 2 thou.

I flatten my chisel backs going from left over sanding belts of 80 ~ 100 grit for really roughing in and through water stones from 220 through 10,000 grit. As previously said, you only have to do that once and there after you just use the 6000 through 10,000 grits. Some folks even go to above 10,000 grit to get a really keen edge but for wood working in general, 4000 and above will do the job nicely.

I highly recommend David Charlesworth's two videos on chisel sharpening and use. Just really excellent IMO.

It needs to be flat for the first 3/4 inch or so. Some older chisels were set up with a tiny concavity above the 3/4ish mark. I prefer to get it where no light what so ever shows between a straight edge and the back. No light usually means you are in the scoring range of where you are less than a thou. Over time the refresh on the polishing only stones will continue to make the back flat as possible in the context of a small shop and not a scientific research center.

A flat, really sharp chisel is a wonderful tool and makes joints and other adjustments a pleasure. Worth the effort so GO FOR IT ! Hoot!

Enjoy the process.

[Mel Fulks]

Right now I'm watchin ONE video on replacing a heart valve....my chisels are already sharp enuf to do one......BUT SERIOUSLY, folks.....A good research project would be for someone to do a paper on the Wintertur Dominy Shop tools as to how flat they are and what they used to grind and hone them.The WITH HAMMER IN HAND book is good but I don't remember it covering this subject .

[Frank Drew]

I agree with the idea that sharp is two polished planes meeting at an edge, but, apart from that, it's much easier to chop straight down if the back of the chisel is straight and flat. I've had more than one chisel with slightly bowed backs.

[James Taglienti]

I'd say most folks need a few dead flat chisels for general cabinetry work, paring things flush, riding chamfers and shoulders, dovetailing, etc. it is easy to become focused on and swept away by the very intimate, precise functions of woodworking, (look at how many people turn their noses up at registered firmer chisels, and then go buy bevel edge chisels that have huge flats on the sides anyway) but it doent mean every tool that I have MUST be able to pare perfectly, take sub thousandth shavings, etc. It is easy to go overboard on tool maintenance / selection.

[Adam Cherubini]

Flatness is a myth. Tools we think are flat are not. Methods we employ to create flatness do not produce flatness. Last, operations we think require tools that are flat, do not require that. This is perhaps the biggest mistake of our generation. Rubbing tools on abrasive placed on a flat surface doesn't produce a flat tool. Placing a flat backed chisel against a piece of irregular wood does not make the wood flat. Nor does malleting a chisel into wood create a flat surface against the flat of the chisel. We've been wrong about just about everything to do with this subject.

Polish your chisels and forget flatness. Maintain the edge and the angle and forget all else.

[Gary Herrmann]

Very well said, Adam.

[Don Jarvie]

Flat means that the scratch marks are even across the 1 inch on the back. Use the lower grits to flatten the back then you just polish the scratches. If the back doesn't get flat it won't give you a sharp edge. With practice you will be able to see if the backs are truly flat.

[James Taglienti]

Flatness is a myth. Tools we think are flat are not. Methods we employ to create flatness do not produce flatness. Last, operations we think require tools that are flat, do not require that. This is perhaps the biggest mistake of our generation. Rubbing tools on abrasive placed on a flat surface doesn't produce a flat tool. Placing a flat backed chisel against a piece of irregular wood does not make the wood flat. Nor does malleting a chisel into wood create a flat surface against the flat of the chisel. We've been wrong about just about everything to do with this subject.
Polish your chisels and forget flatness. Maintain the edge and the angle and forget all else.
I can feel a substantial difference between the chisels in my shop that i have "reasonably flattened" and those that i have simply polished as is. The flat ones have less tendency to dive when paring, are much easier to hone and maintain, and are more keen.


The ones i have left alone are more difficult to sharpen and less predictable. Many times extreme precision doesnt matter to me. Sometimes it does. Perhaps this is a matter of experience or skill, but if all I have to do is spend 5 minutes on a diamond hone to make my chisels more pleasurable and versatile, consider me deluded.

[Tri Hoang]

I like a flat faced chisel because it is more predictable to me than a random back bevel. Other than the slightly longer initial prep time, I don't see anything wrong with it. Yes, you can get a sharp edge by just polishing (power/manual stropping) both sides.

[Jack Curtis]

I can feel a substantial difference between the chisels in my shop that i have "reasonably flattened" and those that i have simply polished as is. The flat ones have less tendency to dive when paring, are much easier to hone and maintain, and are more keen.


The ones i have left alone are more difficult to sharpen and less predictable. Many times extreme precision doesnt matter to me. Sometimes it does. Perhaps this is a matter of experience or skill, but if all I have to do is spend 5 minutes on a diamond hone to make my chisels more pleasurable and versatile, consider me deluded.

Same here, James. It's too disconcerting to use a chisel that doesn't behave predictably, makes me worry about a slip that cuts skin, to say nothing of wood miscuts.

[Chris Vandiver]

Flatness is a myth. Tools we think are flat are not. Methods we employ to create flatness do not produce flatness. Last, operations we think require tools that are flat, do not require that. This is perhaps the biggest mistake of our generation. Rubbing tools on abrasive placed on a flat surface doesn't produce a flat tool. Placing a flat backed chisel against a piece of irregular wood does not make the wood flat. Nor does malleting a chisel into wood create a flat surface against the flat of the chisel. We've been wrong about just about everything to do with this subject.

Polish your chisels and forget flatness. Maintain the edge and the angle and forget all else.

Wow, that's quite a statement. I'll have to give that a try. On second thought, nah....:)

[Mel Fulks]

What about the original advice request? He did not say he had experienced any problem using the chisel.He did not say he noticed any pitting or problem. He has worked on it and sees no distortion except by looking for light while checking with a machinist quality rule.With the flattening system he is using its not going to be perfect. At some point he may spend some money on a diamond hone for some future problem. We have no reason to think there is anything to fix,we just like flat chisels....Let's get a birdhouse, coffee table, gun rack, bombe chest on chest, or something ....FINISHED!

[Jacob Reverb]

Flatness is a myth. Tools we think are flat are not. Methods we employ to create flatness do not produce flatness. Last, operations we think require tools that are flat, do not require that. This is perhaps the biggest mistake of our generation. Rubbing tools on abrasive placed on a flat surface doesn't produce a flat tool. Placing a flat backed chisel against a piece of irregular wood does not make the wood flat. Nor does malleting a chisel into wood create a flat surface against the flat of the chisel. We've been wrong about just about everything to do with this subject.

That's nothing...go over to the General Woodworking forum and you'll see plenty of folks getting worked up into a lather over a couple of tenths (ten-thousandths, that is) of runout ... or (better yet) "non-parallelism" http://d26ya5yqg8yyvs.cloudfront.net/lol8.gif between their saw blades and their miter slots.

If you try to tell them that the blade (or for that matter, the table casting) will move far more than that if they do so much as breathe on it, they'll act as if you were a convicted child molester...


I have no doubt that for sale somewhere there is something to check if the soup has enough salt, but tasting it works too.

That's a great line!

[James Taglienti]

Andy,
Is the sandpaper worn? Try taking the chisel to The finer grits of whatever system youre using. If the coarse scratches clean up consistently, youre likely good to go. Trying the chisel is always a good test of whether or not it is ready to use.
As you can see, everyone has an opinion on the importance of flatness. Weve got the whole spectrum. We also love to debate. There are a few loaded questions that will trigger an avalanche of posts, each one less and less relative to the original. If it goes unchecked in a couple of days this thread will become an argument about different kinds of tool steel. Luckily by then somebody will have posted a brand new thread about flattening and we can all weigh in again.

[Jim Matthews]

I, for one, would take the advice from Mr. Cherubini seriously.

I think the important aspect about your chisel prep is repeatability.
If you can't do it efficiently, find a simpler method.

I'm rarely cutting anything with my chisels deeper than 3/4 inch. Why do more?
Sandpaper on a flat substrate - spritz with water or oil - polish enough to raise a burr - strop and test on Northern Pine endgrain.

If it cuts that smoothly - I'm good to go.
If not - strop a little more.

[Adam Cruea]

Flatness is a myth. Tools we think are flat are not. Methods we employ to create flatness do not produce flatness. Last, operations we think require tools that are flat, do not require that. This is perhaps the biggest mistake of our generation. Rubbing tools on abrasive placed on a flat surface doesn't produce a flat tool. Placing a flat backed chisel against a piece of irregular wood does not make the wood flat. Nor does malleting a chisel into wood create a flat surface against the flat of the chisel. We've been wrong about just about everything to do with this subject.

Polish your chisels and forget flatness. Maintain the edge and the angle and forget all else.

I tend to agree. My father would probably kill me (he's a die maker by trade), but back 200 years ago when things were made with *quality*, do you really think people worried if a chisel was super-duper flat?

Not to seem like a goober, but "flat" is really, for the most part, unattainable. You're working with an organic medium. . .it's going to move. You could create a "flat" dovetail or mortise or tenon. Check it in a year, and it's not going to be flat anymore, no matter how much you cuss, kick, yell, scream, and flatten it down to "tolerance".

It boggled me for the longest time about how flat you needed to get a chisel, and I did some reading that really made me think. Back in the day, people didn't have "flat" references. People didn't true their stones up to a "flat" surface because, well. . .you couldn't measure down that far. Yet look at some of the work that was done back then by people and is still being used today.

I fall back on this, which makes complete and utter sense to me, unlike the "flat flat flat flat dead nuts flat" a lot of people believe.

http://toolsforworkingwood.com/store/blog/243

[jamie shard]

I don't obsess about flat, but if anything I sharpen to make chisels fractionally concave.

This is the page that got me started down the road:



240494



http://www.hand-cut-dovetails.com/tools/9bottomsurface/btm.html

(Does David Charlesworth use the same technique?)

[Mel Fulks]

Jacob,thanks for the compliment.There is something about being around modern stuff that brings out OCD.People see every variation as imperfection.A friend commented that if you give someone a drink in a paper cup they are fine with it ,give them a drink in a fine crystal glass and they look for chips.

[Jim Matthews]

Yep.

I saw Tom Lie-Nielsen demonstrate the same technique which he credited to the inestimable Mr. Charlesworth.
I suspect DC gets better results than most due to a slurry of cigarette ash and ancient limestone falling from the ceiling.

The point is to follow repeatable steps, and get to putting steel to wood quickly.

[David Weaver]

I fall back on this, which makes complete and utter sense to me, unlike the "flat flat flat flat dead nuts flat" a lot of people believe.

http://toolsforworkingwood.com/store/blog/243

Wow.. that's quite a nutty comment list on there. (I hope I didn't make one of the comments).

I will usually polish the first inch or two of a chisel if it's close to flat. If it's not, I'll do as little as I can to make it work for what I want. If there aren't pits in a back that are threatening the edge, and it's not a paring chisel, I often won't do too much to it, especially not beyond the part of it that determines whether or not the chisel is going to go the way you expect it to halfway into a cut.

But on a parer, I always have the back polished well right to the edge and uniformly across. There is little more in woodworking that is more satisfying than cutting a chamfer with a paring chisel, exactly to a mark with nothing but a waxy surface left behind. If you sand things, you may not experience something like that or have no regard for it, but as my guitar teacher used to say "it's just the___" fill in the blank with whatever word you use, you'll probably not guess his.

Something that's going to get hit hard, if the back is close, no, it'll eventually get polished toward the edge through use.

What charlesworth does is different, though. As far as I can tell, David does like to use things like thicknessed guide blocks on cuts like HBDTs, and you have to have some level of straightness if you're going to do that and rely on guide blocks an inch or two away from where the cut is happening.

I tried to find something that holtzapffel said about chisels, but my volume doesn't have it, at least I couldn't tell if it does. I wouldn't be surprised if some of the finer woodworkers of the late 18th or early 19th century were pretty intolerant of out of flat backs, at least for the first couple of inches, because they would use the same backs then to condition the surface of their stones. I think there's little chance we can look at tools they may have used (and that have been totally bastardized since, by carpenters, etc) and get an idea of what they did.

[Pat Barry]

I am having difficulty understanding the comments made about the back of the chisel not needing to be flat. Here is why. It is my understanding that the back of a chisel is the reference surface and is flat to provide a contact surface to guide the tool along a straight pathway thru the wood. Now I understand that certain chisels may have concave surface on the back, but that portion of the tool is incidental in that design. The bearing / reference surfaces are the non-concave portions, particularly the edges, and those are ground and polished as flat as possible. Pits and so on have no bearing (sic) on the performance of the flat back. On the other hand, the back of the chisel can't be convex or raised with respect to the bearing surface. If it is bumped up then it better be flattened or the tool will not function as intended. Being concerned with flatness is therefore important for traditional, er European, chisels.

[Adam Cruea]

I am having difficulty understanding the comments made about the back of the chisel not needing to be flat. Here is why. It is my understanding that the back of a chisel is the reference surface and is flat to provide a contact surface to guide the tool along a straight pathway thru the wood. Now I understand that certain chisels may have concave surface on the back, but that portion of the tool is incidental in that design. The bearing / reference surfaces are the non-concave portions, particularly the edges, and those are ground and polished as flat as possible. Pits and so on have no bearing (sic) on the performance of the flat back. On the other hand, the back of the chisel can't be convex or raised with respect to the bearing surface. If it is bumped up then it better be flattened or the tool will not function as intended. Being concerned with flatness is therefore important for traditional, er European, chisels.

To my knowledge, tools won't go straight through wood. Since you have a wedge, the bevel part of the wedge pushes the tip of the wedge back. I actually have rarely ever gotten anything wedge-shaped to go straight through anything.

With that being said, I don't use the backs of my chisels for reference. If there's one thing I've learned quickly, it was that wood is not the same density and it's way too easy to have a corner of a chisel dive into a softer portion of the wood and screw up what you're doing. If I want to check flat, I find a straightedge.

And to be completely honest, I want to work wood, not metal. Wasting time flattening a tool that's getting ready to cut into an organic medium that will change, to me, is just completely absurd. You'll never get wood flat and keep it there. . .why waste time getting your tool flat so you can get frustrated that your wood isn't flat?

Now, that being said, I don't use tools that are obviously out-of-flat by a mile, but if I see a little light on a chisel, I'm not going to let the OCD go nut-nut and lap it flat. Same with the wood that I work. Do I look for dead-perfect flat? No, I don't have that much time in my life. But if it's an obvious 45* slant, I might take it down to 15*. ;)

[Sean Hughto]

Ha! Love it. My sentiments exactly. These kind of fine points of tool fettling threads amuse me. Build something! you know - you can, even without the most perfected tool imaginable.

[jamie shard]

Wedging is a matter of degree. If you have 4 inches below the flat back of your chisel and 1/64" of an inch on top of the bevel, the shaving is going to peel away and the chisel will essentially go straight.

[Adam Cherubini]

Wedging is a matter of degree. If you have 4 inches below the flat back of your chisel and 1/64" of an inch on top of the bevel, the shaving is going to peel away and the chisel will essentially go straight.

That depends on the grain right? What is more likely to happen is the shaving propagates a crack ahead of the edge and very likely deeper than the intended cut. That's why planes have soles.

Remember too that if you have a flat backed chisel, flat on the wood, you have zero control on depth of cut. You can't rock the chisel to take a lighter cut. The depth is determined by the amount of wood above the reference surface. Doesn't take much for that cut to be too much. 1/64 is very likely too much. My guess is I pare a few thou at a time (like .005).

This is the same issue as the recent discussion of chip breakers. This is something I want to talk about at WiA. With hand tools, the waste can really effect the tool. Most power tools aren't botherd with waste or obliterate it (like a router or planer). Very different situation. The other Adam is right- The wood doesn't care about your reference face- it just sees a wedge and responds accordingly. This is especially true of chopping.

[jamie shard]

(I agree that 1/64 is too thick, but I was trying to come up with an example that wasn't way off at one extreme.)

I think we need an actual scenario to make this more real.

I was think along the lines of paring the shoulders of a tenon or the flat ends of the "tail" side of a dovetail joint . I agree that about .005 will be trimmed off at a time, and usually there will be a few shavings taken to sneak up on the gauge line. It seems like the wood does care about the reference face, but in both of these examples, the grain tends to be oriented perpendicular to the cut. Might not be a good example. What scenario are you thinking of?

[Prashun Patel]

Personally, I think these things are best learned through experimentation. If you have been able to get an even sheen when lapping the back, then I'd go to sharpening/honing the bevel, then make some cuts. If you get results yr happy with, with the effort yr happy with, then you've found yr sweet spot.

[Frank Drew]

I've never used magnifying glasses or microscopes or micrometer calipers to prepare my edge tools, but I like "sharp" tools and prefer a "flat" back to a chisel for the work I would mostly use a chisel for -- chopping mortises or dovetails, etc.; for those operations a convex back would be a hindrance, IMO.

I'm not going to engage in a discussion on whether or not flat or sharp are real or mythical, though, beyond saying that of course we can't achieve perfect levels of either condition.

[Adam Cherubini]

(I agree that 1/64 is too thick, but I was trying to come up with an example that wasn't way off at one extreme.)

I think we need an actual scenario to make this more real.

I was think along the lines of paring the shoulders of a tenon or the flat ends of the "tail" side of a dovetail joint . I agree that about .005 will be trimmed off at a time, and usually there will be a few shavings taken to sneak up on the gauge line. It seems like the wood does care about the reference face, but in both of these examples, the grain tends to be oriented perpendicular to the cut. Might not be a good example. What scenario are you thinking of?

With you 100% (which is why I wrote "depends..."). Have to look at individual scenarios. I use chisels for tons of things. Probably more than any normal woodworker. I love chisels. Removing a tiny step of end grain is different from paring long grain. End grain can create enough force to cause an undercut. I know because I've done it. But I think this is the basic operation that many guys think of when this subject comes up. And I agree a flat chisel seems like a requirement for this to work.

I'm going to do some more work on this subject. I want to know how far out of flat my chisels are in inches. I suspect that the chisel that doesn't develop even scratch patterns or the OPs that you can see light when checked against a straight edge COULD be out less than .001". My "out of flat" chisels, won't develop even scratches on a Shapton, but I sincerely doubt many guys would have any trouble using them or even know they weren't flat.

So i agree Jamie, we need specific scenarios and real measurements of flatness. I can borrow some inspection tools from the machine shop and see what I can come up with.

[John Coloccia]

I suspect that the chisel that doesn't develop even scratch patterns or the OPs that you can see light when checked against a straight edge COULD be out less than .001".

Without writing a treatise on this, I will say for a fact that I can easily see fractions of a thousandth under a straight edge. Now, I have some machining experience, and you start to develop an eye for this sort of stuff, but I think ANYONE with reasonable eyesight can candle a joint to far less than .001", especially if you're candling against a thin straight edge.

My criteria is that I flatten on a course stone (a diamond stone) until I have a consistent scratch pattern. I polish on a ceramic stone (a Spyderco) until those scratches are gone. I polish further on an ultra-fine Spyderco until the scratches are gone. Then I strop. As I use the edge, I strop frequently....very frequently. I have several strops kicking around on my bench pretty much at all times. That keeps them ultra sharp for a LONG time.

IMHO, the straight edge has no real use for sharpening other than possibly the "ruler trick". IMHO, there is a lot of money to be made with sharpening equipment, and I'll admit I went down the rabbit hole and have spent thousands over the years on sharpening contraptions, but it's just not necessary and it's not that difficult. That said, the only thing that gets you there is practice, and sharpening contraptions help you get there in the beginning. It's helpful to hook up with someone that can show you the light.

But toss out that straight edge. For sharpening, you should be monitoring the scratch pattern. If you're not rocking and creating a convex surface, the scratch pattern will tell you everything you need to know about the quality of the edge.

Just my opinion.

[Jacob Reverb]

Build something!

HERETIC!!! (http://www.sawmillcreek.org/showthread.php?191218-Saw-Blade-Run-Out)

[Jim Matthews]

If you really want to bend your noodle around the idea, have a browse through
Paul Sellers videos on YouTube.
(http://paulsellers.com/2012/01/sharpening-chisels-forget-weaker-micro-bevels/)
His technique seems to work, and should be readily repeatable without special tools.
Of note, he uses simple Marples bevel edged chisels for nearly everything.

That means he need only flatten the back of 4-6 chisels for the majority of his work.

His speed at putting an edge on his chisels and plane irons depends on developing a convex (!) bevel.
As mentioned by John Coloccia above, judicious stropping keeps things hopping.

It's not the method I use, but it is compelling.
His demo slicing paper has me flummoxed.