Need advice on planing technique.

[Sam Shankar]

Hi folks. Plane newbie here. I've been trying to flatten my workbench top. No matter how many diagonal passes I make with my jointer plane, I always end up with a bit of a 'hump' in the middle. The good news is that it is reliable and reproducible, so I am guessing it's something wrong in my technique. In the picture below, you can see where I've scribbled pencil on the low spots of the workbench top. Those low spots are about .006" below the middle of the bench, which is otherwise flat. I know it's not easy to see but I didn't want to use a sharpie(!)

The top is 72" long x 28" wide. I am taking overlapping diagonal passes with a jointer plane, starting first on one side, and then switching to the other. I am struggling to maintain consistent fore-and-aft pressure on the plane over the length of those passes, and I suspect that is the issue. Any hints? Or should I just give up on doing big pretty passes and just shave away the middle with my #4?

430468

[justin sherriff]

If there is a hump in the middle just plane the middle. Go with the grain for a bit. Here is a nice video that could help you out may be. https://www.youtube.com/watch?v=2_96gNMMc_g

[Derek Cohen]

Sam, determine where the hump is, mark the area, then plane this first. Ensure all high spots are removed before using winding sticks to square and level the bench top as a whole.

Regards from Perth

Derek

[Robert Hazelwood]

I would dispense with the diagonal strokes and plane along the length. Take full-length passes along the length in the middle where you've indicated the top is high. Avoid doing passes towards the edge where you've marked with pencil.

You should stop, resharpen, and reduce the depth of cut to 1 or 2 thousandths for planing along the grain with a wide shaving. You want something you can control, and you are reasonably close to flat so no need to fight a thick shaving. If the middle is 6 thousands high then 3-6 series of overlapping passes should take care of it. Then do a few series of overlapping passes across the entire surface. The goal is to get full width shaving for the full length of each pass. You should be where you want to be at that point.

Don't forget to check for twist.

[Jim Matthews]

If you put a pencil in the middle, aligned with the long axis- does it roll?

[Rob Luter]

Sam, determine where the hump is, mark the area, then plane this first. Ensure all high spots are removed before using winding sticks to square and level the bench top as a whole.

Regards from Perth

Derek

^^^ This ^^^

[William Fretwell]

It looks like a new workbench. The wood will move over the next few seasons, the top may twist etc so don’t get too obsessed with perfection. Use it for a year then flatten it again, then you will have a stable flat top for a while.
The top is much larger than your long plane so measuring accurately with a straight edge and using a #4 on the high spots makes the most sense.
Whatever finish you apply after flattening, do the same to the underside.

[James Pallas]

It always helps me to have a little reminder. I was taught “ it’s just like raking the yard, first you knock down all of the mole hills before you try to rake the whole yard”. That has stuck with me for many years. I think about it almost every time I work to flatten something, Wood, yard, garden, concrete, drywall, doesn’t matter.

[Tom M King]

It's sort of like sculpture, only flat. Only take off the part that's not flat. Passes all the way across won't ever get you there.

[Tom Trees]

I suggest you make a pair of straight edges at least the length of your bench if not a bit longer.
Make them parallel so you can flip one over to check for any mating bow in the timbers.


I am making do with a composite material like derlin lab countertop resting on a roubo ish bench base, and learned some lessons about flattening tops often,
or in my case shimming it often.

I found that using shorter lengths for straight edges to refer to, and shim accordingly, resulted in a bowed profile in longer stock...like snipe on the ends.
Not nice! quite noticeable and all the rest when your laminating long stock.and the error doubles up when pairing timbers.

Good luck
Tom

[Jim Koepke]

Some good advice above.

If you need to make winding sticks or straight edges this is a good place to start > https://www.scribd.com/document/50097193/Making-Accurate-Straight-Edges-from-Scratch

The notes on the averaging process, not visible unless you sign up for a free trial, are pretty much common sense clarifications.

This method was used for making my winding sticks:

430484

My plan is to make a shorter pair in the future.

jtk

[Bob Jones 5443]

Jim, what wood species are these? Do you see seasonal movement?

[James Pallas]

Hi folks. Plane newbie here. I've been trying to flatten my workbench top. Noo matter how many diagonal passes I make with my jointer plane, I always end up with a bit of a 'hump' in the middle. The good news is that it is reliable and reproducible, so I am guessing it's something wrong in my technique. In the picture below, you can see where I've scribbled pencil on the low spots of the workbench top. Those low spots are about .006" below the middle of the bench, which is otherwise flat. I know it's not easy to see but I didn't want to use a sharpie(!)

The top is 72" long x 28" wide. I am taking overlapping diagonal passes with a jointer plane, starting first on one side, and then switching to the other. I am struggling to maintain consistent fore-and-aft pressure on the plane over the length of those passes, and I suspect that is the issue. Any hints? Or should I just give up on doing big pretty passes and just shave away the middle with my #4?

430468

Hi Sam, How is DC. I grew up there and can imagine it being hectic at this time. May I suggest you read my 4 recent posts “String line use”. You may find some use for a string line as opposed to making straight edges. I’m not denying the usefulness of straight edges. In any case work the humps first. Winding sticks are helpful. Stay off of those edges that you have marked. Working large areas you sometimes need to just get up on it kneel down and work the middle. Good luck with your project. Try to stay safe.

[Nicholas Lawrence]

If you are new to planes you may be assuming a jointer plane works like a jointer, only upside down. That will cause you frustration.

With a jointer you move the wood over the bed, and it comes out flat.

If you just move a jointer plane over your stock and keep doing that until you get to flat you will end up knee deep in shavings looking for another piece of lumber.

As everyone else has said in their own way, with a plane you have to find the spots that are “out” and make them better.

Those full length full width .00001 shavings you read about are more a way of verifying that you have it flat. They are not the way you get to flat.

[Prashun Patel]

When planing across the width of a bench, you have to awkwardly reach to the other side; you can't shift your legs and center of gravity properly. I suspect that's why you can't maintain fore and aft pressure evenly. Personally, I wouldn't worry about planing full lengths. I never do that on table tops or my benches. I just work in local areas on the high spots. Far more critical is winding sticks or another flat reference to provide feedback.

[Warren Mickley]

You don't need to make three straightedges to insure precision. Nicholson's method from 1811 uses only two. Peter Nicholson was a cabinetmaker and a mathematician. I will outline it if anyone is interested.

It does not seem as if Sam is having trouble finding a straightedge; he knows where the high spots are, but needed to know how a plane worked.

[steven c newman]

Ever watch Ishitani? He merely uses a STRAIGHT 2 x 4 x 8' along the top. He is checking for how flat the board/top is.....then, he merely runs a power hand planer straight across to lower the high spots....then rechecks, and planes again where needed. THEN follows up with a hand plane going at a slight diagonal....

Reason for the low spots along the edges....all the pressure on the front knob of a plane when you start it.....
Don't use a "Death grip" and press down when starting
Even pressure all the way across....

A long bodied plane will tell you if there is a high spot in the middle, as it will rock a bit going over that spot...where the toe is up in the air.

I usually go at the 45s across to level a board, I go all the way down from one end to the other...then go the other way back up to the start point. second cut is at 90 degrees to the first cut. THEN go with the grain....until a full width shaving appears. Used to call such a plane doing that a "Try Plane" like a 5-1/2 or a #6 size.

Did I read correctly .....OP is worried about...0.006 high in the middle ...on a WORKbench? Just put the dang thing to work for a few seasons, and then recheck....wasting daylight worrying about a paperthin bump......About like wanting to have the sole of a hand planed machined flat.....BEFORE checking the sole.

[Sam Shankar]

Wow, lots of great advice. Here's what I'm hearing, and what I'll do:

1. Now I get it: even a long plane doesn't work magic on convex surfaces. So I'll stop trying to do full diagonal or cross-grain plane strokes to level out high spots.
2. Sharpen the jointer plane again. (Ugh. A2 steel.) That's a technique issue itself but god knows this forum has enough advice on it already.
3. Take thin shavings, lengthwise, focusing on the middle of the workbench until it's reasonably flat, don't try to achieve perfection.
4. Drill some dog holes and make a dining table. Sort of looking forward to that part.

FWIW--
* I am using wooden straightedges made with my (electric) jointer. If that's no good and I need to learn another hand plane technique to produce a REAL straightedge (and get better at sharpening before I do that), I may get back into motorcycle repair!
* If I put a pencil in the middle of the table, it doesn't roll. (Round things don't roll either!)
* I know that this convex surface came from less-than-perfectly-square subassembly edges, not wood movement.
* It's a bevel-up jointer plane.
* My family is healthy, and my wife and I have job security, so I'm lucky that this is my current frustration.

[Jim Koepke]

Jim, what wood species are these? Do you see seasonal movement?

To the best of my knowledge my winding sticks are made of maple with rosewood inserts. It came from a table found on my way to work on a spring clean up day the local waste management company sponsored in conjunction with the fire department.

The wood has been acclimating to my shop for about a decade and seems to be fairly stable.


You don't need to make three straightedges to insure precision. Nicholson's method from 1811 uses only two. Peter Nicholson was a cabinetmaker and a mathematician. I will outline it if anyone is interested.

It does not seem as if Sam is having trouble finding a straightedge; he knows where the high spots are, but needed to know how a plane worked.

Warren, many of us would likely appreciate an outlining of a way to make a pair of straight edges.

Yes, Sam has been able to find the high areas from side to side. As one gets their work flat side to side and end to end one needs the ability to see if there is any twist.

jtk

[Jim Koepke]

I am using wooden straightedges made with my (electric) jointer. If that's no good and I need to learn another hand plane technique to produce a REAL straightedge (and get better at sharpening before I do that)

If the straightedges have no gaps or humps when set face to face and show no gaps, then you should be good to go.

jtk

[Robert Hazelwood]

You have to consider three dimensions- flatness along the length, flatness across the width, and twist. From your picture and descriptions it looks like you have good flatness along the length in the center portion, and across the width you have a central flat portion but falls off towards each edge.

If the fall off on the sides is indeed only 6 thou, and there is no twist, then you are really very close to being done.

You seem to be getting a lot of advice regarding spot planing that is more appropriate for the initial jack planing stages than where you are actually at, IMO. You have a large broad "high spot" in the center of the width, it just needs to be brought down so that it extends all the way to the edges. So just plane along the length staying away from the edges and that hump will be gone- the geometry of the plane will not allow you to create much of a hollow.

IMO it is more efficient and effective to plane along the grain, except perhaps for the earliest stages to remove gross high points. You can control flatness in both directions, as well as twist, without having to change planing directions.

As for my advice on thin shavings, going along the grain with a 2"+ wide shaving you will struggle to push much more than 2 or 3 thou shaving thickness while maintaining good control. If you want to go thicker you can add camber to narrow the shaving, but there's no need for that at the stage you are at. Looks like you are about 5 minutes of planing away from being done if its just the 6 thou on the sides to deal with.

[Tom Trees]

You can make a pair of straight edges finer than what you can see
by rubbing the edges against each other face side on the bench burnishing the high spots on the edges.
Keep them parallel when fettling using a panel gauge or calipers to do so.

[Robert Hazelwood]

You don't need to make three straightedges to insure precision. Nicholson's method from 1811 uses only two. Peter Nicholson was a cabinetmaker and a mathematician. I will outline it if anyone is interested.

It does not seem as if Sam is having trouble finding a straightedge; he knows where the high spots are, but needed to know how a plane worked.

I'll add a request for the Nicholson method.

I figure that if you match plane two sticks with a good plane, then compare the planed edges against each other, that ought to be pretty good. Match planing should eliminate most of the errors that would normally be revealed with a third stick. For example you shouldn't be able to create one convex and one concave edge if you are match planing. Any error would be duplicated on the other stick and doubled in magnitude when you place the edges together.

[Sam Shankar]

Y Looks like you are about 5 minutes of planing away from being done if its just the 6 thou on the sides to deal with.

Ha! It really is only 6 thousandths away, but that only means that YOU would be about 5 minutes away. This forum exists to narrow the gap between that and the time and effort (and lost thickness) I will require.

[Robert Hazelwood]

Ha! It really is only 6 thousandths away, but that only means that YOU would be about 5 minutes away. This forum exists to narrow the gap between that and the time and effort (and lost thickness) I will require.


Perhaps, but in any case now is a good time to relax if you are feeling frustration, by realizing you don't have a mountain of work ahead of you. Resharpen the blade really well, and set it for a nice fine shaving. Slow down and work methodically. Do a series of passes along the length over the center high portion, then check again. The 6 thou should be smaller now, reduced roughly by the thickness of your shaving. Repeat and check again. Double check flatness along the length and for twist. Etc.

[David Eisenhauer]

Lots of good, real world advise above. I am with Prashtun, the bench is 28" wide and how long? Hard to maintain perfect balance/control/pressure/etc across the width and down the full length. As Derek said, always identify and mark the high spots first for elimination and lots of times what remains is flat or flat-enough.

[Warren Mickley]

Warren, many of us would likely appreciate an outlining of a way to make a pair of straight edges.


jtk

Nicholson planed up two identical sticks as well as he could then fastened them together in the vise and planed the two edges together ("match planing". Any deviation from flat would then be doubled when one piece was laid on the other. If one were slightly hollow, the other would be also and the problem magnified when they were laid edge to edge. So by trial and error one could get them to where there was no gap between them. Nicholson did not recommend this for ordinary glue joints, just straight edges, where the extra time was not an issue.

Here is Nicholson:
430491430492

[Jim Koepke]

Thanks for posting this Warren.

jtk

[Sam Shankar]

It sure as heck took me more than 5 minutes, and a whole lot more than a few elegant swipes. (See the mess below.) But it's good enough.

I sharpened as well as I know how, and tinkered as best as I could, but I just can't reliably generate thin full-width shavings. In fact, my shavings seem wildly inconsistent. Wispy, half-width for several stroke then suddenly substantial, then back again. But by following the instructions here, I just rubbed away at the highest spots gently over and over, started over to fix the things I overdid, and eventually got to the point where I can't slip a 6 thousandths feeler gauge under the straight edge anywhere on the bench. So I'm declaring things done and getting back to making things that will leave the shop. Thank you!

For those who want to speculate on my plane tuning:
* It's a veritas bevel-up jointer. The sole is flat as far as I can tell.
* It's a 37 degree iron, 40 degree primary bevel with another degree of microbevel.
* I think it's pretty sharp for a beginner, at least sharp enough to generate the wispy shavings shown below.
* There's a fair bit of curl to the maple in the bench.

430535430536

[Tom M King]

Sounds like you did good. If you're taking thou and a half shavings, and there is a 2 thousands hump you're going over, you're not going to get a full width shaving. You're doing about as good as you can do. This is where the long plane comes in. It only hits the high parts. Theoretically, you should be able to get to the point where you're going to get full width shavings over the whole surface, but it sounds to me like you have it plenty good enough.

This pile of shavings got a lot deeper, but I didn't take any more pictures.

[James Pallas]

You should be fine now. Put the thought of it behind and make some stuff. You will soon have some marks deeper than .006.

[Jim Matthews]

* If I put a pencil in the middle of the table, it doesn't roll. (Round things don't roll either!)

It's probably close enough.

Your not building an interferometry table.

[Prashun Patel]

Is your blade cambered or square and flat? Is your blade square to the mouth or skewed?

I have that same plane and I find it possible to get full width shavings. (when planing with the grain that is).

Is your adjuster or set screws moving on you?

[Sam Shankar]

Is your adjuster or set screws moving on you?

Good questions. I don't know. Do you tighten the set screws so that they are actually snug to the blade when it's square? And I've been loathe to really tighten the cap screw--or whatever you call the thing that you tighten to secure the iron--because the directions are so clear about not overtightening it. I think the directions actually say 'no more than a quarter turn after it engages.' That feels like it leaves it pretty loose.

[Jim Koepke]

Good questions. I don't know. Do you tighten the set screws so that they are actually snug to the blade when it's square? And I've been loathe to really tighten the cap screw--or whatever you call the thing that you tighten to secure the iron--because the directions are so clear about not overtightening it. I think the directions actually say 'no more than a quarter turn after it engages.' That feels like it leaves it pretty loose.

My method for these adjustments is to first start with the cap screw. Set it lightly, then make shavings on the edge of a board first on one side of the plane and then the other. If the blade moves the cap screw needs a touch more tightness. Use increments of ~16th of a rotation to sneak up on the 'sweet spot.'

With the set screws, they should be tight enough so the blade can not move from side to side but can still be removed without having to loosen them.

jtk