Slight bow in edge when jointing long/large boards

[Brad Swanson]

I'm starting the Benchcrafted split top roubo workbench build and am in the process of milling my stock (8/4 ash). I spent one day last weekend getting my 8" Grizzly jointer dialed in, and I get consistent square edges the entire length of the 8-10' board when first milling the face and then the edge. Unfortunately, the edges are slightly bowed such that when I place a 4' straight edge anywhere along the edge I can fit a 4 thou or sometimes 5 thou feeler gauge in the widest part of the gap. I do have roller stands for infeed and outfeed support set up to be just below the plane of the tables. When I mill a board that is limited to the length of the infeed table, the edges are perfectly straight (straight edge test can't fit a 1 thou feeler between straight edge and board). Any suggestions for why the longer board is slightly bowed? Do I need my infeed/outfeed support to be precisely at the same height as their corresponding tables? Thanks.

[Bob Coleman]

I think the extra infeed/outfeed tables are probably too low. The board would be lightly tilted at the start, so you would not take a full depth cut from the very beginning. As you feed the board, it becomes easier to hold it onto the jointer tables, so you begin to take a bigger bite. Same goes for the other end.


Here's my method for jointing boards longer than my jointers infeed. Starting with the concave side, run it through until it stops cutting, then turn it end for end and run it through that way until it stops cutting. Repeat until you get a full length cut. You need your jointer set for a very fine cut to get perfectly straight boards, and really two people are required.

[Judson Green]

Can you pull the boards together with a clamp? If so I'd leave it. Sounds like a nicely sprung joint. Many try to achieve that as do I for glueing up panels and such.

[scott vroom]

Did you straight line one edge before jointing?

[Mel Fulks]

I agree with Judson, perfect the way it is, at least for the particular job you are doing now. I can't address what the table
extension is doing since I will not use them,prefer skilled helper. But most likely you have the out feed table just a little high. Maybe as little as a thousandth. I would leave it as is until it's a problem ,as the knives get a little wear on them the sprung joint might be increased and,or, possibly leave enough "extra" wood that you feel the board hit the edge of the out feed table. At that point you have to adjust.

[Mike Schuch]

I would start by making sure that when the knife in the cutter head is at the top of it's swing that it is level with the outfeed table. Put a straight edge on the outfeed table so that it projects over the cutterhead. Rotate the cutter head and make sure the knife barely kisses the straight edge at the top of the arc. Usually you will get a bow in the middle of the stock when the outfeed table is too low.

If the cutter head and the outfeed table line up I would suspect the problem to be technique. I have never used in feed and out feed supports. When I start feeding a board I push the stock down on the infeed table with my front hand while lifting the stock behind the table with my rear hand.

When the majority of the board is through the jointer I switch positions. My rear hand pushes the stock down on the table while my front hand supports the portion of the stock that extends past the outfeed table.

If you are relying on in feed and out feed supports that are set just below the table this is probably where the curve in the board is coming from.

[scott vroom]

If you joint a 10' bowed board on a 6' table it is nearly impossible to get a true edge. As the leading edge passes the end of the outfeed table it will begin to lose it's planarity with the table. The same effect will also occur at the infeed as the bowed edge climbs the front edge of the infeed table. I fixed the problem by straight lining one edge on the TS before jointing.

[Brad Swanson]

I may be less concerned if the boards were to be assembled edge to edge for assembly, but they will be glued face to face for the two top slabs, so the "smile" of each board will be facing up (and down if I run them through the planer without fixing). Scott, I did not straight line rip before jointing because my track saw doesn't have the cut depth to cut 8/4 (Festool TS55). The first board I ran the edge "frown"-side down, the second I ran the "smile-side" down, both gave the same result. I may try snapping a chalk line and "free-hand" ripping on the band saw first next time before edge jointing. Other recommendations for straight line ripping would be welcome (I don't have a circular saw, but I could use the Festool track as a reference for a jigsaw (which would be painfully slow, and probably not cut as square as I'd like it to be). Fortunately both boards milled with this problem have plenty of width left to correct it and still be 4.5" wide for the benchtop slab assembly.

Mike, I think the table alignments (cutterhead to outfeed and infeed tables, coplanarity of tables) are pretty good, as a 5' long board jointed in the same fashion gives dead-nuts-on straight edges and perfectly square edges.

Thanks very much for all of the comments. I appreciate them all.

[Kent A Bathurst]

Judson + Mel are correct - often done to ensure tight joint when pulled together under glue-up. 4 - 5 thou means you are off by 2 thou per 96" - 120" board. Plant the flag, claim victory - that's my advice. ;)


On the issue itself - the outboard supports are not level with the table. I dunno about the "just below" bit - you are trying to lengthen a dead-flat surface, so those rollers should be a continuation of that surface/plane. When I do this, I clamp a straight edge on the outfeed [for example] and raised/bump my roller stand until it is just barely touching.

Also - the outfeed support does not need to be way downstream - when your board clears the cutterhead, you want the roller past the mid-point, but only by a few inches. I have found that the closer to the real table I keep it, the easier it is to get it level, and I get good, repeatable results.

I also have found - big surprise - that the concrete basement floor from 1920 is not dead-nuts level. :eek:

So - I am careful during setup to get the roller "level" along the line the board will be taking as it exits the jointer. That spot on he roller can easily be a few thou+ different elevation than the ends of the roller.

EDIT: Started typing after I read the post by The Vroom Dude [tm]. Sorry if I duplicate the thoughts of the intervening posts - they are quicker on the keyboard than I.

[Mel Fulks]

I don't think I've seen a bench more accurate that what you describe now. I would glue as is ,I think it will be fine .

[scott vroom]

I wouldn't settle for a face joint glue up with bowed boards.

A 10' arc passing over a 6' straight surface will not render a straight edge. You need to either straight line the edge before jointing, or build infeed-outfeed extensions that are planar. Using a single outfeed and infeed roller will not fix this problem.

[Pat Barry]

I wouldn't settle for a face joint glue up with bowed boards.

A 10' arc passing over a 6' straight surface will not render a straight edge. You need to either straight line the edge before jointing, or build infeed-outfeed extensions that are planar. Using a single outfeed and infeed roller will not fix this problem.
I think if you are going to straightline them as suggested here and you are doing the face lamination as described, then I don't see ANY benefit to using or worrying about the jointer for this edge. After lamination you will need to dress the bench surface anyway, whether it be a power sander, or planer, hand planing, router jog, what have you. The face surface of the bench will never be true enough anyway so alll the work to get perfectly straight edges is moot.

[scott vroom]

I think if you are going to straightline them as suggested here and you are doing the face lamination as described, then I don't see ANY benefit to using or worrying about the jointer for this edge. After lamination you will need to dress the bench surface anyway, whether it be a power sander, or planer, hand planing, router jog, what have you. The face surface of the bench will never be true enough anyway so alll the work to get perfectly straight edges is moot.

Pat, from a practical standpoint I would tend to agree with you.

The OP PM'd me asking how I straight line on a TS. Seems SMC doesn't allow pics in PM's so I told Brad that I'd post a couple here showing my setup. You can substitute hold downs for the L-brackets and screws shown in my setup. All that's required is a length of MDF with one straight edge to ride against the fence and a method of holding the stock to the MDF. The first 3 pics show the initial straight line, the last pic is shows the other edge being squared up with the straight lined edge.

[Brad Swanson]

Brilliant! Thanks Scott!

[Peter Aeschliman]

You've got correct answers here, but I'll just add another vote. If your jointer is producing perfectly flat, square, and straight surfaces for shorter pieces, then it is probably as tuned in as it can possibly be. If you're working with a long workpiece that has a meaningful crook, there's a point where the jointer can't remove all of it because the leading edge of the workpiece will drop toward the ground as it exits the out-feed table. You could maybe try to rig up some roller stands, but I haven't had much luck getting much precision using roller stands for anything. Best thing you can do is find alternate means of removing the crook before going to the jointer (per Scott's post about straight-line ripping on the TS).

But .004" over 4' is really darn good frankly. So IMHO, in your case I would not worry about it at all since you're going to have to flatten the work top anyway... and even if the pieces were perfectly straight, you're going to have more than .004" of variation due to the imprecision of glue-ups...

So, you're good man. The only thing that you should be worrying about is posting pictures of your build. :)

[scott vroom]

If you're working with a long workpiece that has a meaningful crook, there's a point where the jointer can't remove all of it because the leading edge of the workpiece will drop toward the ground as it exits the out-feed table.

Yep...and the same dynamic in reverse will occur as the trailing edge, starting off lower than the infeed table, rises as the board moves across the table. It's simple physics and a single roller stand positioned at each end of the machine will not correct the issue. You must either extend infeed-outfeed tables planar to existing, or straight line the board prior to jointing.

[Bob Carreiro]

Over 40 yrs ago at a vocational HS, I was taught how to run warped boards over a jointer with the crown up or down. Jointer or board length do not matter. With my 6" Ridgid, I recently edged some 5/4 Hard Maple, some with 1"+ crowns that were all 12' long. Not easy handling boards that long and heavy by yourself, but it CERTAINLY can be done. Need I review the process to this professional audience?

[scott vroom]

Over 40 yrs ago at a vocational HS, I was taught how to run warped boards over a jointer with the crown up or down. Jointer or board length do not matter. With my 6" Ridgid, I recently edged some 5/4 Hard Maple, some with 1"+ crowns that were all 12' long. Not easy handling boards that long and heavy by yourself, but it CERTAINLY can be done. Need I review the process to this professional audience?

One way to do that is to feed alternating ends of the board, a short length at a time each pass, until the arc begins to flatten to the point where the crown portion will not breach either the infeed or outfeed tables. What is your process?

[Bob Carreiro]

Correct Scott... for running edges crown down. For running crown up, place the crown center on the infeed table so both board ends are level to each other and joint maintaining boards orientation. When cutter contact stops, place fresh edge on infeed and repeat. It's not difficult. I've been doing it all my life.

[scott vroom]

Correct Scott... for running edges crown down. For running crown up, place the crown center on the infeed table so both board ends are level to each other and joint maintaining boards orientation. When cutter contact stops, place fresh edge on infeed and repeat. It's not difficult. I've been doing it all my life.

Bob, yeah I'm with you on that. For me, I was looking for a solution not so neander mostly because I was in a hurry so I came up with the straight line sled idea.