How To Edge Joint

[custard]

Edge jointing means joining up narrower pieces of wood to make a wider board. It's something you have to do with nearly all furniture projects, because most boards sold today are less than 300mm wide, and the majority are in the range 150-200mm wide. So if you want a table top, or a chair seat, or a panel for frame and panel work, then you'll have to learn to edge joint.

When it goes as planned it looks like this. You start with a board that has at least one true face, and you then cut an edge that's at exactly 90 degrees to the reference face all along its length,

You also need this edge to be dead straight, or possibly with a minute hollow in the centre, but on no account must it have any trace of a "bump" when tested with either a wooden or metal straight edge,

When you've repeated that process on a second board they should sit together perfectly,

Which means when they're cramped for the glue up there's an invisible glue line and you can rest a straight edge across them, and the resulting table top (or panel, or whatever) is perfectly flat with no visible gap,

So, how do you set about doing this?

There are plenty of different approaches, but I'll take you through the approach I use, which I know is shared by many other cabinet makers.

I guess the first thing to say is that the very best glue lines will come from a sharp bench plane. With the right machinery you can produce a good edge straight from a machine,

But if you look carefully at a planed or sawn edge you'll still find traces of machine marks, these for example are the tell-tale ripples from a planer,

And that's from a good planer, where the knives are always kept sharp. But most hobbyists aren't in that position, they may have a planer with a fence that has a slight twist along it's length, or the knives are bit blunt, or it's prone to snipe. Or they may not have a planer at all, and be relying on a lunch box/bench top type thicknesser, or even just hand tools.

That shouldn't be a problem. For many years I had a small workshop that could only accommodate a bench top thicknesser and a band saw, yet I managed to produce plenty of edge jointed components to the highest standards. And with a bit of application so can you.

I've reached the photo limit for a single post, so I'll continue in the next post.

 

[custard]

Edge jointing will really test your hand planing skills. So even if you've got your plane in reasonable shape and are fairly content with your planing skills, you'll probably need to keep an open mind that you may have to take those skills up a level.

For example I know many new woodworkers (and even some not so new ones!) struggle when starting the cut. They'll tend to let the plane droop down at the back which produces a bit of a divot at the beginning of the cut like this,

If you see this happening then try sharpening the iron and setting a lighter cut. But most of all, get into the habit of pressing down firmly with your thumb, directly centred over the board's edge, with your fingers curled around underneath to act as a fence,

If there's even a trace of unevenness at the start of your planing cuts you'll really struggle with edge jointing, the quality of the cut needs to be immaculate from the very start to the very end. Within reason you can edge joint with almost any plane from an 04 to an 08. But the extreme shortness of an 04 plane's sole ahead of the iron means it's harder for a beginner to master edge jointing with an 04. This photograph of an 04 and an 05 illustrates the point,

You'll find it easier to edge joint with a slightly longer plane, which is why schools always used to teach woodworking with a number 05. At the other end of the scale an 07 or an 08 are pretty unwieldy, the 08 especially. You might think these ultra long planes will guarantee quality results, but if their weight and size compromises your planing techniques then they are counter productive. In these photos I'm using a Lie Nielsen 07, but it could just as easily be a Record 05 1/2 or a Stanley 06, they'll all deliver exactly the same results.

The next challenge is to get the right camber on your iron. This is particularly tricky as you'll only get there after quite a lot of trial and error, unfortunately there's no magic formula that will automatically produce the optimum camber for edge jointing. I have heard some people say a 37" or 38" radius camber works for them, which on a 2 3/8" wide iron would mean around a two thou gap at each side of the iron compared with the centre. Personally I prefer a very slightly more aggressive camber than this, and in any event I've never tried to achieve it "by the numbers". I simply apply more pressure to the edges while honing, until I'm happy with the results. After that I try and maintain that camber by honing with a consistent number of strokes while applying pressure to five points along the iron's edge, so perhaps ten strokes while applying pressure on the far left, ten strokes while applying pressure to the centre left, ten strokes while applying pressure to the centre, and so on.

The best account that I've read of cambering an iron, and the role it plays in edge jointing, is in volume two of David Charlesworth's "Furniture Making Techniques".

One test I apply is that with my plane set for a very fine cut I'll be able to run down a typical 20-25mm wide board and the shaving won't quite span the entire edge, but it will feather out imperceptibly on both sides, like this,

I'll then reset the plane for a slightly coarser cut. But taking great care that the apex of the camber is right in the middle of the plane's sole. If you can't reliably do this by eye the try using a sliver of scrap wood to stroke down the sole of the plane, checking how far in from both edges you are before you start taking cut.

You can tell from this how a plane's camber is both critical and also a personal preference. You'll often tend to adjust that camber at every honing, so it's worth spending a fair bit of time experimenting to find out what works for you.

 

[custard]

You'll recall that successful edge jointing means achieving two tasks. Ensuring the edge is at 90 degrees to the reference face, and also that the edge itself is straight or perhaps very very slightly hollowed. The purpose of the camber on the iron is to help us achieve the first objective, ensuring the edge is at 90 degrees.

To illustrate how this works look at the following example. Here we're testing the edge with a square, with the stock resting on the edge we can see a gap next to the reference face. Importantly this gap is at the top of the blade.

This means the far side of the edge is higher than the front. One way of being sure of this is to imagine an even bigger gap at the top, to produce a bigger gap you'd tilt the square towards the front, therefore lifting the back of the stock up and off the far edge. So the challenge is to plane away more of the back edge than the front edge. We'll remind ourselves of what needs to be done by applying a pencil mark on the high side, like this,

In practise you'd repeat this check at about 6" intervals all along the top edge, leaving pencil marks accordingly.

Now if we shift the plane over towards the right hand side the camber means we'll cut deeper on the right and shallower on the left. Remember we'll be using our fingers curled underneath the plane as a fence to ensure we're consistent throughout the cut,

And you can see the proof of this with this shaving. It shows the pencil mark that's been removed, but the cut fades out to nothing on the left,

In reality I wouldn't move the plane quite this far too the right, I prefer to take a shaving that runs right across the edge. I do this to avoid the risk of the edge becoming faceted, but I show this exaggerated cut to demonstrate just how effective the technique is.

We'd check again with the square, marking the high side and repeating as necessary, until finally the edge is dead square to the reference face all along its length. Here you can see there's no gap whatsoever, at the top or at the bottom,

Sometimes you'll be faced with a top edge that's in wind, ie it's slightly twisted so it might start with the left side high, but finish with the right side high. In that case you'd have pencil marks on the left at the beginning, but on the right towards the end. Faced with this (and it is quite common) you have to move the plane across while you're making the cut. So you'd start with the plane over towards the left, but finish with the plane over towards the right. All I can suggest is you keep practising these strokes until you're confident you can execute it smoothly and accurately.

 

[screwpainting]

Thank you Custard, after an horrendous day, your thread has me restored.
=D>

 

[LancsRick]

I had no idea about the camber for edge jointing, and had always assumed it was preference of grind for scrub planes. How wrong was I...

Thabk you again Custard, hugely informative.

 

[custard]

The second challenge in edge jointing, after getting the edge at 90 degrees to the reference face by using the camber, is to ensure the top edge is dead straight or possibly slightly hollow, but under no circumstances must there be a high spot.

The problem is that planes don't always plane a dead flat surface. In particular hand pressure alone can flex the sole by a few thou, which means it may conform to a pre-existing curve in the top edge. With practise you'll get better at planing flat, but one way to guarantee the minute hollow that delivers a sprung joint is to use "stopped shavings".

This is a technique where you start the cut about half an inch in, and finish the cut the same distance before the far end. It sometimes helps to mark the top edge like this, and thereafter aim to retain those marks at both the start and finish,

I should emphasise that the way to stop the shaving at the far end is to lift either the front or back end of the plane by a fraction of an inch, just before you judge that the cutting edge of the iron is about to remove your mark. Don't just stop and lift the plane away or you'll find a long shaving is still attached to the workpiece! Getting this right, stroke after stroke, takes some practise, but don't get despondent if you find you miss and plane away the end mark, just pencil it back in and within a few strokes you'll be back to where you should be.

Imagine it as deliberately planing a hollow into the edge. If you repeat this for eight or ten strokes you'll find the plane simply stops cutting. The edge has become so hollow that the length of the sole prevents the iron making contact, first in the absolute centre then progressively outwards from that point.

Once you've arrived at this position you take through shavings (as opposed to stopped shavings) until you get your first full length shaving. I normally find that one, or possibly two further shavings gets me to the exact degree of hollowness I want. If you prefer a dead straight edge then stop at the first complete, full length through shaving.

You test with a straight edge, either metal or wood as you prefer,

Personally I rarely check for light under the straight edge. Instead I pivot the straight edge about the centre and feel and listen for feedback.

If it pivots easily, and silently, then there's a high spot in the middle. Heavy resistance or a complete refusal to pivot says the opposite, it's too hollow. But a modest resistance with a small scratching sound means it's right where it needs to be.

As you can tell from this account, a lot of this is about feel and judgement, things that only arrive with practise. Furthermore, there's the frustrating possibility that during the stop shavings you'll loose the precise 90 degree angle, so then you have to start again.

In reality I'll tend to combine elements of the two stages together, and ideally when I've finished getting the 90 degree angle I'm actually already at the degree of hollowness I want, or very very close to it.

You do this for each edge of the two boards then bring them together and check the final result. Repeat the same pivoting test you did with the straight edge, and check that the two faces are in a perfect 180 degree line, if they tend to tip forwards or backwards you've more work to do.

When you first try this I can pretty much guarantee you'll be tearing your hair out. But believe me, it does get progressively easier and a lot lot faster!

However, if you feel it's all to difficult I'll follow up with some posts giving some alternative and easier options for edge jointing.

 

[custard]

If you don't have a decent planing machine you can still machine a reasonably accurate edge using just a router.

One way is to do the job directly on a router table. The difficulty here is you need a split fence, with the outfeed fence being exactly in line with the cutter, and the infeed fence being set back according to whatever depth of cut you want. Setting the outfeed fence precisely is certainly possible, but in practise it isn't as simple as it sounds. And unless it's absolutely accurate, you'll get some snipe to your machine edge, meaning you'll then have to cross-cut off a few inches from each end.

A slower, but surer outcome, can be achieved using the router hand held.

You'll firstly need a bearing guided straight router bit, like one of these,

These are top and bottom bearing cutters, the top one has a skew cut which is cleaner but they're more expensive. The bottom one has disposable blades, so there's no excuse for not having a sharp bit!

You'll run the bearing against the "factory edge" of some MDF. The factory edge is the name given to an edge that was cut by the manufacturer, you can identify it because it often will have words or a code stencilled on it. You can be confident that it will be absolutely straight, so straight in fact that you can even use a factory edge as an inexpensive straight edge. The only thing to watch out for is it mustn't have been bumped or dented.

You need a piece that's at least eight or twelve inches longer than your workpiece. It's critical when copy routing (which is basically what this technique is) to have a "lead in" and a "lead out" for your router, in other words the MDF template must overhang the workpiece at each end. When you've got your ideal bit of MDF place the workpiece on your bench with the edge over hanging by half an inch or so. Then place your MDF template on top, but set back from the workpiece's edge by a whisker, say about 1/32" or 1mm. It should look like this,

The problem you'll probably have (assuming you don't have fancy kit like vacuum clamps) is work holding. So here's a method that works and doesn't require specialist equipment.

You're going to copy route the edge in stages. Start with the cramps off-set to the right, so you can cut the first stage,

You'll only probably cut about 8-12" before you get blocked by a cramp. So stop and switch of the router.

Now here's the critical bit. Before removing any cramps you must first add one cramp at the far left, like this,

Only now can you remove the cramp that's blocking your way, like this,

Then you pick up the copy routing from where you left off,

You progress along the edge in stages, always remembering to add a cramp before removing one.

If your workpiece is very wide, and if you have a bench configured so that you can cramp from the back, then you might be able to avoid this stage by stage approach. But for most jobs this is how you'll have to do it. It might be slow, but you'll get decent results, almost certainly better than you'd get from a circular saw for example. However, it's still a machined edge, so it won't be quite as good as an impeccably executed edge joint with a bench plane. But if you really can't spare the time to learn how to edge joint by hand, then this is an acceptable option that's well worth considering.

A couple of final points regarding router cut edge jointing. Firstly, don't use Solo Cramps, which are this style of cramp,
https://www.axminster.co.uk/axminster-t ... p-ax364432

Solo Cramps are a good cramp, cheap and with plenty of workshop applications, but they have a tendency to let go under vibration. So for securing templates use F Cramps or good quality Speed Cramps.

Secondly, you can use the MDF factory edge approach with a router table, the quickest way is to use double sided tape to stick the MDF template to your workpiece. With the same style of bearing copy bit in the router, you can then run it straight through on your router table without needing a split fence or any off-sets. The only thing to be aware of is that many commercial workshops are now frowning on the use of double sided tape to attach templates on safety grounds...but it's your call. Your workshop, your rules!

 

[custard]

Another alternative for edge jointing is to shoot the edge.

You're probably more used to shooting boards than run across the bench, like this,

Obviously this arrangement makes it impossible to shoot any workpiece longer than the width of your bench. Which is why I tend to prefer shooting boards that run along the bench, like this,

It's far easier to make a much longer shooting board in this fashion, which may well allow you to edge joint on it. For example, this is how I edge joint drawer bottoms, which are 8mm thick boards of Cedar of Lebanon, a bit too thin to edge joint any other way.

Building on this general principle it's possible to put together a very simple shooting board type arrangement that removes a lot of the skill from edge jointing, but still delivers the impeccable results only achievable with a bench plane.

Start by laying a long length of Birch ply or MDF on your bench like this,

Then cramp your workpiece on top of it like this,

As long as your bench is reasonably flat you can then progress through the edge jointing stages previously discussed like this,

For example here's one way to test for straight by supporting a straight edge on small blocks,

If your bench allows cramping from the rear you can reverse the entire arrangement like this, which is more convenient

And if your bench isn't that flat then you can place a length of MDF underneath everything for the plane to run on. I can guarantee this method works very well, it's basically how I edge joint saw cut veneers or other very thin but long components, but it works just as well with thicker workpieces. Going this route you don't need a cambered iron, as you can adjust for square using the lateral adjustment on the plane. However, personally I still prefer a cambered iron, I just take great care that the apex of the camber is perfectly in line with the centre of the workpiece.

 

[screwpainting]

This should be part of a book Custard, it really should. Passing on and explaining skills in simple terms with such clarity is a rare gift.
Very professional, thank you.

 

[galleywood]

+1 for what 'screwpainting' said.
Thanks

 

[LancsRick]

Thinking it through in my minds eye custard, I assume you must have a plane where you are certain the side is perpendicular to the sole. If you try to adjust by lateral blade adjustment presumably its going to be a heavy cut to maintain 90 degrees?

I always ensure I flatten the soles of my planes but I never check for the sides being perpendicular!

 

[Sideways]

Superb post Custard. Thank you.
This feels like the start of a course on fundamental woodworking / furniture making skills
For me at least, it's pitched at exactly the right level.
You've done your good deed for the day + week + month in writing this. Cheers !

 

[Buckeye]

Awesome again Custard. Thanks so much for sharing your learning and skills on this site in such a practical and easy to follow way

 

[custard]

I assume you must have a plane where you are certain the side is perpendicular to the sole. If you try to adjust by lateral blade adjustment presumably its going to be a heavy cut to maintain 90 degrees?

I've never checked if the sides of my bench planes are 90 degrees to the sole (a shoulder plane is a different matter entirely, there it's critical). Honestly, as far as bench planes are concerned don't worry about it, I guarantee the technique works!

 

[Just4Fun]

Custard, can you please comment on a technique I use that you have not mentioned?
What I do is to put 2 boards side by side in the vice, and plane the 2 together. Then if I don't plane at exactly 90 degrees it doesn't matter because the error on one board has an exactly compensating error on the other board. I still try to plane at 90 degrees but don't worry too much about it, so long as I am somewhere near. I find this quicker, partly because I don't spend time getting to exactly 90 degrees and partly because I plane 2 boards at the same time.

 

[LancsRick]

I don't doubt you for a minute custard I just can't figure it out in my head! Unless the next board you put in upside down so Amy variance cancels out?

 

[thomashenry]

Great thread Custard.

 

[AndyT]

Agreed, that's a lovely clear explanation. Thanks again for what must have been a non-trivial amount of time and effort.

If I could add one little tip from my own experience, if you are making a table or something else with an edge jointed top, make the top first. Then measure it and size the supporting parts to suit. If you need several goes at getting the edges to match, your top may end up somewhat narrower than you expected - at least, mine did.

 

[Sheffield Tony]

Custard, can you please comment on a technique I use that you have not mentioned?
What I do is to put 2 boards side by side in the vice, and plane the 2 together. Then if I don't plane at exactly 90 degrees it doesn't matter because the error on one board has an exactly compensating error on the other board. I still try to plane at 90 degrees but don't worry too much about it, so long as I am somewhere near. I find this quicker, partly because I don't spend time getting to exactly 90 degrees and partly because I plane 2 boards at the same time.

This is what I was taught to do at school - and with thin boards at least, it seems easier to plane square when they are doubled up. But I guess with thick boards and a cambered iron, the effect of the camber becomes less useful ?

 

[John15]

Thanks for the above posts Custard. I struggle with the edge joint and it usually takes me several attempts to get it right. Often my problem is a gap at each end. I haven't tried the hollowing technique before but will do next time.

John