Joinery Question: How Would You Accomplish this Notch?

[John Crawford]

Hi Everyone:

I'm working on a coffee-table build here. I'm planning to lift an element from a Krenov table that I like (see below): a split top table, which has stretchers rather than aprons on the ends. Krenov writes that the split table top is supported by a "cross-piece partially notched into the sides." See pictures of the table below, with arrows pointing to where the supporting cross-piece protrudes from the side aprons.

My question: How would you accomplish the notch that accepts the cross-piece? Would this be a variety of dovetail? My main concern is that I would like the cross-pieces to add some rigidity to the frame, to compensate for the lack of the end aprons. I'd also be open to hearing suggestions that don't require the cross-piece to protrude through the apron.

The table is being built in walnut with neander methods. I'd like to do the cross-pieces "nicely," meaning I'd prefer not to screw them on with L-brackets

Thanks for your suggestions!

Krenov table.jpg

(from Krenov's Cabinetmaker's Notebook)

[Nicholas Lawrence]

What is the thickness of the apron? If you are under 4/4, I think I would simply notch it as it is in the photo.

To make the notch in the photo, I would cut the shoulders carefully with my best backsaw (mark with a knife, deepen the mark, use a chisel to cut a "V" shape on the inside edge to form a shoulder for the saw); and clean out the waste with a chisel and router plane. If I did not have a router plane, I would pare carefully with a chisel (pare halfway down, flip the board, pare the other half), and once I got close I would perhaps finish the bottom with a shoulder plane. I have a router though, so that would be my first choice.

[Prashun Patel]

I would cut the sides with a saw, then I would use one of two ways to complete:

Coping saw, then chisel,

Or

Saw more vertical kerfs then chop out the sticks starting from the end.

The grain direction will make the paring very easy. Come in from both sides of the apron to avoid blowing out the insid

[William Fretwell]

The way its done in the photo if the cross pieces have a shoulder inside the apron they would add lateral strength to the frame but I can't tell from the photos.
You could use a blind dovetail and a large shoulder short of the apron. Even a small rabbet for the shoulder with a bottom edge pocket would add strength for the day some 200lb oaf stands on the table.
If you want to go crazy make the cross pieces L shaped then you add lateral and longitudinal strength. The top flat edge of the L could be blind dovetailed into the top of the apron and the bottom of the vertical part of the L just dowelled to the apron.

If you make the L into a U with two vertical components then several oafs could stand on the table.

[John Crawford]

Thanks for the suggestions. The rough stock for the aprons is 4/4, so they will end up a little thinner than that.

Nicholas and Prashun: It sounds like you all are suggesting a square notch (with a shoulder on the cross-piece), rather than some sort of dovetail-shaped notch?

William: Interesting ideas; I like the blind dovetail idea (I haven't tried this before). Do you think an apron that is shy of 4/4 would be strong enough? I don't quite get the "U" idea; I do have an 8-year-old who won't be encouraged to stand on the table but may end up there!

Chris

[Patrick Chase]

Nicholas and Prashun: It sounds like you all are suggesting a square notch (with a shoulder on the cross-piece), rather than some sort of dovetail-shaped notch?
Chris

You have a long-grain to long-grain interface at the bottom of the notch (assuming you cut it accurately) so I don't think you'd need to dovetail it. That joint doesn't carry racking loads in this design as far as I can tell.

[William Fretwell]

I don't know the size or proportions but walnut is strong. That Krenov table works at a small size (but I hate the legs). You can certainly make it hold coffee!
My own coffee table has a 3/4x2" stretcher glued to the apron bottom. The stretcher is mortised to the legs as is the apron. On the drawer side an extra stretcher at the top is single blind dovetailed to the top of the leg. It's a good look and very strong without looking heavy. It's fun to have three joints at each end of an apron, but that's why we make our own stuff!

[Gene Davis]

Do you know how Krenov fastened the battens to the underside of the top slabs? I used to own all his books, but lost them.

I am wondering how he dealt with seasonal wood movement.

[William Fretwell]

That is true, it could just hang in the notch and still function as a top support.

[John Crawford]

You have a long-grain to long-grain interface at the bottom of the notch (assuming you cut it accurately) so I don't think you'd need to dovetail it. That joint doesn't carry racking loads in this design as far as I can tell.

Patrick: Thanks for your observation here. Racking is really my main concern. I have prototyped some of the pieces, but I didn't make a prototype of the whole frame, and I was trying to imagine if a joint here was necessary to prevent it.

If you have a second, could you explain about the long-grain to long-grain interface? I think there's a structural principle there I need to learn.

[John Crawford]

William: ha ha, I love the "gumby" legs. I made numerous models in scrap, attempting to imitate them, but decided to go with a more ordinary taper. The table is only about 17" high, not enough to achieve the nice (to me) swoop that he gets. Mine just looked sort of short and piggy.

Gene: Unfortunately he doesn't really provide details on the table, other than to say that he was just being playful with the split top, and realizes that it is not for everyone.

[Nicholas Lawrence]

If you have a second, could you explain about the long-grain to long-grain interface? I think there's a structural principle there I need to learn.

He is assuming you are going to glue it in. Glue joints are strongest when you have long grin glued to long grain (as at the bottom of that notch). Long grain to end grain is weaker (the sides of the notch), and end grain to end grain is weaker still.

[Gene Davis]

I built a table like that but not with a split top, and I used a different detail for fixing the top to the frame. Mine was sofa-table size, 17 x 48 x 29 inches high.

Looking at the original poster's images from the Krenov book, I have re-designed (for me) the way I would do this. See the attached images.

The fixing cross-bars are dovetailed to the top long rails, while the top feature of those cross-bars is a pair of sliding dovetails (a single that's been cut away at its center), and a pair of holes bored for a dowel or screw that fixes the top board in place laterally.

My model has goofy extra lines on the legs because of the way Sketchup dealt with the curves and surfaces. Try to ignore them and focus on the overall shape.

I did not model this feature, but Krenovian design logic would dictate the end surfaces of the dovetailed ends of the crossbars be pooched out a little. James would have extended this face about 1/8" and then planed and filed three of the arises back to "pillow" the ends.

[lowell holmes]

I would pin it with a vertical dowel into the stretcher.

[John Crawford]

Gene: Thanks for taking the time to sketch this up. I was thinking about a dovetail arrangement like yours, but Patrick's comment above has me wondering if it is necessary. Perhaps I'll wait until the frame comes together, and see how sturdy it is before the cross pieces are on. I do like the "pooch" detail you mention, and will likely steal that as well, if I go this route.

It looks like your table has the top floating a bit over the aprons, is that correct? I think mine will probably sit atop the apron. I notice the gap between you sliding dovetails will also mute the view of the cross-piece (which can be seen through the split top); I've been playing with ways of doing that.

Did you make the "gumby" legs on yours?