Thread: Bed time!

will be following this...

I want to build my self a new king bed and want to do castle joints for the legs as well.

Originally Posted by havabeer69

will be following this...

I want to build my self a new king bed and want to do castle joints for the legs as well.

I'm including some of my stuff ups as well. Not a terribly complicated joint and it's gone pretty quick for me - definitely room for improvement though, I'm not too keen on some of the gaps. If I were doing a king sized bed I think my biggest hurdles would be the physical space required for test fits and moving parts around. I've got way too much stuff in my 2 car garage/workshop. I've gone with a lot of hand tool use but a router template to cut out the castle joint itself after bandsawing would give pretty good results I think.

This update has a positive start and and a positive end but a little hiccup in between.

So the next step was the cross laps on the header/footer. I may screw or glue these in to the castle joint since these boards are around 1.1m, so easy enough to move/store if required. I used a marking gauge set to half the width of the board for the depth of the cross lap and a morticing gauge set to the width of the boards and run down the endgrain of the board to give me my cross lap width to remove. I will be changing how I mark this out for the side boards though. Perfect time to break out the 12tpi cross cut tenon saw. Have I mentioned how nice a timber Qld Maple is? It saws and chisels just so nicely. The cross laps are easy enough to cut and having the morticing gauge lines for the width made it a breeze to chisel down. Again, offcut used for testing the width.

cross lap.jpg

Four cross laps later, the header and footer are done. I wanted to have these done and then do a test fit to get my cross lap positions for the side boards. Having cut the joinery by hand, I am guaranteed to be pretty far from perfection, so basing all my mark out lines on measurements alone would be a great way to not have parts line up. And speaking of not lining up, this is where I goofed. I have no idea how or why I changed my morticing gauge settings... I went to check that the castle joints all lined up and cross laps were good so far and found this:

out of alingment.jpg

So yea... the castles look hood and the footer board is looking great, but how did I shift the header board cross laps 5mm too close to the board edge! Same mistake on both sides too. I considered gluing a shim and the recutting the cross lap, but I think I'm going to live with it and see how tight the end result will be, if needed I will hammer down a darker colored timber wedge. If the client has complaints, I'll get him to speak to the manager

To rectify for the moment, I just traced the protruding timber with a knife and chopped it to the new line with a chisel.

Next step, cut the side boards to length, to give me 2080mm from inside edge of one leg to another. One of the side boards had bowed a bit, so I clamped the boards together and cut them with a track saw. On the board that is bowed, I'll put the crown facing upwards, so the cross laps should seat ok and for a bed, the couple of millimetres height difference between the sides is not going to matter.

cutting side boards.jpg

Yes, the TS55 and guide square aren't cheap but can't argue with the results. So any stuff ups in cutting these cross laps will be on me

side boards square.jpg

Now the exciting bit, taking over the entry way and dry fitting so far.

test fit.jpg

Happy to report that using the offcut to test the castle joint and cross lap widths worked perfectly. Boards needed just a little persuasion and in they went. It feels very rigid, I don't think robustness of longevity will be issues. I just need to make sure not to sand or plane too much of the boards - there is some extra thickness allowed for this but it's just a smidge as I don't want to have any of the castles crack on me, I've done that to enough dovetails

As for the error cutting the header cross laps, it is not that terrible and hasn't affected joints negatively (that I can judge). I think I will stick with the idea of a wedge and living with the mistake. Wabi sabi and all that.The thought of the gap killed me yesterday but I had a sleep and I'm over it now. I mean, I'm still impressed I managed to screw up marking out such straight forward joinery.

gap from alignment error.jpg

Overall though, outcome is ok, room for improvement but the frame is 4 cross laps away from being done. Using hand tools, there is a bit of finessing to get the castles right, if the 4 arms of the castle aren't parallel and the tenon/board inserted isn't of uniform width, there can be issues with gaps at the top of the joint or the bottom as they arms will be wedged apart. If my cuts didn't need to be so deep, this would be perfect to cut vertically on the table saw, which seems to be the predominant method on the old Youtube.


castle joint fit - Copy.jpg

So sometime next week - side board cross laps, slats, and finishing. I like the colors from the test fit so I think I'm going to go with a neutral look. Tossing up between blonde shellac, hardwax oil (could be too yellow? Maybe the white pigmented stuff), or water based poly.

Alkahestic

That is coming up very well. I have made several beds, but never used the castle method for the legs and rails. As it is looking as if I may have one more bed to make, I would like to try this technique, so I am watching carefully.

Regards
Paul

Originally Posted by Bushmiller

Alkahestic

That is coming up very well. I have made several beds, but never used the castle method for the legs and rails. As it is looking as if I may have one more bed to make, I would like to try this technique, so I am watching carefully.

Regards
Paul

Thanks Bushmiller. In retrsopect I think I should have cut the side boards and header/footer with tenons so there are shoulders on either side of the castle. The shoulders would cover any small gaps from not paring/chopping perfectly. I would have just had to focus on the ensuring no gap at the top. That said, apart from the silly layout mistake, the joint hasn't required a great deal of finesse and the test fit was very rigid, better than bed brackets by a good bit.

Originally Posted by Alkahestic

Thanks Bushmiller. In retrsopect I think I should have cut the side boards and header/footer with tenons so there are shoulders on either side of the castle. The shoulders would cover any small gaps from not paring/chopping perfectly. I would have just had to focus on the ensuring no gap at the top. That said, apart from the silly layout mistake, the joint hasn't required a great deal of finesse and the test fit was very rigid, better than bed brackets by a good bit.

"A"

The tenons would work to cover the inside joints, but no help with the outside unless it was made in the "log cabin" overlapping style. however, I don't think the new incumbent is going to be complaining very much. How are you going to find a three-year-old in that size bed? Be careful when stripping the bed in case he has had a lie-in! Maybe a tracking device could help.



Regards
Paul

I suggest you reinforce the short grain in the ends of the rails with a dowel. They're likely to snap off at some point without reinforcement.

Just drill them from the underside so you can't see them.

Originally Posted by Bushmiller

"A"

The tenons would work to cover the inside joints, but no help with the outside unless it was made in the "log cabin" overlapping style. however, I don't think the new incumbent is going to be complaining very much. How are you going to find a three-year-old in that size bed? Be careful when stripping the bed in case he has had a lie-in! Maybe a tracking device could help.



Regards
Paul

I should have been clearer - not exactly a normal tenon, but like a bridle joint that is not at the end of a board, so there are shoulders on both sides of the castle like fig 72 on this page Timber Frame Tools >> Bridle Joints. It would introduce the element of having to make sure that the shoulders are seated nicely but to my mind it would be easier to sneak up on the fit using a side rabbet or shoulder plane on a tenon that is few mm deep as opposed to tuning an arm of the castle that is about 25mm wide and 72.5mm long with just a chisel.

As for finding him in the bed, I'll have the prey come to me. He'll hear the crinkle of the foil around a chocolate bar and be by your side before you can blink, suddenly all full of good manners and politeness. But you're right, when I did the test fit he was also conducting his inspection and it really put into perspective just what a wee lad he is now. And how much space a bed takes up. We could save a lot of floor space going the Murphy bed or tatami mat route!

Originally Posted by markharrison

I suggest you reinforce the short grain in the ends of the rails with a dowel. They're likely to snap off at some point without reinforcement.

Just drill them from the underside so you can't see them.

Thanks, great idea and easy enough to execute. Plus I can try out my new dowel maker and report if it's any good!

Originally Posted by Alkahestic

I should have been clearer - not exactly a normal tenon, but like a bridle joint that is not at the end of a board, so there are shoulders on both sides of the castle like fig 72 on this page Timber Frame Tools >> Bridle Joints. It would introduce the element of having to make sure that the shoulders are seated nicely but to my mind it would be easier to sneak up on the fit using a side rabbet or shoulder plane on a tenon that is few mm deep as opposed to tuning an arm of the castle that is about 25mm wide and 72.5mm long with just a chisel.

As for finding him in the bed, I'll have the prey come to me. He'll hear the crinkle of the foil around a chocolate bar and be by your side before you can blink, suddenly all full of good manners and politeness. But you're right, when I did the test fit he was also conducting his inspection and it really put into perspective just what a wee lad he is now. And how much space a bed takes up. We could save a lot of floor space going the Murphy bed or tatami mat route!



Thanks, great idea and easy enough to execute. Plus I can try out my new dowel maker and report if it's any good!

Ok. I'm with you on the bridle joint and that too is now noted.

We had a child who would wake up from another room if he heard paper crinkling. Funny thing is he no longer has the sweet tooth! Must be a middle age thing.

Small update - cross laps cut on the side rails and the whole test fitted.

castle angled view.jpg

castle top view.jpg

Overall looks ok. Not perfect and gap less but I'll put it down chopping and paring better next time.

I'm really up in the air about how/whether to shape the legs or just do bevel or round over on all the edges and call it done. It's a bed so I don't want or need it to be the most decorative element in the little man's room and the square blockiness of it does have its own appeal...

Happy to report that with the castles complete, the frame is rock solid. I will not need glue on this one and the frame can be dragged around from one corner with no racking. So definitely would give the castle joint two thumbs up for sturdiness and ease of making. Just would go with tenons in the joint itself to enable hiding of some fumbles.

Hi Alkahestic,

Enjoying the build. I'd like to vote for bevels all round on those castle turrets!

Originally Posted by homesy135

Hi Alkahestic,

Enjoying the build. I'd like to vote for bevels all round on those castle turrets!

Thanks homesy - the consensus at home was bevels as well, so bevels it is!

More progress and we're getting close to done.

First step was to route the bevels along all of the edges except the castle interior. I went with a 45° chamfer bit in the trusty Makita hand held. This was the most comfortable position I found for working on the long boards - if you have any suggestions for alternatives please let me know. Didn't feel sketchy or dangerous as the hands and dangly bits were well clear of the router bit. Board held in vice, me at the end of the board and just riding around the edge. The side boards were way too big to handle on a router table.

routing edges.jpg

And the castle chamfered as well - there was a bit of burning on the end grain, almost none otherwise. The bit is just a Skil branded that is part of a set but hasn't been used much, so is still pretty sharp. Fortunately no blowouts but it was a very shallow cut.

turret burns.jpg

First attempt at clearing the burns was a sanding block with 120grit paper but I didn't like the scratched surface and that I'd have to work up the grits to at least 240.

turrets sanded.jpg

So instead, out comes the little #101 plane. It's tiny, the end of the leg is about 77x77mm. Worked really well for cleaning up the burns.

LN 101.jpg

And the planed turrets:
turrets planed.jpg

Now there are still some fuzzy bit at the base of the bevels. The fuzzy bits on the outside of the castle will be removed when it the surface is planed and I'm not touching the inside. The joints were left just a tad tight so I can plane/sand and still have a good fit. I don't intend to have a heavy coating of finish so I don't think the finish coating will impact the joint much - but we're getting to the point where we will find out! I think this part is also another point in favor of using 'tenons' inside the castle. One can prep the surface as much as they would like without impacting how the joint goes together, whereas I can not.

But on to what is literally the best part - smoothing of all the exterior surface. It's Qld Maple, it'd be a crime to just sand it, so out came the #4 1/2. This is the one of the worst faces, some light tear out but nothing too serious.

leg tear out.jpg

What can I say - sharp iron, closely set cap iron, and the shavings were shooting out of the mouth. It was about a minute per side to smooth. Have I mentioned just how much I like using hand tools on Qld Maple?

leg smoothed.jpg

So now, the boards are smoothed, the joinery is done and I really need to pick a finish to apply. I used offcuts to test - one thing to note is that the legs are made from a different stock that has a more reddish/brownish tinge than the sides/header/footer so will look a bit different.

I will be applying a second coat to these test pieces tomorrow (maybe even a third coat). They are:

1- Control piece (no finish) offcut from a side board
2- Tung oil on offcut from leg piece
3- Osmo PolyX Extra Thin clear
4- Haymes polyurethane (oil based satin)
5- Blonde shellac (2 coats)
6- Cabot's water based poly gloss
7- Osmo PolyX Raw (has white pigment for use on lighter timbers)
8- Tung oil

sample finishes.jpg

I'm going to drop #7 (Osmo PolyX Raw) as I don't think the white pigment will go well with the legs and even on the paler timber, it just doesn't do it for me. The water based poly changes the colors least. Trouble is, I like the look of the water based poly, oil based poly, tung oil, and blonde shellac... decisions, decisions. Maybe a couple more coats and a few sleeps will shed light on the matter.

A little talc powder in the dry joints will help prevent squeaking in the future.