Garden Gate

[Dewald Swanepoel]

Aah, gotcha. Makes perfect sense and I think that's what I'll do. I'll just have to get the right router bit first. Only one I have has a bearing at the tip. I'm gonna need one with a bearing at the base.

[Janello]

When you buy it, consider the depth you need for the tennon...you can play with the template thickness too if need be. A shallow mortise bit may be the ticket.

[Dewald Swanepoel]

Sure. Thanks.

[wtnhighlander]

Looking great, Dewald! That is some lovely lumber you have chosen.

[Dewald Swanepoel]

This is slow going as it is, so a 1 year old with tonsillitis isn't helping. But I've finally found some time for the workshop again and have some progress to report on.

The next big step, which I took with some trepidation I might add, was to cut the tenons for the vertical boards to join with the curved mortises. I decided to go with @Janello's advice and router with a flush trim bit against a template. First the tenons were cut roughly with the band saw.

Then I traced a template on MDF and cut roughly with a jig saw and sanded down to the line

And then the router work against the template. It worked like an absolute charm.

This all worked wonderfully but of course I had some cleaning up to do with the chisel to tidy up the short shoulders. And then disaster struck. I went in a little too hard and split the board a good 3cm down the length of it. So I thought I'd try a trick that I've only read of but never tried before. Pry open the crack, taking care not to crack it further. Apply glue generously from one side while holding the shop vac hose up against the other side. The shop vac sucked the glue nicely into the crack and I could clamp the failure and wipe of the mess.

I forgot to take a picture of the fix afterwards but, suffice it to say, right now I can't tell which of the four boards is the one that underwent this fix.

Anyway, I tidied up all the shoulders and dry fit the tenons to the mortises and... Wow! I was really worried about these joints but they turned out better than I had hoped for.

OK OK, you got me. Not all of them were that snug and required some fine tuning first

And here's the final result with all four joints dry fitted

The next step will be to cut the top curve on the yellow board and the lap joints where the green boards join with the dark blue board - for those who can still remember the colour coded design

[Janello]

Looking awesome man! Great repair job and shop vac trick!

 

[Brendon_t]

Great looking joints and nice fix.

Whenever chiseling into end grain,  take it slow. I'll usually use hand pressure if at all possible.

[Dewald Swanepoel]

42 minutes ago, Brendon_t said:

Great looking joints and nice fix.

Whenever chiseling into end grain,  take it slow. I'll usually use hand pressure if at all possible.

Thanks and yes, I think I've learned that lesson now.

[Chet]

Dewald the fit on the motive and tenon joints look great. Nice work.  Cool trick with the vac

[Dewald Swanepoel]

Alright, I've got some catching up to do. The last month has been hectic on this side but I did get some work done in between two toddlers making respective visits to the ER and a 3rd birthday party that had to be planned, prepared and executed.

So next order of business was to cut the cross lap joints into the mullions (green boards). This was easily achieved by laying them side-by-side, setting up two fences and routing out the material with incremental passes.

I was very careful not to router these lap joints too deep. The plunge depth stop on this Ryobi Router of mine is not the most precise and so I rather cut the laps a smidge too shallow and then paired away the fibers to the exact depth with a chisel.

And so here is the dry fitted gate so far.

Then to the router table where I cut grooves for tongue and groove joints into the stiles (red boards), mullions (green boards) and bottom of the mid-rail (dark blue board).

To end up with this

and this

The tongues were also readily cut with the router table

And here are the tongue boards (teal boards) and mullions (green boards) laid out.

I've test fitted them and it's a bit snug at the moment. I'll have to shave off some more wood before assembly. But I also think I should cut the tongues a bit longer so as to break the flat surface that the tongued boards and the grooved mullions form when it's a snug fit like this. Maybe this picture will explain. Any thoughts on this?

Either way, I'll finalise these tongued boards once the mullions, rails and stiles are glued up as they will simply slide in from the bottom.

Apart from the gap which I'm considering to leave between the tongued boards and mullions (which I'd love to hear your opinions on), I'm also a little unsure as to how these tongued boards should be glued in place. The obvious choice would be to spread epoxy into the grooves and then slide the boards in. I'm not sure this is the correct choice though as I suspect that one will have to somehow leave room for wood movement across the grain over time. I'm worried that, if I epoxy everything together, I might get failures over time as the wood absorbs moisture or dries out. Should I perhaps just slide them into the grooves and only epoxy them onto the bottom rail where they overlap with it?

Anyway, so all of the above happened about a month ago. Then the project ground to a halt as my attention was diverted to hospital visits and birthday parties. But I did get to pick up the strings again the last two evenings and will post some pics of the further progress tomorrow.

[TerryMcK]

When you fit the tongue and groove boards to the gate use a spacer of around 4-5mm between each board. When the boards expand (and they will) that space will soon be taken up. I made my tongues about 10mm (3/8") long and the grooves 12mm (1/2") deep.

Those dimensions also take into account shrinkage so the tongues will remain within the grooves. I used stainless steel screws to hold the boards to the gate (no glue).

 

Looking great so far Dewald.

[Dewald Swanepoel]

2 minutes ago, TerryMcK said:

When you fit the tongue and groove boards to the gate use a spacer of around 4-5mm between each board. When the boards expand (and they will) that space will soon be taken up. I made my tongues about 10mm (3/8") long and the grooves 12mm (1/2") deep.

Those dimensions also take into account shrinkage so the tongues will remain within the grooves. I used stainless steel screws to hold the boards to the gate (no glue).

 

Looking great so far Dewald.

Thanks Terry, that's exactly the kind of advice I was looking for. How does it work with these loose elements and the sealant/varnish you eventually put on it? If the boards move around, surely the varnish will crack between the boards and mullions? This will allow water to penetrate. Or is that just par for the course with outside projects?

[TerryMcK]

You can always prefinish the boards around the tongue/groove/slot area if you are concerned about water penetration. Then apply another couple of coats after assembly. I didn't bother as it is an outdoor project and I will apply varnish periodically (as I already have). 

The varnish I have used is a flexible one and doesn't crack. It is Epiphanes spar varnish but most spar varnishes/yacht varnishes will have flexibility.

[Dewald Swanepoel]

Alright so I then returned to the workshop after a hiatus of three weeks and thought I'd start with a dry fit again. Oh bother! This particular three weeks happened to be when the heavens opened after an extended drought in the country. The air was pregnant with lovely moisture... but so was the wood and I couldn't get all the mortise and tenons to slide in anymore. Notice how those bottom M&T's just won't go in all the way.

There's no indication that the weather will change soon and the advice I got was that African Mahogany, once dried again, would be unlikely to return to its shape. So I removed some more material from one of the tenons which sorted out the problem. All this is probably OK but my lovely mortise and tenon joints on the curved surface is now completely out of whack. It seems that the curved upper stile saw the most movement and now those shoulders just doesn't sit flush with the stile anymore. And seeing as the lap joints have already been cut, there's nothing I can do to fix this.

I'm really bummed about this and the best I can think of is to mix some AfMahog sawdust with epoxy and filling up these gaps at final glue up. The other option would be to widen the lap joints with the mid rail by a couple of millimeters but I just feel that would be even more noticeable and unsightly.

Oh well, at least the dry fit works again

Getting close to the first glue-up step now. Just some rounding off of the edges on the top rail, mid rail and upper sections of the mullions.

And breaking the sharp edges of the lower sections of the mullions with a block plane.

And we're ready for glue up. Stage one will be to glue the internal elements; That is, the 8 lap joints that connect the mullions (green boards), the mid rail (dark blue board) and the bottom rail (purple board). I did this by gluing the joints and then assembling the entire gate, even the unglued elements. This way the top rail and stiles, though dry fitted, help keep the internal elements exactly in the configuration they will eventually be in when all is glued up.

My weapon of choice, Epidermix 372. This is a binary epoxy with a nice thick consistency and a monster grip.

So the nervous task of gluing begins

And clamping that puppy

Small mistake. As I wiped off the squeeze out with acetone, I clearly didn't use enough of it and now there's a thin film of epoxy around the joints (as can be seen on the above picture). Looks like there is some intensive sanding in my near future.

And that's as far as I am. The next step will be to glue up the 6 mortise and tenons on the stiles and the 4 mortise and tenons on the upper rail. Which brings me to my next question that I will end this post off on; I've always planned on doing wedged through mortise and tenons for the six joints with the stiles. It seems there are two ways in which this can be done. The first type of wedged through M&T is the way @TerryMcK did it by cutting wedges off the corners of the tenons and driving them in at glue up. The second type involved cutting a kerf into the tenons and driving wedges into these kerfs. Here's a picture to illustrate the two types:

I'm half playing with the idea of using the second type now if for no other than aesthetic reasons. I like the idea of using a contrasting colour wood for the wedges. But I imagine that means I'll have to do some more work on my mortises first to give them an inner bevel so as to make room for the expanding tenon? I'm not particularly keen to return to the mortises with a chisel now though. Can Type 2 be pulled off without giving the mortise an inner bevel or should I rather just stick with Type 1?

While on the topic of wedges, would I derive much benefit from using fox tail tenons where the mullions join with the upper stile?

[Dewald Swanepoel]

Alright, so after that glue-up, this is what I had.

Next, I got cracking on these wedged through mortise and tenons. I decided to play it safe and go with the first type, the ones that @TerryMcK also used. I made a template from a piece of scrap MDF.

And then roughed up the tenons to give the epoxy more purchase.

I decided to cut the wedges from some Panga Panga I have lying around. I wanted to experiment with having a contrasting colour wood for the wedges (thought I think it would've looked better on the type 2 wedged mortise and tenons). I also figured that it might be good to have wedges from a harder wood than the tenons so I could really drive them in.

And then I clamped. I only have two clamps that are long enough to clamp the full width of the gate and only after I had applied them did I realise that they were pulling the gate concave. I really should have used 2 more (if I had them) and clamped the gate from the back as well but I decided to just use smaller clamps and pull the gate against the sash clamps to straighten it. This seemed to work.

After the epoxy cured I cleaned up the squeeze out with a plane iron and chisels.

Then flush cut the protruding tenons (I just love every stroke of this ryoba saw) and planed them down with a block plane.

The next step will be to cut the tongued boards that fit between the mullions and get them in place. It's actually been done already. I just need to get the pictures sorted out.

[drzaius]

That's a great looking gate. But I'm wondering why you'd make the wedged tenon that way. Putting the wedges on the outside of a sloped tenon provides no mechanical resistance to the wedge pulling out if the glue weakens years down the road.

The type 2 method will lock the tenon in the mortise much better.

[Chet]

Nice progress, its really starting to come together.

[Dewald Swanepoel]

16 hours ago, drzaius said:

That's a great looking gate. But I'm wondering why you'd make the wedged tenon that way. Putting the wedges on the outside of a sloped tenon provides no mechanical resistance to the wedge pulling out if the glue weakens years down the road.

The type 2 method will lock the tenon in the mortise much better.

Yeah, I was uncertain about that as well. i actually did ask about it in my previous post but when I got no answer and I had to move forward with the project, I thought I'd just go with what seemed like the easier wedge and the one which @TerryMcK had used in his gate.

I'm not sure how much mechanical strength it adds to the join but I'm sure it must add some. I used a harder wood than the tenon and so when I drove it in, it would have compressed both the tenon and the mortise wall. I guess time will tell .

I'd agree though that the Type 2 method certainly looks stronger. My physical intuition tells me that it would be the better join to use. I was hesitant to do it though as I really didn't want to go back into the mortise walls with a chisel as I was worried that I'd botch it. There were just too many unknowns about this type of join, like how far from the edge to cut the kerf line, how deep to cut into the tenon, what angle to give the mortise wall, how to make sure I get that angle accurately etc.

One day, when I'm big, I'll fearlessly make a type 2 wedged through mortise and tenon and nail it.

[TerryMcK]

17 hours ago, drzaius said:

That's a great looking gate. But I'm wondering why you'd make the wedged tenon that way. Putting the wedges on the outside of a sloped tenon provides no mechanical resistance to the wedge pulling out if the glue weakens years down the road.

The type 2 method will lock the tenon in the mortise much better.

 

1 hour ago, Dewald Swanepoel said:

Yeah, I was uncertain about that as well. i actually did ask about it in my previous post but when I got no answer and I had to move forward with the project, I thought I'd just go with what seemed like the easier wedge and the one which @TerryMcK had used in his gate.

I'm not sure how much mechanical strength it adds to the join but I'm sure it must add some. I used a harder wood than the tenon and so when I drove it in, it would have compressed both the tenon and the mortise wall. I guess time will tell .

I'd agree though that the Type 2 method certainly looks stronger. My physical intuition tells me that it would be the better join to use. I was hesitant to do it though as I really didn't want to go back into the mortise walls with a chisel as I was worried that I'd botch it. There were just too many unknowns about this type of join, like how far from the edge to cut the kerf line, how deep to cut into the tenon, what angle to give the mortise wall, how to make sure I get that angle accurately etc.

One day, when I'm big, I'll fearlessly make a type 2 wedged through mortise and tenon and nail it.

The thing about this joint in this particular application is there is a massive gluing surface area as it is a through tenon. The wedges do slightly compress when driven in and as they too are also heavily coated with glue it makes for a weatherproof joint. They also take up all the space cut by the slope cut into the tenon (presuming you have cut them at the same angle). There is an issue though that the clamps need to be on before driving the wedges in as the wedge can push the joint apart. When the epoxy is cured this is no longer a problem.

If the joint was used on a smaller item like a piece of furniture then sloping the mortise walls and keeping the tenons straight would add strength when the wedges are driven in - fox wedging by cutting a kerf in the end grain is an option too. But on furniture the thing to remember is that sloping the walls of a mortise is relatively easy as the sections are not too thick.

On doors, gates etc with comparatively thicker stiles there is a problem. Have you ever tried sloping mortise walls accurately at 5 degrees through a section 4.1/2" thick or greater? It's not easy to do. Hence the compromise that is made by sloping the tenons and keeping the walls of the mortise straight. The reliance is then upon modern epoxy resins to secure the joint. They make boats using this stuff and it really is stronger than the wood itself.

I am currently making two pairs of driveway gates bigger than the garden gates using exactly the same wedging technique. However I've used a belt and braces (suspenders) approach this time around. The joints are still wedged through mortise and tenons but are also drawbored. Marc Spagnuolo has a video on the WoodWhisperer website explaining how to do this if you are wondering what drawboring is - here is the link. 

What drawboring brings to the table is that you don't actually need any clamps when making the joints as the drawbored holes and dowel pull the joint together when the dowel is driven through the holes. It also addresses the potential issue (if remote) for mechanical joint failure. Then the wedges are driven in as before with plenty of glue and that joint is going nowhere.

 

[Dewald Swanepoel]

Aah, thanks Terry, that does give me some peace of mind. I must say though, when I'm feeling how tight these M&T's were before I even wedged them and knowing this epoxy, I just can't imagine how those joints could ever fail.

But yes, I guess time will tell. I'm happy to submit this entry to the test of time. It's my first gate so if it lasts only 10 years, I'll be more than happy

[Dewald Swanepoel]

Moving along; My next step was to tweak the tongued boards that fit between the mullions to sit nicely and have some room for movement. I've already illustrated in a previous post how the tongues were cut and it was a simple matter of adjusting the depth and width of the cut slightly so that the tongues were 5mm longer than the grooves on all sides and also have some play inside the grooves, ending up with this

And where the boards meet the middle rail

Next I had to fix these in place and, as discussed before, I didn't want to glue them into the grooves. The whole point of these loose tongues were that the panels can shrink/expand over time without causing problems. I didn't want to chase screws through the back of the bottom rail so I decided to glue them only in the centre. First I cut a 5cm long groove on the router table down the middle of each board

Then marked that groove onto the bottom rail (the play between the tongue and groove on the side can be seen nicely here)

And chiseled out a corresponding groove on the bottom rail

So this gave a nice pocket between each board and the bottom rail

I left a little dab of epoxy in the middle of each board along the groove in the middle rail...

...slid the boards into position and epoxied a key into each of the five pockets to fix the boards to the gate but still allowing them to expand and contract across the grain.

When the epoxy was cured, I simply cut the keys flush with the bottom of the gate

And no sooner had I completed this step than I realised my mistake. I should have sealed these boards before I glued them into place. Those loose tongues in the grooves are raw wood and will never be sealed properly now. I don't know what that might mean in the long run for a gate that will be out in the rain and sun. Bugger! I'm not sure how big a problem this is but there isn't really much I can do about it now.

[Janello]

Fill all the gaps with epoxy after you apply finish. You'll be alright, it's a gate, not a castle door. 

The overall project is looking great and I'm sure you'll walk away with some great knowledge from this. You have some interesting joinery in this build and overall I think you've executed it pretty darn well.

[TerryMcK]

Just seal the ends with epoxy and you'll be good to go.

Dewald just an observation with the last photo I hope you have enough clearance at the bottom of the slots and the peaks of the tongues. It appears to be minimal despite the clearance between boards at the root of the tongues looking adequate. Hopefully there are not too wide humidity changes in your neck of the woods.

Keep your eye on the mid and bottom rails too as the glued in mullions may crack off a lump of dado on the rail when they expand. I didn't notice you had done that until the last picture (long grain mullions at right angles to cross grain dados in the middle and lower rails). Fortunately if that happens it isn't a big deal to fix and with good fortune it might not happen.

[Dewald Swanepoel]

25 minutes ago, TerryMcK said:

Just seal the ends with epoxy and you'll be good to go.

Dewald just an observation with the last photo I hope you have enough clearance at the bottom of the slots and the peaks of the tongues. It appears to be minimal despite the clearance between boards at the root of the tongues looking adequate. Hopefully there are not too wide humidity changes in your neck of the woods.

Keep your eye on the mid and bottom rails too as the glued in mullions may crack off a lump of dado on the rail when they expand. I didn't notice you had done that until the last picture (long grain mullions at right angles to cross grain dados in the middle and lower rails). Fortunately if that happens it isn't a big deal to fix and with good fortune it might not happen.

Oops, I may have a problem then. That last picture is pretty accurate on how much clearance I have inside the grooves. Our humidity in Johannesburg does swing a bit from around 50% in winter to 70% in summer. I'll have to keep an eye on those boards then.

I didn't even think of the long grain mullions at right angles to cross grain dados being a problem. I'm not sure how the design should have been different to avoid that though. Can you point me to an article or some pictures perhaps that explain this issue? But yes, so we learn. I suspect I've made some other similar rookie mistakes in this gate. In fact, I know of at least one. Just spoke to another woodworker here who told me that I should have sanded the individual pieces down to a near final grit before assembling the whole thing. I guess I will have to buy an extra jar of elbow grease then...

[Chet]

These has been enjoyable to watch Dewald.  I just hope your garden is worthy of this gate.