Exterior Door Advice please

[Mannyroad]

Hi,

I'm new to the forum, yesterday actually, and have been pouring over the forums which are a treasure trove of advice and ideas! I'm recently 'retired' early so have plenty of time now to pursue my hobby interests, one of which is DIY woodworking.

I'm planning to make a new front door for my cottage and my material choice is oak or idigbo. I was originally going to buy one of those engineered wood versions till I realised that they come in standard sizes and they are no good if they are exposed to the rain!! :lol: So I'm going to have to make one.

The style of door I want is a cottage style with a small vision panel. I would like the door to be double boarded which means there won't be room for a mid-rail. What concerns me is the likelihood for the door to twist/move etc. To try to combat this I'm proposing to make the door frame as follows:
Stiles: 144x44, top rail: 125x44, bot. rail: 219x44. All joints through tenon. Panel to be 12mm ply sandwiched between 16mm oak/idigbo facing boards. I'm not sure of the best way of fixing the facing boards. I was figuring on gluing them to the ply with Cascamite, given that they are only 16mm thick. Or is it better to use 18mm ply with 13mm glued boards?

Below are 'typical sections' through the door, showing my proposed construction. The ply core would be edge sealed with slow cure epoxy to help prevent end grain moisture absorption and I'm proposing to overlap the facing boards at rail/stile junctions to reduce water ingress. Where the facing panel rebates meet the head of the vision panel I was proposing to 'plug' these with Sikaflex to prevent water holding behind the top of the vision panel frame.

What I'd like to ask, of those guys who know their stuff regarding door construction, is whether the proposed construction, whilst maybe a little unorthodox, should do the job. The materials aren't cheap and so I don't want to go to all the trouble of making the door only to find it's going to fail. If not can anyone suggest a better way to build the door?

I'm not a joiner so my apologies if my terminology isn't spot on. I've made a few woodwork projects before, including a pair of pitch pine louvre doors for a woodshed I built last year along with a pair of sapele gates (see photo below), so I'm hoping my experiences with these doors, gates and other projects will carry me through.

Thanks for taking the time to read this guys. I look forward to your thoughts.

 

[dzj]

It isn't the usual frame and panel door, that's for sure.
It probably would be OK.
I'd make the bottom rail in the manner of the gates in the pic.
(Water might get through the rabetes of the face panels.)

 

[ColeyS1]

I wouldn't go gluing the boards to the ply. If the boards expand or shrink its likely to split them if its glued. when I've made this type before I've ran a groove around the top rail and stiles and the boards just slot in. You can secret nail the boards through the tongues to the rails.This also does away with the need to groove the ply panel in. If the boards are grooved in both sides the ply panel in theory can just be left loose. If water somehow got behind the boards it would sit in the groove on the bottom rail and rot away ! When I was an apprentice I grooved a bottom rail by mistake, 6-7 years later it was back in the workshop being remade cause it started to rot It was softwood which didn't help I would definitely stick in a middle rail , probably same size as the bottom rail. It'll be the same thickness as the ply so should help keep the stiles nice and straight.

 

[billybuntus]

May I suggest that you don't use ply. If I were you I'd be buying marine ply to do the job which will be a small fortune. Spend a few quid more and get some Medite Tricoya.

I'm currently making up a paneled exterior door surround and have finished to large molded bay window tops and used meditie tricoya throughout.

It's bomb proof. I left a piece outside for two months. It was saturated with water (no coatings) but the material didn't move at all. When compared to a dry piece it was the same size. It's also guaranteed for 50 years against rot. With a service life of around 60 years.

If water gets at your ply it will rot. The multilayers of it are perfect for the rotting process to began and once water gets into it, it will be a goner in no time at all. As I have experienced twice now despite a lot of time and effort in prep and sealing all surfaces.

 

[RobinBHM]

I now use tricoya for fielded panels, flat panels or mock T&G in doors. I still prepaint panels before fitting in door, but have much more confidence about longevity.

The only frustrating thing is that there is only a subtle difference in appearance between tricoya and plain mdf, so any unmarked part boards or offcuts are hard to tell apart. I have put test pieces in a jug of water, 9mm tricoya expands by about 0.2mm after 4 hours, I forget the figure for plain mdf -it was a lot more!

 

[Mannyroad]

Many thanks indeed for your replies guys. I haven't come a cross Tricoya, I'll definitely check it out. The matchboards/battens, being 16mm thick so any midrail would only be 12mm max. thickness. Is there any point using a mid-rail this thick? The reason I intended using the large frame sizes was to ensure good heavy joints at the four corners to help reduce movement and the stiles are 144 deep simply because I was proposing to avoid a mid-rail. Any more thoughts on the need for a mid-rail if I stick to the sizes I'm proposing?

One thought I had was to up the main panel core to 18mm with 13mm matchboards, or even 25mm with 9mm mtachboarding. I was wondering if by using the thinner matchboarding would it be ok to glue them to the Tricoya, like thick veneer boards? I've seen this done by some door makers on bespoke contemporary doors, albeit those matchboards were only 6mm thick.

Thanks again for all your help. I want to make sure I only do this job once :wink:

By the way, is this Tricoya stuff more expensive than the oak or idigbo? If so I may as well just use full thickness (i.e. 40mm-to allow 2mm setback) matchboards.

 

[ColeyS1]

If twer me I'd use 18mm plywood and have the door finish 57mm. Even though your middle rail will be only 12mm it'll still hold the stiles and help stop them bowing.
I tried adding a pic via tapatalk but it's not working .... :|
I can't help feeling plywood would be just fine for this job- give it a lick of something first.

 

[Mannyroad]

Hi ColeyS1, cheers for your ideas, and for taking the time to draw it out.

The door will need to stay at 44 thick because the frame rebate is this size. I'd rather avoid replacing the frame because its well rendered into the reveals and being set into an old thick stone walls, with stone lintols and jambs, it has to come out internally rather than externally and this would mean removing much of the reveal render/plaster.

To be honest, I thought marine grade ply with epoxy sealed edges would have been ok but I quite like the idea of this Medite Tricoya, having looked at how durable and stable it is.

 

[billybuntus]

I now use tricoya for fielded panels, flat panels or mock T&G in doors. I still prepaint panels before fitting in door, but have much more confidence about longevity.

The only frustrating thing is that there is only a subtle difference in appearance between tricoya and plain mdf, so any unmarked part boards or offcuts are hard to tell apart. I have put test pieces in a jug of water, 9mm tricoya expands by about 0.2mm after 4 hours, I forget the figure for plain mdf -it was a lot more!

Any decent offcuts always get "MT" across them in pencil. Failing that it smells of vinegar.

 

[ColeyS1]

The price of it may influence your decision. From memory I'm think 75 quid for an 8x4 12mm sheet.
I must comment on your other projects. The gates look really good ! You obviously take alot of pride in your work

 

[Fat ferret]

I think you need to be careful mixing ply and real wood. Before I became a joiner we had a door made and the man had used ply behing tounge and groove rebated into the frame from outside. When the wood moved the shoulders of the m and t joints opened up because the t and g had expanded. Binned that and made another without the ply. Still fine.

Doors do pose special problems but my advice is get good straight timber don't make your design too complicated rebate t and g into the outside and paint asap to try and keep movement down. Make sure your gap is an even 2-3mm all round and sort out your threshhold. I just make triangular drip strip for the bottom although you can buy fancier patterns. I think you may aswell use through m and t joints for doors and always use a diagonal brace, the right way round! Finally if your door does twist a wee bit after a few months no one (except you) will notice if you move the stops to compensate. Good luck.

 

[ColeyS1]

That just says allow a decent gap between the boards to me. You can sneak your last few nails in the sneaky gaps between each board

 

[Mannyroad]

Many thanks for the advice guys, it's much appreciated, and ColeyS1 thanks for the kind words regarding my gates, yes I do take pride in my work.

From what I'm hearing I need to dump the ply, use diagonal bracing in bottom half and a mid-rail (vision panel prevents diag. brace in top half), though the brace and midrail would only be 12mm thick. Is this a fair assessment and will the bracing and mid-rail be effective at this thickness (bear in mind I'd have very little shoulder on the mid-rail)?

 

[ColeyS1]

Personally I'd use ply imagine your stiles are only 2inchs wide, if the ply panel is cut in nice and snug there's no way at all it'll be able to drop- plus its easier than cutting in braces and..... and...... and it's better for insulation :lol:

I dont think I'd bother having shoulders on the middle rail at all. If you keep it 12mm (same as mortice) your haunch will keep it the right distance.

 

[Jacob]

Doomed to fail - over "designed"
It's a trad looking door so why not go for a traditional construction? I'd take the boards down to the bottom and not stop at the bottom rail, make the middle and bottom rails thinner than the stiles and top rail, by the thickness of the boards. Make the M&Ts line up flush with the back of the boards. The boards t&g to each other and a slot in the stiles and top rail.

PS a more common term for "vision panel" :roll: would be "light".

 

[Mannyroad]

Jacob,

Thanks for your comments and I appreciate what you're saying, which seems to be to simply make a normal framed, ledged and braced door, with the ledge and braces visible on the interior side, but that's not what I'm wanting to achieve. I'm wishing to double board the door so that I have matchboarding/battens both sides. The effect I want is as the photo below, but on both sides, and for a 45mm thick door, including the vision light, which mucks up bracing the top half of the door.

I'm not sure I understand why different means 'doomed to fail'. If properly considered and constructed failure should be avoidable.

I guess what my thread was hoping to achieve was to establish why, if at all, the proposed construction would not be a sound choice of construction. It would certainly be buildable, but what will the problems be, if any, and HOW do I solve these.To my understanding, the main issues are water ingress, matchboard movement and fixing and if a solid panel core is considered a good choice, what should this be, marine ply/Tricoya etc, and it was really these matters that I was looking for member advice.

 

[Jacob]

Well what you have drawn looks like a normal door so why not make it in the normal way? Not braced. Just framed. That's the trad pattern for door like yours (but with the boards reaching the bottom, not stopping at the rail.
If you want it double boarded just add boards to the inside, perhaps held in with a bead.
Your weakest detail is the idea of gluing to a ply core. Bound to fail with differential movement (moisture, heat) - they need to be loose to allow for movement. Anyway you don't need a ply core it doesn't do anything except add weight.

PS I've just remembered - a few years ago a friend of mine made a gate with a similar construction to yours. She assumed that if everything was tight, glued, well sealed etc it'd withstand the elements. After the first bit of rain it started expanding and gradually opened up huge gaps in the joints between stiles and rails.
She was trying to copy the original so I told her to look closely at the original and copy it in every detail. But she had already scrapped it. Simple lesson there!

 

[Mannyroad]

Thanks for your comments Jacob. Presumably then you'd secret nail the matchboarding to the bottom rail? I ask this because without any bracing, which the boards would ordinarily nail to, the only thing holding the boards would be the bottom rail. I was only using the ply to fix the boards to, as one would to any ledges and braces and to stop the shutting stile dropping. I guess I could stop it dropping by using one lower diagonal brace gunstock jointed into the stile. That should work, and I could nail the boards to this brace as well. You reckon this is a good alternative then?

 

[ColeyS1]

If you were to make the top rail how I drew it you can easily nail the boards to that, and the middle rail, and the bottom rail everyone has different ways of doing things, pick which one you think works best for you.

 

[blackrodd]

Doomed to fail - over "designed"
It's a trad looking door so why not go for a traditional construction? I'd take the boards down to the bottom and not stop at the bottom rail, make the middle and bottom rails thinner than the stiles and top rail, by the thickness of the boards. Make the M&Ts line up flush with the back of the boards. The boards t&g to each other and a slot in the stiles and top rail.

PS a more common term for "vision panel" :roll: would be "light".

+1 I must agree, Always best to allow timber to expand and contract with the seasons, forget the ply.
I'd allow 4mm between shoulders, on the T&G or rebate the meeting edges should you think the T&G tongues not strong enough. Rodders