Building some bunk beds that will be pretty permanent and my plan was just to glue everything to make it nice and sturdy. If I'm gluing is there any reason to use screws at all, trying to make it look nice?

  • 3
    A lifetime master of Japanese joinery could probably make a strong bed without even the glue, based on years of training and experience in how wood joints behave under stress and how to work with the natural materials' strengths. If you use good joints and braces where needed, glue alone might well suffice. But for the rest of us mere mortals who are not PhD structural engineers, screws are cheap insurance. If it were my kids, I'd have dovetails, glue, screws, corner braces, and steel reinforcements. Jul 10, 2018 at 22:38
  • ...and a crash cage for the kid on the bottom bunk! Jul 10, 2018 at 22:49
  • @LeeDanielCrocker you should post that as an answer, I'd upvote!
    – Kromster
    Jul 11, 2018 at 10:25
  • 1
    One point about screws worth noting is that they act like internal clamping, so screws can be used to pull certain joints tight which is important for strong glue joints to form. And if left in place they act as security should the glue joint ever fail for any reason, e.g. shocks, regular movement stresses and expansion/contraction due to moisture changes year by year. Using screws does not at all have to mean you compromise on looks, the screws can be sited on surfaces that don't show or if used on show surfaces hidden by wood plugs (search for "counterboring" for more) making them invisible.
    – Graphus
    Jul 11, 2018 at 11:22

1 Answer 1


Lee's comment is right on Ryan. But in any case, we couldn't answer your question without more detail about the construction. Depending on what you're planning, even screws+glue might not be strong enough.

In particular, glue is only strong in shear, and not tension. It's also not effective on end-grain. With either screws or glue (or both), you'll still want the mechanical strength of the wood to be working for you, and that will mean non-trivial joints like mortise and tenon. Just butting pieces up against each other and joining them with glue and screws risks ending in tears.

A bed especially is subject to jumping and climbing stresses and has unconscious occupants much of the time.

If you can provide drawings we could give you something more specific. If it were me, I'd be thinking hard about a quick trip to IKEA :)

  • That point about glue is a bit of a red herring as there wouldn't normally be situations where there you'd create a joint that's only in tension. And anyway in woodworking it's not about the glue itself usually, it's about the glue joint, which isn't exactly the same thing (thick glue line being weak, thin being strong, signifying that it's not about the adhesive per se).
    – Graphus
    Jul 11, 2018 at 11:33
  • @Graphus I guess I was thinking of maybe gluing a plywood "bottom" (that a mattress would lay on) to a simple perimeter frame. But it's an interesting point. And I think you're right that bets are somewhat off when talking about wood (rather than say metal) where all the adhesive has to do is bond more strongly than the lignin holding the fibers together :)
    – scanny
    Jul 11, 2018 at 17:14
  • What exactly is your source on glue joints not being strong in tension? Modern wood glues used properly are stronger than the surrounding wood in every test I've ever seen. Jul 11, 2018 at 18:23
  • Dan Gelbart: youtu.be/EeEhS3zmnDg?t=832. But in fairness, his discussion is in the context of engineering materials, not wood. Also, he makes a distinction between peel mode failure (like lifting Formica from corner of countertop) and "pure tension" where the entire joint is stressed equally. As I mentioned in my reply to Graphus, these characteristics are perhaps less relevant to wood given its lower material strength, but it would apply to joints themselves, like lap joint can peel where mortise/tenon can only shear. Joint fails whether it was the glue or lignin that gave way.
    – scanny
    Jul 11, 2018 at 21:42
  • 1
    Extrapolating from the advice generated by past experience does tell us that even using a very strong adhesive as they were back in the day, and cutting joints generally to a very high standard, that it was recognised how much weaker a halving joint (lap joint) was in the real world. This wouldn't have been any real issue for something that just sat there, but the flexing that goes on when furniture was moved stresses all joints in a structure and these were known to fail. The higher seasonal movement in the wood due to the finishes used historically would have contributed to the problem.
    – Graphus
    Jul 12, 2018 at 11:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.