I’m building a bookshelf out of MDF (18mm thick) using 6mm dowels (Silverline brand like the dowelling jig). Each shelf will be dowelled in place at the back and on both sides.

I’m concerned about whether the dowels will be strong enough for a heavily loaded shelf.

Is there a formula or rule of thumb that says how many 6mm dowels are required (or even what distance between dowels is required) to support a given shelf load (in kg) assuming the shelf is supported only by dowels on 3 sides?

I’m planning for the dowels to be glued. Are there times when dowels are not enough and the shelf should also be glued, screwed or nailed in place?

  • 2
    There's no formula that I know of, and if there were it would need to be taken with a pinch of salt because dowel strength can vary enormously due to variation from piece to piece (see this recent thread for a little on why). Additionally the length of the dowels and whether they go through the case or are in shallow holes are important factors, as is the quality of the MDF (it's not all equal, and the worst of it is pretty crumbly).
    – Graphus
    Commented Jun 26, 2019 at 13:12
  • I think because of these many reasons, a lot of people switch to biscuits for MDF work like this.
    – user5572
    Commented Jun 26, 2019 at 13:15
  • 2
    One last pithy comment is that whenever I think about MDF work I always consider flat-pack furniture construction (the good and the bad) we've all seen too much of. How do Ikea, et al, engineer this sort of thing? What construction do they use for their items intended for more massive static forces? This will give you an idea of how they tune their construction for the material at hand and the loads involved.
    – user5572
    Commented Jun 26, 2019 at 13:23
  • 4
    @jdv "I think because of these many reasons, a lot of people switch to biscuits for MDF work like this." tee hee. I'm laughing because testing shows that dowel joinery is, and this is a surprise to most, actually hard to beat in terms of strength. I believe what happens when pros are weighing whether to go with a biscuit jointer or dowels that speed becomes the sole deciding factor in the majority of cases, despite the much higher initial spend and the higher ongoing costs with biscuits. Which is perfectly legit of course, clearly both are strong enough for most purposes, and time is money.
    – Graphus
    Commented Jun 26, 2019 at 16:57
  • 1
    Metal would certainly be stronger.... but it's possibly overkill for a shelving unit where the shelf material itself may be likely to turn out to be the weak point ^_^ That aside, this doesn't Answer what was asked. It's fine to suggest alternatives, but it's considered good form to answer the Question as asked, then go on to give the alternate suggestion(s) if you have some and ideally with explanations for why they're better/more desirable, (e.g. stronger, less work, cheaper) or necessary (only safe to do it by a different method than asked).
    – Graphus
    Commented Jun 30, 2019 at 5:05

1 Answer 1


You can model each dowel and the shelf itself as beams. The dowels would be like a uniformly distributed load cantilever, while the shelf would be a simply supported beam. You can look up the material properties for your material type and input the parameters of your shelf in the calculators below.


Reasonable values would probably be yield strength of 50 MPa and an elastic modulus of 10 GPa.

Typically deflection or "sag" is the first area of concern for beams so I recommend solving for that first. I have attached a few calculators you can use for that.




You can also find the maximum moment and use the flexure (bending) stress equation in the link below to solve for stress and see if it is above the yield stress for your material type. https://mechanicalc.com/reference/beam-analysis

  • This is great, but hard to apply here. The strength of the dowels is an unknown, and as mentioned in the first Comment above can vary greatly even for dowels of a known type (breaking strain in some could be as low as half that of the strongest).
    – Graphus
    Commented Oct 18, 2019 at 6:59
  • Not to mention that those web refs will be moved or removed in an internet moment. To make this answer have a longer lifetime, maybe edit it and focus on the specific terms and notions that someone would use as jumping off points for study. That is, call out terms like "maximum moment" etc. and focus on those in your answer, with the links presented as short-lived examples.
    – user5572
    Commented Oct 18, 2019 at 14:54
  • @Graphus The link I provided has material properties for a wide variety of wood types. The actual strength doesn't need to be accurate, just double or triple the number of rods predicted to support the load. Breaking strain is a measure of displacement and isn't really relevant for this question. Stress is probably the word you are looking for. Commented Oct 19, 2019 at 0:49
  • @jdv Hopefully things are a little clearer now. The links are useful because you don't really need to understand the mechanics, you can just plug in values. Commented Oct 19, 2019 at 0:51
  • I will defer to your engineering knowledge 100%, but I know wood in a hands-on way backed up by published info that has looked at the material since the late 19th. c. The strength of the dowels is an unknown. If strength (however one needs to term it) can vary by a factor of 2 or more values given in some table are of limited practical use (esp. if not confirmed elsewhere). So say the solution on paper is to double or triple the number of dowels, that seems fine until you get to actually trying to do this, e.g. perfect alignment of 36 holes for one shelf! So hard to apply here is fair.
    – Graphus
    Commented Oct 19, 2019 at 9:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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