Drywall screws vs. other types of wood screws

From time to time, people criticize me for using drywall screws in so many of my projects. But I really like drywall screws, and not just because they are cheap. They have a fairly flat head, which I think is less prone to getting pulled through the wood than the wedge-shaped heads of other screws. And they are fairly thin, so they are less likely to split the wood.

After my shed building articles, people once again criticized me for using drywall screws. So I figured it was time to put them to the test.

I tested four different types of screw, each 2.5" (65 mm) long. The drywall screw is a #7 screw, while the other screws I tested are #8 screws.

#7 drywall screw (black)
#8 wood screw (silver)
#8 deck screw (tan)
#8 self-countersinking deck screw (green)

The drywall screw is slightly thinner than the other screws, though its head is just marginally larger.

Based on my experience with screw joints, the limiting factor with wood screws is often that the head sinks into the wood. So with this test, I wanted to see how much force each screw head could exert on the wood.

I rigged up a test piece, which spreads the load between a fixed fulcrum and a bathroom scale. The point of contact with the scale is three times further from the fulcrum as the screws, so the scale registers one third of the force applied by the screw.

I pre-drilled 11/64" (4.4 mm) holes for the screws. That way, the screw's thread won't engage the top piece of wood. Otherwise, the screw's thread would pull down on the screw, pulling the head into the wood, even without pulling up on the piece below.

To get multiple tests out of the same piece of wood, I put five holes side by side. After testing a screw in each of those, I flipped the work piece around and made another five holes. After that, I cut an extra set of notches, offset from the first set by 5 cm so that I could then get another line of holes into the wood, also with a 1:3 leverage to the scale.

I started each screw with an impact driver, but switched to a manual screwdriver to do the final tightening. I then recorded the largest value that I could get to stay on the scale for ten seconds after the last tightening.

And the results are....

Screw type	Scale readings	Average reading
Drywall screw (black)	75, 90, 82, 82	82
#8 wood screw (sliver)	70, 80, 72, 72	74
Deck screw (beige)	75, 70, 85, 73	76
Deck screw (green)	60, 55, 92	69

I was expecting the drywall screw to do better than the regular wood screw in this test, but looking at the numbers, it has a relatively small advantage. Considering that the drywall screw's head has a 7.3% larger area than the #8 wood screw, and the average force exerted by the drywall screw was just 10.8% more, it's a very marginal advantage.

Screw	Head diameter (mm)
Drywall screw (black)	8.33
#8 wood screw (sliver)	8.04
Deck screw (beige)	8.20
Deck screw (green)	8.10

The self-countersinking cutters on the back of the green deck screw's head help to dig a countersink into the wood as it's driven in. But that same countersinking action means there's less force to exert on the wood as it's cut away. This makes it easier to drive the screw's head below the surface, but it will also exert less force as it's tightened.

I also noticed that the readings towards one edge of the test piece seemed to be larger than on the other side. And averaging the readings towards either edge, I calculated 80 pounds at the top edge (in the picture) and 69 pound towards the bottom. I tried to mix where I placed each screw to average that bias out of my results. But it's interesting that which side of the wood I tested had as much of an effect as which type of screw I used.

Looking at the piece of wood from the end, you can see that one side has slightly darker growth rings closer together. The readings on that side of the wood were higher. The dark summer growth is much harder.

This is the block I was screwing into. It's a relatively low density piece of spruce, and the tips of the screws only screwed about 1" (25 mm) into it. But the screw thread held well enough in it, except for a few times where I accidentally screwed into a previous hole.

There is also the question of how easily the screws might break. So I put each screw in a vise, and bent it over sideways.

I tried several drywall screws, and they all broke before I could bend them 90 degrees. I tried some #8 wood screws, and they also broke, although one of them still barely held together at 90 degrees.

The deck screws were much more bendable, and neither of them was close to breaking at 90 degrees.

So I would say the drywall screws are slightly more brittle than #8 screws, and the decks screws were the least brittle. Although I later repeated that test with another type of deck screw, and that screw also broke.

What's nice about drywall screws is that they are relatively thin screws, and the head doesn't sink into the wood as easily. But when working with hardwoods, the strength of the screw itself becomes the limiting factor. At left, I screwed two 1.5" screws into a piece of ash hardwood as far as they would go, without any pilot holes. Once the head of the drywall screw started sinking into the wood, it snapped apart near the head.

The #8 wood screw, however, sunk a little below the surface, but then the thread started to slip in the wood. The screw also caused a crack in the piece of wood, which the thinner drywall screw did not.

I can't say one type of screw is "better" than the other, but when working with soft woods or when splitting is a concern, drywall screws certainly have their advantages. But in hardwoods, a drywall screw will usually fail before the wood fails.

Further thoughts
This article has created quite the reaction from people. It seems, ome people are very passionate about what sort of screw to use, and insist that drywall screws are not "proper" screws and should only be used for drywall.

That said, I'm no fan of construction screws used with 2x4 construction. Though they are less likely to break than drywall screws, they DO break, and often, there will be a screw head or segment of screw left in a scrap piece of wood, with no metal protruding on either side. That sort of thing never happened with nails. This makes it riskier than it used to be to use recycled lumber, especially because the harder screws will cause more damage to blades than nails would.