Spent the weekend on the deck and finally stopped stripping my screws

Sep 25, 2026 Last reply: 3 hours ago 6 Replies

I spent the afternoon swapping out hardware and finally figured out why I was rounding off so many heads lately. I used to just grab a #2 Phillips becasue it's the standard, but I realized that if there's even a tiny bit of play when you seat the bit, you're going to have a bad time. Now I make sure to test the fit by hand first. If I can wiggle the driver at all inside the screw head, I swap it out. It sounds obvious but taking those few seconds saved me so much frustration today. The real game changer was being disciplined with the metric vs SAE mix-up on the lag bolts. I had a 14mm socket that felt "fine" on a bolt, but when I grabbed the 9/16th, the difference in the snugness was huge. It's that tiny fraction of a millimeter that ruins your day once you start putting real torque on it. Same goes for the hex bits on the trim. If it's not a perfect, zero-play fit, I don't pull the trigger on the impact driver anymore. It's way better to walk back to the toolbox than to spend an hour drilling out a snapped screw.


I learned that lesson the hard way while rebuilding my back stairs last summer. I used to assume all star-shaped bits were basicly the same until I tried using a T20 on what turned out to be a T25 GRK screw. It felt like it caught, but as soon as the impact driver kicked in, the head just turned into a smooth crater. Now, I always keep a small magnet on my driver; if the bit doesn't stay perfectly seated by itself without me holding it, I know it's the wrong size.

Another thing that helped was switching to Wera Impaktor bits. They have a tiny bit of texture on the tip that bites into the metal. It also helped me realize how many Phillips screws are actually Pozidriv, especally on some of the imported hardware I got for the railing brackets. If you see those tiny tick marks coming off the center of the cross, a standard #2 Phillips will slip every single time.

For the lag bolts, I stopped guessing and bought a cheap digital caliper. If I'm unsure between 13mm and 1/2 inch, I just measure the flat sides of the bolt head. It takes ten seconds but prevents me from rounding off a head that's buried four inches deep in a 4x4 post.

I actually stopped relying on the 'wiggle test' becasue some of the cheaper deck screws I get from the local hardware store have such inconsistent coatings that a bit might feel snug but it's actually just sitting on a layer of paint or zinc. Instead of just checking the fit, I started obsessing over the pressure I apply. I realized most of my rounding issues came from me being lazy with my body weight. Even with a perfect Phillips #3 or a 5mm hex, if I'm not leaning my full shoulder into the back of that drill, the bit is going to cam out the second I hit a knot in the pressure-treated lumber. I also switched almost exclusively to impact-rated bits from Milwaukee or Wera. They have a bit more 'give' in the torsion zone which seems to prevent that sudden snap-and-spin that ruins the head. It's less about the perfect millimetric fit for me and more about making sure the bit stays bottomed out in the recess no matter how much the wood fights back.

While mechanical fitment is paramount, one must also account for the cam-out phenomenon inherent to the Phillips geometry, wich was intentionally designed to slip to prevent over-torquing in industrial applications, a characteristic that becomes a liability when driving 3-inch structural screws into pressure-treated lumber. I have found that applying a small dab of valve grinding compound to the tip of a high-quality JIS or ACR bit provides the necessery friction to bridge the tolerance gap in those instances where the fastener manufacturing is particularly substandard. Furthermore, when dealing with hex-drive lag bolts, I prioritize six-point flank-drive sockets over standard twelve-point variants, as the former exerts pressure on the flats of the fastener rather than the corners, significantly reducing the probability of rounding under high-torque loads.

One thing that finally stopped the stripping for me was cleaning out the screw heads first. On my last deck project, I realized half my 'bad fits' were just gunk, sawdust, or dried finish packed into the bottom of the drive. I keep an old dental pick or a stiff wire brush in my pocket now. If the bit doesn't bottom out completly, I don't even try to drive it. Scraping out that bit of debris makes a loose-feeling bit suddenly lock in solid. It’s a pain to do it for every screw, but it's faster than grabbing the vice grips later.

Glad to see folks finally paying attention to the fit. Used to be we'd just switch everything over to Robertson square drives and call it a day—never understood why Phillips stayed the standard for so long. One thing I'll add: check your bit for wear. If the edges look even slightly shiny or rolled under a light, it’s scrap metal. Back when we used Yankee drivers, you could feel a bit failing, but these modern impacts are so fast they'll melt a head before you can blink. Don't be cheap; toss the bit the second it looks polished.

Quick update—cleaning out the gunk with a wire brush like you suggested relly helped seat the bits deeper. However, I'm still hitting issues with those thick zinc coatings making the #2 bit feel like it won’t even enter. Since the Phillips design is made to cam-out anyway, are you guys finding that certain brands of bits acutally grip those coated heads better than others?

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required