Hacker News Reader

Comment by blhack | Hacker News Reader

As far as I'm concerned, torx is the only reasonable screw anymore. It seems like most construction screws (at least in the US) have moved to this.

Just to make the point: I've stripped out more torx drivers than I have torx heads.

I can say I truly don't understand how phillips ever even took hold over flatheads. Cheap phillips head screws seem designed to strip out.

Somebody elsewhere in this thread made the point that they are self-centering, and make industrial manufacturing easier. I guess that makes sense.

View on HN

As someone who has just remodelled an old house with loads of flatheads: you've got to be kidding me. I hate those things with a passion.

Flathead screws only restrict the movement of the screwdriver in one dimension and allow it to move in the other one. This leads to the screwdriver to keep sliding out when you put force on under even the slightest angle.

Phillips screws improve on this design by restricting movement in two dimensions and force the screwdriver to always be in the center, but they slip easily and wear quickly.

Torx are superior because they fit much more "snugly", though at the (minor) cost of being less forgiving of using the wrong size of screwdriver.

Yeah that's fair. I guess I was thinking of flatheads as small machine screws and things like that, where only a few rotations are actually necessary.

An advantage of flathead is that it is easy to unfoul the screw head. A minute with a utility knife or a flathead screwdriver will clear the channel of dirt or paint. I would not want to deal with a torx that has several layers of paint dried into it.

I have fortunately not had to experience that. I guess a special screw extractor tool might be the only option then.

Break out the drill and drill a tiny hole down the centre to crack the screw. Now pull it out and can put in a new clean one.

Using a reverse drill (if you have one) might also be helpful

The drill will bite into the screw and start unscrewing it

I think a better solution to this problem is a different tool: a bulldozer.

Use a needle or a knife.

As someone who has just remodelled an old house with loads of flatheads: you've got to be kidding me. I hate those things with a passion.

Came here to say the exact same thing. Rusty flatheads, ah, such joy. Literally lost hours because of it.

I’m sure you have worked this out but my technique is to scrape all the paint off very carefully, then crack the lock by turning the screw in the wrong direction. Sometimes I hit the screw driver with a hammer too. After the lock is broken, turn the screw the right way. This way, stripping the head isn’t such a problem as it’s burring off the head in the wrong directions, saving the important edge for actually getting it out.

Yes, I also quickly developed some skill with a utility knife, getting the paint out and creating space for the screwdriver head, but I never found a good way to prevent the screwdriver from sliding sideways. Especially with large screws that require a lot of force to remove.

Heat and then a crack with a hammer have been my method. Always using the absolute tightest getting screw driver. I still snap the screw maybe 1 time in 5 due to the torque I can get on it. I’m repairing rusted out 1940s window hinges, and it’s truest awful. Have got about 100-150 to go, but I’ve done maybe 300.

I use a cheap chisel the same width as flathead screw, thoroughly clean out the slot with it and use as a screwdriver with very heavy pressure as I turn. This was for very old (circa 1900) screws in old growth wood.

If I was in charge of these things, I would enact a 'Fahrenheit 451'-style policy, except for Flathead screws instead of books. Any of the other myriad style of screws is preferable to those ancient fasteners.

Agreed. There are two things that I feel should be destroyed and replaced on sight for the betterment of humanity: flat head screws/drivers, and network hubs.

It would probably even cause a measurable increase in lifespan due to removal of a significant stress source. But I can dream...

I see your point, but if the remaining problem to solve in our civilisation is what type of screw heads then we're doing pretty good :)

Do you mean in the strict usage of a Ethernet repeater hub, as opposed to a switch? Where do you still encounter those dinosaurs?

Everywhere in large/old enterprises. I swear they multiply when nobody's looking, and they tend to cause massive headaches when they're plugged in by the unaware.

Hence: Kill on sight.

You probably have a cheap flathead screwdriver which has tapered edges. They are significantly more expensive, but a good flathead with edges that are square to the screw slot works much better. (I'm not sure if I described that well)

To make them work even better you can burnish a slight edge one the blade which will dig in and stop slippage.

They don't even need to be expensive. I have no clue why these aren't more common, but keep a cheap set of gunsmithing screwdriver bits around. Never seen them in a hardware store, but they are far superior to regular flathead bits with a chisel shape. Or just regrind all your screwdrivers when you buy them, which also makes it really easy to identify which flathead screwdrivers are reserved for hammering/prybar duty.

Another issue with torx is that the driver has to go in mostly square. Flathead or phillips screws can be turned if there's an obstacle directly behind the screw and you need to put the screwdriver in at a slight angle.

>I can say I truly don't understand how phillips ever even took hold over flatheads. Cheap phillips head screws seem designed to strip out.

It's not cheap ones; Phillips heads were designed to strip out. It's the very reason they're used.

The whole idea was to use them in factory assembly, so that workers wouldn't overtorque them. You put them in and tighten, and when they cam out, you're done.

If you want a fastener that can be easily removed, and maybe reused several times, Phillips is the wrong choice.

And if you have a torque-limiting screwdriver, as should be common in any modern factory, then Phillips is simply obsolete and unnecessary.

Well, they’re designed to cam out. Not strip out.

It only takes a small number of cam-outs before it's stripping out.

Only if the fastener has mechanical properties on the order of butter.

Ah yes I prefer my screws to be made of a hard cheese myself, something along the lines of parmesan.

... a nut butter?

<I'll see myself out.>

This was my understanding as well, but the wikipedia article directly addresses it and say that it is a common myth with no supporting evidence.

> Extensive evidence is lacking for this specific narrative, and the feature is not mentioned in the original patents. However, a 1949 refinement to the original design described in US Patent #2,474,994 describes this feature.

That makes it sound like a software "feature" that gets added to the docs because it turns out to be useful in an unexpected way.

If you design a feature and it doesn't work, you don't get credit for the design. Who wants a screw that will self-destruct when it is correctly torqued? Which, by the way, is a value that widely varies and you can't just design one screw head to limit torque for every situation.

I've had Phillips over-torque and eat threads. I've had them strip to a raw steel crater long before the head even meets the surface it's being screwed into.

Definitely agree that Phillips is obsolete. God I hate them. I may have a problem. I've worked on a lot of old engines in my life and seeing a torn-out cross is like a big middle finger from the previous person who worked on it, usually the manufacturer. An engine is something that deserves an easily removed fastener that is applied with a torque wrench. Who knows how many newton-meters a Phillips head takes before it cams? Does anybody? Probably a huge range.

>I've had Phillips over-torque and eat threads. I've had them strip to a raw steel crater long before the head even meets the surface it's being screwed into.

>Who knows how many newton-meters a Phillips head takes before it cams? Does anybody? Probably a huge range.

Yep, that's the problem with them. Maybe they made sense decades ago when assembling aluminum aircraft and they didn't have widely-available torque-limiting screwdrivers. But now they're ridiculously obsolete, and the only reason we still use them is because of sheer inertia (i.e., everyone has a Phillips-head screwdriver handy, but lots of people do not have a set of Torx drivers at hand; also, the inertia makes them cheap). They should have been retired ages ago. If I were dictator, I'd simply ban them (not allowed in new designs, no new Phillips screws allowed to be sold, but drivers of course are fine for removing old ones to replace with better fasteners). There is simply no good reason to use Phillips now.

I prefer square drive to Torx for larger screws. But I agree with Philips... yikes.

If by flat head you mean “-“ shape then the main reason is that if you don’t have the drill bit centre plain in the exact same plain as the screws then it’s a nightmare to remove. For metal of pre threaded holes this isn’t an issue but drilling into wood where even with a pilot hole you can be a few degrees of perpendicular to the surface flat heads are significantly slower to work with.

Square heads are also acceptable. Kreg uses them for pocket hole screws and I like how easy it is to get the bit in when you're in a tight spot. They resist cam out pretty well also, I've not yet had a screw of either type I couldn't easily remove.

A good flat head screw is better than a crappy Phillips but a bad one is worse. The bad ones seem to strip by design.

No, they don't just seem to: Phillips was specifically designed to cam out. It was invented for this very purpose. When you're cursing with a Phillips screw that's stripped-out, you need to know that the fastener is doing exactly what it was designed to. The problem is people expecting these screws to be easily removable; that was never the intention. And the other problem is people using them for applications where they might want to remove them one day; that again is contrary to the design goal and intended use of the Phillips screw.

The linked article says that there isn't a lot of evidence that was a deliberate design decision.

https://en.wikipedia.org/wiki/List_of_screw_drives#Phillips

"There has long been a popular belief that this was actually a deliberate feature of the design, for the purpose of assembling aluminum aircraft without overtightening the fasteners. Extensive evidence is lacking for this specific narrative, and the feature is not mentioned in the original patents. However, a 1949 refinement to the original design described in US Patent #2,474,994 describes this feature."

My issue with kreg is that the driver isn’t a perfect fit for the screws so when you take them out you have to knock them off the driver tip somehow. It’s like it’s an imperial metric mismatch but they are all official kreg parts so don’t know.

I think that's a feature, not a bug. You can stick the screw on the driver, and even though it's not magnetic, it'll stay on, allowing you to get into a tight space easily one-handed.

You don't really remove pocket hole screws much (though you certainly can with much less cam-out than Phillips head) so it's optimizing for the presumed one-way trip. I would assume anyway.

Almost every time I remove a pocket hole screw it's because I'm trimming something glued up (like a table top) and didn't plan for the saw blade when I put in the screws. Every time I think "heehaaw" and then I do it again the next time.

Reply on news.ycombinator.com