Question: 6 point vs 12 point Sockets

[leehljp]

People will tell you 6 pt sockets are less likely to round a head, but it's "urban legend" and has been disproven in testing.

The error in this statement is probably because of the way that the "disproven" was done - in some kind of controlled climate in which each test was done exactly the same under the same conditions. If you keep doing the same thing over and over again exactly the same you will get the same results again, over and over again.

I agree from pure experience what these other fellow have said. I am pretty sure that under controlled conditions that I could prove what you are saying - under specific controlled conditions. However, I know that I have had more heads rounded with 12 points than with 6 points. But that is reality where I grew up and situations where I worked.


A scientist worth his salt would try to find out why one group of people (non-scientists) consistently have similar results that disprove the consistency of the scientific study. What needs to be done is for those that did the testing to not make everyone conform to their tests or try their tests, but take the facts of numerous users and figure out just why this does not happen always out in the real world of "making a living" for every one.

So, my question to you is: Why do people out in the real world of everyday living get different result from your scientific experiment/study. Are you saying that we are "stretching the truth," as it relates to our experiences? Are you saying that we are exaggerating our experiences and favoring one side over another? What can you say about the fact that quite a few of us experience one thing that is contrary to the "scientific" study? Is it possible that the study is not an absolute for all situations and conditions that we, as users out in the field, are experiencing something that is NOT being taken into account for those scientific tests?


I have a high regard for scientific testing, but Boeing's testing only works for me when I am flying not when I am riding my lawn tractor.

[Cochese]

The problem with controlled testing is that it is designed to exclude what you're most likely to see in real life - user error or variables.

Which is why it really doesn't apply here.

[LCHIEN]

FWIW I find some merit in Woodturner's arguments. But...

Sometimes in life the truth lies somewhere in between and people's perceptions take over. Personal Experience is sometimes biased to the conditions under which one works.

In this case lets just say there are good technical reasons for both to exist, independent of which are better for high torque, bolt-rounding, corroded fastener performance. And, since there appear to many varied opinions as to which are better under torque, and there are other reasons to prefer one over the other, its easier for manufacturers to offer both rather than try and persuade customers to buy one or the other.

I imagine the OP has already gone out and bought his BIL's Xmas present and it sitting under the tree in all its 6- or 12-point glory. Time to rest your keyboards now.

Thank god for the moderators this isn't a debate over whether the tree should have a 5- or 6-point star on top...

[Dal300]

"Scientists investigate that which already is;Engineers create that which has never been.''

“Information is not knowledge.”

“Insanity: doing the same thing over and over again and expecting different results.”

“The only source of knowledge is experience”

“Great spirits have always encountered violent opposition from mediocre minds.”

“If the facts don't fit the theory, change the facts.”

~Albert Einstein

[chopnhack]

I'm surprised your last line didn't get moderated Loring ;P

With all this heated debate I am wondering if my question raised anyone else's interest or if it was lost in the battle? This has always been my pet theory as to why the six point purportedly does better. Hopefully one of the engineers can weigh in on that.

[Mr__Bill]

Hmmm
I measured the inside of a 3/4 socket, both 6 and 12 point and guess what! The 12 point has a larger inside diameter or a larger diameter round rod will fit inside a 12 point than a 6 point. This tells me that more steel has to be rounded off with the 6 point than the 12 point to get to the point where the socket will not turn the bolt or nut. To put it another way it's possible to round off a head with a 12 point socket and still be able to turn it with a 6 point.

Now I'll grant you that I haven't measured all sockets or even a mathematically significant sampling of sockets but.......



Bill

[leehljp]

Hmmm
I measured the inside of a 3/4 socket, both 6 and 12 point and guess what! The 12 point has a larger inside diameter or a larger diameter round rod will fit inside a 12 point than a 6 point. This tells me that more steel has to be rounded off with the 6 point than the 12 point to get to the point where the socket will not turn the bolt or nut. To put it another way it's possible to round off a head with a 12 point socket and still be able to turn it with a 6 point.

Now I'll grant you that I haven't measured all sockets or even a mathematically significant sampling of sockets but.......

Bill

Thanks Bill for a little scientific application (actual measurements) to this process.

A properly tightened Crescent wrench and a properly fitting open end wrench do as much gripping on the flat side as it does on the corners. I can't tell you how many times that I have used a crescent or open end to tighten in tight places - with the wrench pointed down and just gripping the middle - no where near the edges. Proper fitting grip will add torq towards the middle of the head instead of the points. Sure, a socket is looser but no looser than an open end wrench. But the 6 points have more gripping area than a 12 point and the 6 points have the mass in the middle to act like the crescent and open ends that are pointed downward.

[LCHIEN]

actual dimensional model

I modeled a hex bolt head or nut in a 6 and 12-pt socket.

I got dimensions from about a dozen new bolts (.428 to .4375), nominal 7/16 head for a 1/4-20 bolt is .4375 - So a spread of about 9 mils.

I checked a craftsman wrench, two SK sockets and a off brand "KD" socket and got .443 to .444 opening across the flats - pretty tight.

So I modeled a largish .4375 bolt (blue) in a socket .445 unrotated and this is what I got - not the very small amount of clearance (see 1st attachment) - about 4 mils clearance all around or the thickness of a dollar bill.

Then I rotated the bolt about 0.1° so that it contacted the walls of the socket. (see second attachment).

Then I did the same for a smallish bolt (.428 which gives around 9 mils or clearance all around) and I found that it would take about 3.8 degrees of rotation to make it contact the walls.

Anyway this should give you an idea of the tolerance and gaps involved. Study the contact points and note where the point of contact moves. ALso note that the point of contact in a properly fit 6- or 12-pt and a good hex head is really the same...

My observation here (and this is based upon a limited study and sample size) is that tools have much tighter tolerances than the bolts four tools varied by only 1 mil and 12 bolts spanned 9 mils.

Its clear that as the bolt head gets smaller to some point - would apply to a rounded off bold as well, that the 12 point will give up contact sooner than the 6-pt.

This is only for one size bolt head and socket but I find that these things (standards and tolerances) are usually scaled up quite well so it should apply similarly to all sizes.

As for strength, given the same outside diameter socket the thinnest and weakest point is at the points, this is equally thin for both 6- and 12-pt sockets, but the 12-point will have twice as many weak points.

So my conclusion is that the 6-point will work better on severely rounded bolts, but that the 6- and 12-points will work equally well on good bolts.

But, you draw your own conclusions. (note click on the thumbnail and maybe enlarge your window depending upon browser and image viewer for a full size image)


P.S.
From Machinery's Handbook 26th ed. the standard ANSI wrench opening s for 7/16 (.4375) head nuts and bolts is .440 to .446
and the allowable tolerances on nuts dimensions across the flats is .428 to .4375
So my measurements and observations are very correct.
It also says you should have a .750 opening for socket clearance to the nut.

[jdon]

Let me add another 0.02, based on some calculations I made. Keep in mind that 1) I'm not a mechanical engineer; 2) I had to delve into very dusty recesses of my brain to resurrect geometry and trig knowledge; 3) I've spent my share of time on my back underneath cars, rounding over countless frozen nuts.

Consider the following: in rounding off a six sided bolt, how large would the resulting circle (disk) would the bolt head be? Secondly, how much of the bolt would be "shaved" off the corners in the process?

For simplicity, I assumed a 1 inch bolt.

I calculated that the area of the bolt is .866 sq in.

A 1" 6 pt wrench would round off to a circle 1" in diameter (0.5" radius), with an area of .785 sq in., a decrease in area of 9.4%.

For a 12 point wrench, each "point" of the wrench is separated by 15 degrees. By repeated diagrams and calculations, I conclude that the radius of the innermost points is 0.5"/cos(15 deg), or 0.518 inches. The resulting circle is .842 sq in, or a decrease of the bolt area of 2.8%.

To me, it seems much easier to remove 2.8% of a bolt head than 9.4%.

In other words, it's easier to round a bolt with a 12 pt wrench than a 6 pt, all other things being equal.

I'm avoiding terms such as "shear" to reduce confusing with bolt shear strength already discussed in prior post. Also, I've not considered force vectors, etc, exerted on the bolt facets.

Attached is a rough drawing, withe the hatched areas representing the "shaved" areas of the respective bolts. I apologize for the poor quality of the picture- cell phone in ambient light.

[Mr__Bill]

The problem in this discussion is that two different things are being discussed and intermixed.
It's plain from Loring's graphics that an undamaged nut in an undamaged socket have the same surface area in contact. End of that discussion.
It's also plain from Loring's graphics and jdon's that a lot more of the corners of the nut must be rounded off before the socket can no longer turn it with a 6 point socket. End of that discussion.

Thanks Loring and jdon for the great graphics makes it easy to see that apples and oranges were under discussion at the same time.

Bill
over here where the sun sets last

[Mr__Bill]

6 point sockets are intended for use on 6 point bolts.

12 point sockets are intended for use on 12 point bolts.

It's really straightforward, but for some reason folks insist on using 12 point sockets on 6 point bolts. I don't understand why. Use the right tool for the right job. I bet many of the folks who use 12 point sockets on 6 point bolts would freak if you used their chisels as screwdrivers. It's the same concept though, just because the tool works doesn't mean it's the right tool for the job.

I have to say that this made me laugh, good one!

Bill

[Stytooner]

Good information here. Many different thoughts and results are posted. My whole point was that it is not myth that sometimes a 6 point can work better than a 12.
I agreed that in good quality hardware and tools, there is likely no difference.
The result of a test using those type things would probably be bolt shear.

You have to lower the test standards or change some variable to get to the correct testing bed. What is true for the high quality test may or may not be true now. My point is that the results from these might be different and when they are, that is not myth.

Now an easy test. Using low quality tools and fasteners. Round off the head of a bolt with a 12 point socket while tightening. Now try to remove it with that socket. It's iffy. Force a 6 point on there and you can generally tighten some more as well as remove it. I know what makes that work and why.

This method does abrade the test results, but it's just another of the Real World variables that were probably not included in a controlled test and that any Joe with a tool and a stubborn bolt can observe.


My conclusion from reading through this thread? Add good quality Vise Grips to any quality tool box just to be on the safe side.

[All Thumbs]

This entire thread has been fascinating.

I just spoke to Snap-on engineering. I posed the following question: "When working with 6-point fasteners, which tool, 6-point or 12-point, can apply more force without risk of rounding the head of the fastener?"

Snap-on engineering answer: "6-point. A 12-point tool carries a greater risk of slipping or rounding the head."

Do I trust their answer? You bet.

[cabinetman]

My conclusion from reading through this thread? Add good quality Vise Grips to any quality tool box just to be on the safe side.

Vise grips..yes. Years ago I got a gift of "universal wrenches", and at the time I thought they were just a fluke. But, for that bolt head that's destroyed, better than vise grips. They'll grab whatever flats are there.



.

[cabinetman]

This entire thread has been fascinating.

I just spoke to Snap-on engineering. I posed the following question: "When working with 6-point fasteners, which tool, 6-point or 12-point, can apply more force without risk of rounding the head of the fastener?"

Snap-on engineering answer: "6-point. A 12-point tool carries a greater risk of slipping or rounding the head."

Do I trust their answer? You bet.

+1. I agree...it's been fascinating. The 6pt are regarded as the better socket. Impact sockets are predominately 6pt for a reason. But getting back to the question of this thread, which wasn't which socket will round bolt heads or all the testing discussion, but which point count to buy as a gift. So, I repeat myself, that as a gift, I would give both.

.