Strength of aluminum threads

...

Rule of thumb I've used is four threads. 20 TPI would give you nearly five.

1/4 x 20 might be a good choice. 1/4 x 28 might be slightly better, both common sizes. There may be other reasons in your design to go to #12, #10 or #8 fasteners. I'd consider stainless steel machine screws.

As too how many you need to hold the force, there are tables in Machinery Handbook. If there's room, I'd way over design. Perhaps others can chime in.

Karl

I think he wants an external thread on the 2.5" OD AL tube and the inside of a larger than 2.5" dia end cap, not a multiple bolt / screw assembly. So essentially a 2.5" dia thread.

Just a theory of failure: The coverdisk (the aluminium-part) gets bent outwards by the pressure. So it is lifting a bit off of the o ring. The result was a small gap into witch the o-ring was pressed into.

What I would try: Make some ribs onto the disk to make it stiffer. Make the circular grove a bit shallower, so the pressure onto the o-ring gets a bit higher and just right when the disk bends outwards. Maybe you can reduce bellowing by adding a screw in the center of the disk.

Dimensions and tolerances for o-ring groves get critical with high pressure!

Nick

Yes exactly correct.

The new assembly idea is to thread the outside of the 2.5" aluminum.

The old assembly was 10X #8-32 in shear and it was strong enough, but it messed with the o-ring on assembly.

Its actually an aluminum tube around the out side. You can get a better idea of the construction here,

the black carbon from this drawing has been replaced with solid copper. the part I'm trying to fix is on the right side of the linked picture.

By putting a threaded cap over the outside of the tube I'm reenforcing it.

I've also comtemplated a snapring closure, but it's can still mess with the oring and is probably not strong enough.

Paul

Oh! That first link to that "technical" drawing was quite misleading! :-)

I bet I know what you made wrong: You had no camfer on the copper throat. You just slipped the O-ring over that part into the grove thinking that it is just stinking rubber. Right? It ain't! Read the manufacturers recommendations *carefully* how to design the "ramp" where the o-ring has to slip over. Also get the right hardness. More pressure, harder O-ring. Don't buy them in a shop where they sell two or three a year and store them away in the opened bag somewhere near the window. They have to be fresh, out of the sunlight and away from oxygen. That *makes* a difference. BTDT.

Nick

What about a shoulder or step between the ring and the bolt holes, to constrain the ring from expanding outwards?

my copper throat.

loads on a hard start

You want to thread the OD of the end of the tube?

3/4" sch40 pipe has a wall thickness close to that of your tube; a 3/4 NPT thread is 14 TPI, so that's a starting point. If you look a the table of Unified Screw Thread dimensions in Machinery's Handbook you'll find several pitches for 2-1/2 threads with the relevant dimensions.

The pitch of the screw won't have much impact on the shear strength of the threads, assuming you get a reasonable number of threads in your available 0.24 inches. I'd be comfortable with at least four threads, so a 2.500-20 UN thread seems reasonable. A very crude calc of the shear area of the threads yields about 1.8 in^2 of thread in shear. 2400 lb / 1.8 in^2 = 1300 lb/in^2 which looks OK, depending on temperature and the alloy and temper of the aluminum.

Looking at the minor diameter (2.437") of that thread, the remaining area of the pipe wall would be 0.68 in^2 => 2400 lb / 0.68 in^2 = 3500 lb/in^2. Again, OK depending on temperature, alloy, and temper. A finer thread (larger minor diameter) would give you a bigger margin here, at the expense of a thread more prone to galling and seizing.

Ned Simmons

I can't help with the thread query, but this site will give you anything you need to know about O-rings and the design of the parts holding them. Tom

Although the black carbon from this drawing has been replaced with solid copper.

I do a lot of instruments at that size using 0.75mm pitch threads.

Machinery's Handbook will help you analyze the thread strength for a given geometry. And of course you have to consider the bulk that will be left on your 0.125 wall tube after the threads are cut into it.

If you split the 0.125 evenly between bulk and thread, as a rough take on maximizing strength, then a 0.0625 thread height would be 0.0625*2/sqrt(3) = 0.072 pitch = 13.9 tpi. Nearest standard sizes up are 2.5-14 UNS and

2.5-16 UN.

Sorry 'bout that.

Plagiarism is the sincerest form of flattery.

My Limb Lopper brand pruning shear has exactly the O ring and seal you're talking about. Its 3" diameter and runs for years at 150 PSI or a bit over

1000 lbs. force.

The ring and cap use 24 TPI and 8 threads total. The O ring sits after the threads in a groove and the tightening sleeve compresses the ring in quite tight. The ring is mashed by the assembly and must be replaced every time.

I'll be glad to take pics and measurements if you decide to copy this idea.

Karl

A rough take on maximizing strength would have it nowhere near an even split - you want shallow threads.

John Martin

Depends on how much thread length you have, and how close you can tolerance them.