threaded inserts for holding take down archery limbs

[jcs0001]

I have made threaded inserts in the past from low grade steel 1/2 in. x 13 tpi bolts - cut one inch of the threaded portion then drill and tap for 5/16 - 18.  They work ok but will rust if there is any moisture. 

I very recently bought some 304 stainless 1/2 in. rod and used it today to make some inserts.  Threading the outside to 13 tpi under power in my lathe worked well and gave decent threads.  However drilling it with an F drill then tapping it was a real chore.  It was so obnoxious that I don't want to use the 304 again for this purpose.

Inserts are just under 1 in. long and are pictured with a 5/16 - 18 bolt:


They are installed along with a pin for each limb of the takedown recurve or longbow:


The insert and pin are 2 in. apart and the limbs are about 22 in. (in a straight line) from the insert to the where the string goes around the tip of the limb.  The length along the limb is a couple of inches longer because of the curve.  The draw weight of the bow will never exceed 70 lb.

I can get stainless 303 in 1/2 in. rod and possibly other types of stainless locally and I can also get brass (do not know what alloys are available at this point).

1) Is 303 likely to be much easier to work with (particularly with drilling and tapping for the 5/16 bolt) than 304?
2) Is there any other stainless alloy that would be easier to work with and fairly readily available?
3) Would some type of brass alloy be strong enough for this purpose and if so what should I ask for?

Note: even though making these three inserts was a real bear, the positive part was that the 36 tooth change gear I made a couple of days ago for my lathe, worked like a charm and I learned how to use the threading indicator.

Thanks for any information you can provide.

John.

[DaveH]

John

Excellent piece of work, really well done

303 ss is much much easier to drill and tap than 304 ss. The trouble with the 304 ss it has a nasty habit of work hardening especially with drilling. 304 ss is best cut with lots of coolant and sharp tools. You will find the 303 sometimes called free cutting stainless a lot easier.

As for "strength" and this is a guess the weakest part seems (to me) the insert pulling out of the limb material - not sure what the limb material is.

 
DaveH

[jcs0001]

Dave:

Thanks for the reply.  I suspect you are correct - the riser (handle) into which the insert is "inserted" is wood - in this case laminated "actionwood" which is really a fancy name for birch/maple plywood.  The riser is drilled and tapped for 1/2 - 13 and the insert is screwed in using epoxy to provide better holding power but there is only so much that can be done for holding power.  Some bows I have seen actually use an insert with a wide bottom and it is screwed into place from the opposite side.

Sounds like I need to go out and mortgage the house for some ss 303 rod.  Is it fairly corrosion resistant?

John.

[jcs0001]

Just to be clear - I build the mold for the limbs, laid up the bow limbs and made the hardware however my friend did all the shaping and finishing on the riser and limbs.  Don't want to take credit when it isn't due.

John.

[Tony Wells]

303 will be fine for what you are doing. Easy to machine, and is corrosion resistant. From one online source:

Resistance to Corrosion
 The addition of certain elements to stainless steels to impart better machining characteristics also slightly lowers corrosion resistance. For dry conditions, and in most mildly corrosive environments, the free-machining grade is practically comparable with its corresponding unmodified type. Where moist atmospheres are involved, some free-machining grades may tend to form a rust film, and in certain severe environments, they may show somewhat increased corrosion as a result of the free-machining additions. In a majority of cases, they will perform nearly the same as the basic parent composition. Because the free-machining grades have a slightly reduced corrosion resistance compared with unmodified basic stainless steel, bright machined parts may be susceptible to surface dulling or etching by final treatment with nitric acid solutions. Caution in the use of such treatments is suggested. For mild action, mixtures containing as little as one or two percent by volume nitric acid, with additions of an inhibitor, may be allowable for short periods of time at 120 to 140°F (49-60°C). A solution of 12 percent nitric acid and 4 percent copper sulfate is also satisfactory where mild action is required. In some instances, the complete elimination of treatment with nitric acid solutions may be desirable.

[75Plus]

John, While you are looking for better machining stainless you might look at 416. It is easiest  machining of all the stainless alloys. Its not as corrosion resistant as 303 but it should be fine for your application. It is the material used in gun barrels.

Joe

[PekkaNF]

One thing comes into my mind. Looks like this part is heavily loaded static and dynamic

Does this joint relies on mechanical fastening (i.e. thread outside of the insert) or glue (and the thread is primary there just increase bonding area)?

Aluminium can be prepared fairly readily for bonding i.e.
http://www.boatdesign.net/forums/materials/bonding-aluminum-epoxy-2629.html

But stainless steel does not glue well. Because it will not etch well and therefore it does not corrode well either! Anyway to glue stainless steel you pretty much rough/grind surface and create a mechanical bond!
http://www.ybw.com/forums/showthread.php?t=235725

When you read those you will notice that the procedure is essentially same, but principles and chemicals should be all different. Wether this matters on this application or not I don't know. But if things start to disintegrate this might provide a starting point for further experiments.

I managed to bond aluminium pretty well for model airplanes this manner: Turned inserts, washed with hot water & detergent , etched them with Liquid Plumber, wash with hot water, immediate application with epoxy.

PekkaNF

[jcs0001]

Tony and Joe - thanks for the information about 303 and 416.  They sound like good options for the purpose.

PekkaNF - thanks for the info on bonding aluminium and steel with epoxy - it wasn't something I had considered.  I believe the strength in this application is due to the course threads and the corresponding threads in the wood.  The epoxy should help to strengthen the wood fibres and keep the insert from screwing out of the wood (although there isn't a lot of force in a counter clockwise direction on the insert).

The bottom line is that I should likely do some test pieces to see how much it takes to pull an insert (or break it or the bolt).  I doubt I will do this in the near future as I have so many other things to do but it would be an interesting small project.

John.

[Jonny]

They are available off the shelf but you may need to change to metric.
Think they are called threadserts.

Know where you get the laminate from, i have imported direct in the past inc that Royal Cammo.

[Tony Wells]

There is an off-shelf item that lock into place. I've used them a few places without problems.

http://www.newmantools.com/kee.htm

[jcs0001]

Tony and Jonny:

I've never used this style but have used some other thread in ones that may be some type of aluminium alloy (maybe with zinc).  I'm not keen on using the aluminium/zinc ones due to possible strength limitations and the internal thread seems a bit loose on them.

Tony - The ones you provide the link for appear to be 0.43 in. long in the 5/16 in. bolt size.  I feel that closer to 1 in. in length gives better gripping power in the wood.  Going to a larger than 5/16 in. bolt to hold the limb on would cause other difficulties.

Fortunately I don't need a lot of these, so given easy machining stainless I can make up a few here and there as needed.

Thanks again,

John.

[Lew_Merrick_PE]

John,

I've been somewhat busy over the weekend with a new (being paid for it) project that had to be ready for presentation today.  However, (1) if you go to http://www.scribd.com/Lew%20Merrick you will find the paper UN Thread Data Chart that will give you values for the shear area of internal and external UN primary threads listed as square inches per inch of engagement.  If you multiply that value times the shear strength (stress, in reality) of the material times the length of engagement (in inches), you end up with the pull-out strength of that side of the threaded joint.

I suspect that you will find that aluminum is much stronger than the crushing strength of the wooden laminate -- but that is a gut feel rather than a researched fact.  Key-locking inserts (Keensert is a brand name) do not do well in wood.  Among other things, the thread on the OD of key-locking inserts is actually only a 50% thread.

304 ss work hardens (as you discovered).  The trick is to be aggressive enough with your feed to cut away the area that is work hardening as you cut it.  Few home-shop machines are rigid enough for that type of machining.  303 ss is the better choice as it is much harder to work harden than 304 ss.  400 series ss (such as 410 or 416) actually machine better and more easily with a light heat treat (say Rc 32-38) as the chips will form better.

The companion paper to UN Thread Data Sheets is Pilot Hole Considerations (also posted at ScribD.com).  They may help you evaluate things.

[jcs0001]

Lew:

Thanks very much for the information. 

I did up a quick spreadsheet with figures for yield strength found on the net (Online Metals) along with the figures in your paper for area.  For all the metals I looked at, the "ultimate tensile strength" was higher so the relationship between them (noted below) would hold if that figure was used.

Interestingly enough it appears that 303 has a higher number than 304 and brass (c360 and particularly c464) has even higher numbers than ss 303 and 304.  I won't go into details on the actual numbers as it is the comparison that is important to me.

Does this make sense in comparison of brass vs. 303/304?


John.

[Lew_Merrick_PE]

John,

Developing good material properties values is seriously non-trivial.  Most of the information floating around the web is typical properties rather than minimum properties -- which may vary greatly.  Next, nearly all material property values are based on normalized condition of metals.  They are not accurate for work hardened materials or, in most cases, heat treated materials.

The thing to remember about thread pull-out strength is that it is based on the shear strength of the material.  In many cases, this value may be approximated as being .632 * UTS (ultimate tensile strength).  The thing to remember is that this is not always the case.  Most brasses have a shear strength in the 36,000 psi (36 ksi) range.  Normalized 304 ss has a shear strength of 46,250 psi.  Normalized 303 ss has a shear strength of 56,800 psi.  Minimally work hardened (25%) 303 ss has a shear strength of 79,000 psi.  Work harden it a bit more (50%) and the shear strength grows to 94,800 psi.  The problem is figuring out how much work hardening you have done.

I have about 30 inches of bookshelf devoted solely to properties of metals.  Getting actual values is far from trivial without taking samples and testing them individually.  Just remember, we are doing this for fun!

[jcs0001]

Lew:

I think I would die of old age before learning all that is available on this topic.  I will use my time for more productive activities   Since I am dealing with at most a few 100 psi force (say even 1000 psi just in case) it certainly appears that the wood is the weak point by far. 

I've ordered some 303 and when it's in will give it a try.  On my own bow I may use brass just to add a bit of bling.  Otherwise I will use stainless as it should be a bit more durable.

It's been an education and I certainly appreciate your responses and those of the other posters.


John.

[Lew_Merrick_PE]

John,

I do understand, I just try to give complete answers.  You probably do not want to ask me what time it is...  ;-)

Laminated wood products are in their own class when it comes to materials properties.  I have worked around the edges of them, but am no expert by any means.  Typical high-density hardwoods have allowable stresses in the 5000 to 7500 psi range (off the top of my head and without looking it up).  I have worked with high-pressure laminated wood products where allowable stresses reached into the 18,000+ psi range.  I would assume that archery products fall into that class of materials -- but that is an assumption based on no actual knowledge.

But, as a design and development (mechanical) engineer, I prefer to use (my favorite material) unobtainium!