1 Large Nut vs. 8 smaller ones

[morbark]

I need some help deciding between two different methods of sandwiching 8 rods (axially) between two plates. Currently, we use ONE 8 1/8" nut on the outside of the plates, that's threaded onto the shaft. I'm thinking that we could try using EIGHT 1" bolts that would go through the plates and into the rods (ie, need to thread the ends of the rods). I 'feel' these bolts would be stronger than that one 8 1/8" nut...especially due to the fact that the nut only has about 1 3/4" to 2" of thread engagement.
Please Help!
Thanks

 

[Metalguy]

It's difficult to envision just what you have, but bolting issues have quite a few factors. What materials are involved, and what are the loads--just for starters.

 

[desertfox]

Hi morbark

From my experience of designing joints it has always been good practice to use smaller bolts or screws and increase the number of fixings, in preference to using larger and fewer fixings, one reason for this is that it spreads the resulting clamping load more evenly over the joint you are designing.
If you could furnish us with some more information about your plates and rods ie:- materials, torque loading on existing single nut etc as the previous writer has requested
I maybe able to assist you further in your quest.

 

[TheBlacksmith]

Many smaller fasteners also allow you to acheive equal or greater clamping forces with reasonable torque values on the fasteners. On steam turbine applications lately I have seen the emergence of multi-bolt/jack fasteners where a large nut is hand tightened, then the force is applied by torquing 6-8 small fasteners through the nut to about 100 lb-ft, instead of tightening the big nut to about 1500 lf-ft with a slugging wrench, torque multiplier or bolt heaters. Same principle should apply here.

Blacksmith

 

[CoryPad]

To answer morbark's original question, if the old and new fasteners have constant mechanical properties, and constant total cross-sectional area, then the 8 small fasteners are not stronger than 1 big fastener.

Good points mentioned by desertfox and TheBlacksmith. One disadvantage to multi-bolted joints is complicated/unknown fastener interactions (sometimes referred to as crosstalk). Tightening one fastener can unload another fastener. Remember to use a pattern that tightens bolts located across from each other.

 

[morbark]

OK...I have a drive shaft intended to turn a grinding hammer mill. There are two plates on the shaft that that 'sandwich' rods that run between them. The ends of the rods are at the inside faces of the plates (the faces facing each other) and the plates are intended to squeeze together as much as the rods will let them. The method of moving/squeezing these rods together is currently to put a large 8 1/8" nut on the outside of these plates and thread it onto the shaft...thus squeezing the plate toward each other and the rods in between. Oh, and the rods are radially, equally spaced around the shaft at a given radius.
The nut is currently made from 1 1/2" thk, T-1 steel w/ a yield of 100k.
The shaft is made from 4150 steel.
The rods...that would be threaded to hold the 1" bolts that will go thru the plates and into the rods (this is what would hold/squeeze the plate together on the ends of the rods...proposed) is made from 8620 Cold Finish Round...and is then heat treaded to 44-46 RC.
The loads are VERY difficult to envision. It's a dynamic pounding/beating that this mechanism is going to be accomplishing. The 'hammer mill' turns at about 1800rpm and is meant to grind/shred wood waste, stumps, trees, etc.
ANY help that any of you can provide would be GREATLY appreciated!
Thanks very much,
Bill

 

[Metalguy]

OK, here goes. *IF* you are measuring the nut correctly (thread size), your 4150 shaft is almost certainly fairly soft (UTS~130K, YS~80-85K). But if you're measuring the OD (flats), then the shaft is smaller and could be harder.

Your small 8620 shafts apparently have been case-hardened, or they were quenched in water or brine. Makes a big difference when you're going to be tensioning the whole cross-section.

More when these points are cleared up.

 

[morbark]

Right...the nut is 8 1/8"...the 'flats' are at 12" b/c that's what we make them (we mfr. the wrench too). So yes, the shaft is a bit soft (32RC). But it's UTS is listed as 156k and it's yeild is listed as 138k...so it's pretty heavy duty stuff.
And you're right on the rods...they're hardened to 44-46RC and quenched & tempered.

 

[morbark]

Thank you all for the help.
The 'cross talk' would be an issue here b/c there will be a fair amount of disassembly and reassembly with this devise...it's rather high maintenance. Thus, one of the things we always try to incorporate into our designs is simplicity.

 

[Metalguy]

This may be academic, but if you have the ability to check the shaft hardness near the center of one end I think you'll find it a lot softer. It's difficult to harden something that thick in 4150, short of a very cold quench in brine-which would likely crack it.

I'd also check the ends of the 8620 rods-same story. Once you know what you REALLY have-strength wise- you can start to make an intelligent choice about what you need to do. It certainly sounds like you CAN use 8 smaller bolts/rods to achieve what you need.

 

[morbark]

Metalguy,
Actually, I'm calc.ing that the 8 bolts will be weaker than the one big one. I'm using the equation in the other tread that you and I are chatting on "tensile capacity/min engagement", given by CoryPad. Maybe I'm missing something here, but I show the eight 1" bolts holding quite a bit less force than the one big 8 1/8" one.
What do you think?

 

[Metalguy]

I suspect you are performing the calcs. (which I'm not) using the strengths you have reported here. BUT, I suspect those numbers are too high-given the size and materials.

Since you have so little thread engagement on the large shaft, it's full cross-sectional strength isn't going to limit anything. Measure the hardness near the OD threads and use 60% of the equiv. UTS (they relate fairly well) to determine the shear strength of the shaft threads. I suspect THEY will be your limit (the 1 1/2" length of them), not the nut-which is *probably* harder.

When you end up with the pounds/tons those shaft threads will withstand, you can then figure out how many bolts/rods of what ACTUAL strength will equal/exceed the shaft value.

 

[diamondjim]

Morbark
A very rough calculation would be
1 times 1.00 inch squared = for one inch bolt
8 times .125 squared = 8/64 inch squared,
they would be very weak in comparison.
It would not be as great as what I have
shown. But do the values suggest something
similar?

 

[morbark]

Metalguy,
Could you please explain how to relate the hardness that I would find, with the UTS...(that I would then take 60% of to get the shear strength).
Thanks

 

[TVP]

Hardness has a linear relation to ultimate tensile strength for steels across a broad range of strengths. You can obtain the relationships by viewing the information in standards like SAE J417 Hardness and Hardness Number Conversions or DIN 50150 Conversion of hardness values for metallic materials.

The 60% value is the ratio of the ultimate shear strength to the ultimate tensile strength. Shear strength is often the limiting factor is threaded fasteners, not tensile strength.

 

[morbark]

Understood...and thanks.
I just have never seen the relationship b/w hardness and UTS.
Thanks again!

 

[Metalguy]

There are charts avail. that show the relationship of hardness to UTS. Note that there is NO direct relationship between YS and hardness.

If you can directly measure the Brinell hardness (BHN) which presses a 10 mm dia. steel or tungsten-carbide ball into a surface with a load of 3,000 kg, you can take that number and divide it by 2. That will be close to your UTS.

Simple, huh? <g>

 

[morbark]

The calc should be easy enough now that I will have the correct data. Though, there are still issues such as the one CoryPad presented w/ the tightening procedure. Sometimes things like this are almost impossible to direct for us. Our customers are going to do whatever's easiest. Thus, we usually over-engineer to accomodate.
Actually, in doing to calcs. previously, the material shear strength would have to be A LOT lower than the #'s I was using to get down to equivalent stength as the 8 bolts....
I think I'm going w/ the one big nut (which I feel like right now).
Thanks!