NAS/MS/NASM Fastener Head-Shank Fillet Clearance 2

[EnginerdNate]

Hi all,

First post here, let me know if I'm not in the right section.

I'm working on some bolted shear-tension joints. I'm currently planning to use NAS6206 bolts and MS21042 deformed thread lock nuts. This application is my first exposure to this series of fasteners and I have a question regarding clearances between the washer and the head-shank fillet under the fastener head. I did a bunch of searching and found this thread:

https://www.eng-tips.com/viewthread.cfm?qid=441917

In which WKTaylor states that the minimal interference between the small head-shank fillet on these fasteners (The NAS620X series being the reduced thread length non-CRES version of the NAS670X series) and the standard NAS1149 washer is acceptable. That said, the OP never answered WKTaylor's question regarding the calculated interference. Using the worst case scenario, with the washer at 0.380" and the maximum fastener OD and root radius of 0.3745" and 0.025" respectively, I'm calculating a maximum clash of ~0.022" which seems like a lot.

I don't mean to question the conclusion here as much as understand it. In my previous job, we used tension head fasteners with large root radii almost exclusively on the design details I was a part of, primarily MS21250s with MS21299 countersunk washers under the head with NAS1804 nuts + flat washers on the nut side. In my current application I'd like to avoid having separate types of washers under the fastener head and under the nut, because I know they're going to get mixed up during production. So I want to believe the conclusion regarding the NAS1149 washers, but I haven't found any info so far that gives me a "warm fuzzy" about it, other than the single post referenced above. Can anyone else shed more light on this issue?

Cheers and thanks,
Nathan

 

[Sparweb]

The ID that Wil quoted was the minimum threshold of the tolerance, so yes, it is possible.
The average dimension is 0.390" with a tolerance of +/-0.010"
Digging through my pocket I find a number of NAS1149's picked up on a recent FOD clean-up.
On close examination I see a radius on most (tho not all) of their ID's.
The NAS1149 spec doesn't give a radius, but there's a radius nonetheless.
Chalk it up to "deburr and break sharp edges" which is specified for NAS1149, but without the specific "how much".

So if ID=0.380" falls outside the 6-sigma band from the washers factory, and 0.3745+0.025=0.3995" is the average from the bolt factory, then the odds of having a bolt head radius on a sharp washer ID are about 1 in 100. Can you live with those odds?


No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[EnginerdNate]

Thanks for your input! I didn't think to look at it from a probability standpoint.

If I were assembling these things myself, 1 in 100 would be perfectly fine. The issue is figuring out how to write a work instruction/tech manual that prevents someone else from assembling one with a clash.

As a tangential question-how bad a practice would it be to buy all countersunk washers and use them under the nut as well as the head? I'd be less concerned about instructing them to get the countersink oriented correctly on the head side if there wasn't the chance they could assemble things completely backwards (flat under head, countersunk under nut).

I didn't worry about this as much at my last job as it was a prototype environment and I was directly involved with the build in as a technical lead so I got eyes on almost all of the joints I worked on before they were buttoned up. My current project could be in production long after I've moved on and I'd rather not get a phone-call that a design decision I made caused an accident due to a joint loosening after a washer or bolt failure due to interference.

Thanks,
Nathan

 

[SwinnyGG]

Considering that the first thread on the underside of the nut acts like the nut itself is countersunk, I don't see much penalty for having a countersunk washer under the nut.

It shouldn't change the actual contact area between nut and washer very much, if at all.

 

[EnginerdNate]

I've been trying to remember if we ever did that at my previous job, but it's been ~4 years since I left there (Went to grad school inbetween jobs) and my memory is fuzzy.

 

[Sparweb]

As a tangential question-how bad a practice would it be to buy all countersunk washers and use them under the nut as well as the head?

Actually, Bombardier does that. In some places. Beside other places that don't. With similar bolts. When I figure it out, I will try to tell you.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[WKTaylor]

'Nate...

You are comparing apples and oranges...

The fastener installation You initially described, and what You described have experience with are vastly different parts for vastly different purposes.

NAS6206-* bolt + [2] NAS1147#06** washer + MS214042-6 nut = general high precision installation with high shear strength and moderate tensile strength and lowest possible joint weight [for steel]

MS21250-6-* bolt + MS21299C6 + MS21299-6 + NAS1804-6 nut = general high precision installation with high shear strength and high tensile static and fatigue strength and moderate joint weight [for steel].

NOTEs.

'Pull-up/review' the specs for all these parts and see how they actually fit together. The physical differences will jump out-at-you.

IF the bolt is installed with no washer under head [assumes BOLT does NOT rotate to torque], then the edges of holes will have to be chamfered to accept the bolt fillet radii.

Washers isolate the structure from rotational torque of nut and/or bolt-head which damages finishes and may score base-metal.

Washers intended for shear application are usually 'paired' with hex-head or even double-hex-head bolt/nut styles... which have similar small bearing shoulder-areas under the heads. Generally speaking, too, washers for use in shear applications are less critical for bearing strength... hence thickness is usually not critical. Installation torques are relatively 'modest' to insure adequate tensile preload... without excessive preload that could compromise shear capacity.

Washers intended for tension application are usually 'paired' with tension head bolt/nut styles that either have a deep socket or a double hex with wide flange/washer-bases... which have wider bearing shoulder-areas under the heads. Generally speaking, too, washers for use in high tension application are bearing strength critical... hence thickness and static bearing-strength is critical. Installation torques are relatively 'high' to insure adequate tensile preload... that mitigates effects of tensile-fatigue cycles with far less emphasis on shear-fatigue.

NOTE.
Shear-joints could be made satisfactorily with tension rated parts... but there would a high $$-cost for parts [intended for higher performance] and a high 'cost' for the unnecessary added weight/mass... and of-course special installation practices to reduce installation forces preventing damage to lighter/lower-strength structure.



Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[EnginerdNate]

So is the big factor the lower torque? I figured that an interference was an interference, though obviously the magnitude of the interference is smaller with the smaller root radius.

 

[WKTaylor]

EnginerdNate...I see my previous response was TLDNR.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[EnginerdNate]

I've been over the specs numerous times I did read your post. We've got a design review coming up and I'm in the process of buying a house, so apologies if I didn't spend enough time crafting my response.

Here's my takeaways from your post:

1. The parts are different. I mentioned this in my first post when I stated that the root radius on the MS21250 was larger.

2. I should pull up the specs. I've got all of them here and I'm well aware of the physical differences. MS21250s have a huge head-shank radius by comparison. The MS21299C* washers have a correspondingly large countersink that's meant to go under the bolt head to relief this radius. Part of the reason I started looking into this in the first place is that I noticed, in the specs, that the bog standard AN6 has the standard AN 0.010 +-0.005 head-shank radius. I know that this fastener can be used all day every day with your standard AN960/NAS1149 washers. The NAS6206 on the otherhand, has what I would call an "intermediate" root radius by comparison, at 0.020 nominal +-0.005. This is what setoff my initial inquiry-the program I'm working on is still using the old AN bolts all over the place and I wanted to move to something a bit better for this particular joint without breaking the bank. NAS6206's are definitely pricer than AN6 bolts but not anywhere near what some of the high strength tensile or tensile-fatigue rated stuff costs.

3. Regarding installation without a washer - was this meant to address both bolt series or just the 21250s? I could easily require a 0.050" chamfer on all the holes in this assembly and call it done. I originally wanted to use a washer instead because even 1 minute of chamfering on a part is going to cost more than a $0.05 (bulk cost) washer. Torquing will be done on a workbench so there's no need to spin the bolt in either case.

4. Shear joints are generally torqued to a lower value to preserve the shear capability of the fastener. This is what inspired my last post after my first quick read through of your post - does the lower torque have any impact on the interference or lessen it's impact? My intent with the last post was to further discussion, not be flippant. If something I said came across as dismissive or hostile it wasn't the intent.

5. Regarding the bearing shoulder areas - (I see you've gone back and bolded those statements so it must be important) - I can see from the specs that the tension rated stuff has a much larger outer diameter. I don't see, at the moment, how this impacts the interference at the head-shank interface. I apologize if I'm being dense here.

General comments:

In certain combinations, even with the smaller root radius, if the washer is not chamfered (even if they tend to be chamfered a bit by the finishing/deburring process, it's not called out on the spec, I have no means of quantifying the size of that chamfer, and therefore can't reliably count on it from a design standpoint) there is a potential for an interference between the washer and bolt head, even with the smaller head-shank radius on the shear rated parts. a 0.015-0.025" radius is small, but it isn't zero.

I realize that the very outer edge of the radius doesn't stick out axially very far (kind of the nature of a radius...) so the actual interference will be somewhat less severe than it sounds just looking at the numbers.

Looking at the tension rated parts it's very obvious that the root radius is going to need some clearance. Because, as you said, it's large. Going to the shear rated stuff, yes, the radius is smaller, but based on the numbers I'm pulling from the datasheets, there is still the possibility of two pieces of metal trying to occupy the same physical space. I even went as far as to CAD up a parameterized version of both parts so that I can look at the clearances/interferences at different tolerance conditions visually. (I needed a library of parameterized fasteners anyway...)

There are a couple possible explanations I have thought of that could allow for this interference:

1. In practice, the washers have an inner radius large enough to clear the typical head-shoulder radius. Over time, companies/engineers have found the production to be consistent enough that though a countersink isn't called out in the spec, there isn't a significant clash.

2. If an interference exists, it's small enough that the washers yield a tiny bit without damaging the bolt, and everything is gravy.

Cheers,
Nathan

 

[WKTaylor]

'Nate

What are Your corporate deburring spec requirements?

NAS6206 requires a very tiny chamfer to clear the bolt head-to-shank fillet Radius... generally within the range of an aggressive deburr... as opposed to a break sharp edge deburr, needed on fay surfaces edges of holes. Obviously the nut side of the hole can 'live' with the 'break-sharp edge deburr'.

IF YOU are close reaming the bolt-hole, then deburring with a light 45-deg chamfer, then You would meet our corporate spec... and bolt head would seat tight/flat against the mating surface [assumes 90.00-deg holes or spot-faced-holes on-axis].

IF material stack-up has substantial tolerance variations, then added washers... under nut or head & nut for overly-long grip length... or thinnest possible washer under just the MS21042 nut for absolute min-GL... and Your job is done at the lightest possible joint mass. Minimum mass parts and parts cost = cost effective joint. For a small number of bolt installations, then You can go anyway You want. For a large number of bolt installations, the added costs/mass for extra washers, extra grip-lengths and 'minimally deburred sharp edges' will 'add-up'.

NOTE. This Joint design assumes that bolt-head will be restrained and only the NUT will be torque-turned.
---
Out of curiosity... RE this structural joint design... are equivalent Hi-Lok pins/collars unacceptable? HL's generally make the lightest/most inexpensive joints in large quantities... especially with titanium pins and aluminum collars.
---
Added Fud-4-Thot...

Washer crushing/bearing strength MUST be matched to the bolt/nut to avoid washer crushing which allows the joint to loose preload in-service... I have seen this happen in service. This problem is insidious and can be very destructive to joints.

I have an MS20002C16 countersunk 160-KSI washer that was inadvertently installed backwards under a tension bolt head [BACB30US16*?...a 220-KSI bolt]. The bolt fillet radius smoothly rounded-of the contact knife-edge of the lower strength washer ID. I'll see if I can get a photo of the washer posted 'for grins'.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[EnginerdNate]

Thanks Wil!

That's about as I expected. Our corporate policies here are not as strict or well laid out as I would like and they are different program to program. Part of the reason I have this question is that I didn't ever have to worry about this at my last job... there was a fastener installation spec callout and I knew the basics (partnumbers, don't leave threads in the shear area of the joint, etc) but the actual hardware choices were made by the program.

Since the last post, I have been pursuing the hi-lok route as it eliminates the washers and their contribution to the stackup. Add to that the fact that they are more "idiot proof" when it comes to installation torque and I think they're a good solution. I'd originally gone with the 6206/21042 combo because the stick out on the nut side was less than the equivalent size hi-lok pin, but as I refine my loads it's looking like I can simply downsize the pin one size to keep within my available space.

Thanks,
Nathan