MS20426AD3-5 1

MD...

MS20426 [spec NASM20426] PART specification is pretty clear...

MS20426AD3-4*-5 length = 4.5 x L [+/-0.010]
MS20426AD3-5* length = 5.0 x L [+/-0.010]
MS20426AD3-5*-5 length = 5.5 x L [+/-0.010]
* = finish code if/as applicable
L = 0.0625

Seems a bit odd specifying 1/2-length [1/32] increments for #3 Dia rivets... but OK...

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

The engineering drawing (B/P) calls out a -5. That is what must be used unless there is MRB disposition, or if for an in-service repair a note in the SRM allowing different grip lengths. Doesn’t matter what is in the MS spec; drawing supersedes the spec.

consider nas1241/1242 rivets

sorry, you're thinking about length ... is it That critical ? go up a size and have a slightly oversize head.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

Okay so the print specifies rivet length...
What are the tolerances for the shop head dimensions... depends on what spec you use.
I've been known to take a long rivet, eyeball it with side cutters and drive it. As long as the shop head is in spec who cares.

That said, I find that when rivet length is on a print, it's largely for procurement. Production will buy those rivets. There will be pounds of those on the line... vs those 4.5 and 5.5 are going to be in a small box controlled by someone who knows what they are. Go talk to your lead sheetmetal tech, ask him to do a few samples of what you're trying and see what you get. I have a job that needs 426AD3-3 if a tech is really good... the shop head dimensions are right at minimum. Some guys like 426AD3-4 but new techs sometimes "nail them over". So yeah we Specifically put -3.5 on a print for procurement.

with a default tolerance (on length) of 1/32" (= 0.03") ? (smile)

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

EDIT: keep defaulting to diameter.

#3 rivet shop head minimum diameter is 0.122, driven head thickness .038 to .050.

One of the subtle aspects of assembly is 'pre-tightening the parts/layers together to eliminate gaps/slack'. IF designers have done their job right, then grip-length should be 99% accurate.

In a few cases 'high-side-thickness tolerances'... materials and finish [primer, sealant, etc] and material/surface waviness... can bugger thickness calculations... forcing the 'practical need' for longer-length fasteners. Grrrrrrrrrrrrr.

In most cases I'm aware of, standard shop practice allow mechanics latitude to use 'next standard longer length' fasteners as required. The real problem comes when purchasing buys 100,000 fasteners of length X AND REALLY NEEDS LENGTH X+1 for practical purposes... then availability of the longer fastener... and 'waste' of the shorter fastener... becomes a real problem.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

Just get long rivets and cut them down a bit.
Every structures tech will have this tool in their toolbox:


To the OP's specific question, my answer depends on what process control you have in place to guide your riveters. If there is no defined riveting technique in your company, then each person defaults to what they learned from school and the guy working beside them last week. When riveting technique is intricately defined in your company, and the personnel are trained, then you have these "process tools" available to give them definition on cutting rivets to size, when they need to, and the drawing can now call up rivets longer than necessary and people can just deal with it.

I've only been designing aircraft parts for 20 years, and I don't think I've ever specified the rivet length on a drawing or in a 3D MBD part/assy. We just specify the diameter and the location of installation. Then they install the rivet per the company's installation specification, which tells the installer to measure the grip length once the hole is drilled.

I can see value in calculating the expected grip length for ordering / SCM purposes, but not for installation purposes.

K_W_K...

Simply remember... company design/drafting practices and Bean-Counter [procurement] policies vary all-over-the map.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

Korey with a K said:

I don't think I've ever specified the rivet length on a drawing or in a 3D MBD part/assy. We just specify the diameter and the location of installation. Then they install the rivet per the company's installation specification, which tells the installer to measure the grip length once the hole is drilled.

Click to expand...

Yes, there is very good reasons to try very hard to never put rivet lengths in approved data since its sensitive to the installers technique plus your level of process control and design never has sufficient time to chase though all the changes that could result, though I have spec'ed length for field kits but with a bunch of +1 lengths but that was mainly to reduce the numbers of form 1's that came with the kit.

Hope this makes sense...

Hmmmmm... the OP... Miguel D... appears to have stepped away from this thread...

KWK, all... in my experience...

Prototype/pre-production assemblies... should give fairly accurate 'sniff-check' for making last-minute adjustments to drawing sheets and PLs for 'grip-length'. AFTER That... Production should make for fairly stable 'stack-ups' and grip length should be monotonously stable.... and procurement can get the 'exact right parts' to the shop floor

'Hard fasteners'... Bolts, Lock-Bolts, Blind-Bolts, Blind-Rivets, 'specialty pins', bushes/spacers, etc... MUST have all fastener elements perfectly described/specified... long-lead times and Big-Buck$... demand it.

I expect that experienced mechanics could select their own/team's solid/driven rivets from 'shop-floor-bins' for small repairs and modifications fairly reliably.

I have noted that in shops where the assembly mechanics are 'in-charge of selecting the right GL rivets for every riveting task' for bigger jobs... after they get into the 'rhythm and swing of things'... THEY, with the help/blessing of their immediate leads/supervisors... will set-up an informal-but-organized 'system of rivet bins for every job/Assy to speed-up their work'. YEP, NOT STUPID.

BUT-BUT-BUT... trust, BUT verify!

Here is [just-one-of-several] PRECAUTIONARY 'noshit' experience that I had [+28-years ago]. In one USAF sheet metal shop... filled with young mechanics... I was asked by one mechanic "why some rivets are 'easier-to-shoot' than others??". THAT was an ODDLY disturbing question!! Soooo I went poking-around in the shop rivet bins... and found identical size/color aluminum rivets... EXCEPT: (a) most had a 'Raised Teat' head-code marking = 2017 'D' rivets... OK, the correct rivet for their job. HOWEVER, I ALSO found... mixed-into the same bin-trays... a lesser percentage of rivets that had indented 'Dimple' head-code markings = 2117 'AD' rivets... which are significantly lower strength that the 'harder/stronger' 'D'-rivets. Collectively, we discovered other bin-trays [different diameters and GLs] with this SAME mix-up-of rivet types.

OMG... NCOs and I inspected several aircraft in-work... where work was on-going replacing solid rivets in a flight-critical splice-joint. We found randomly mixed/installed 'D' and 'AD' alloy rivets the full lengths of every joint. The junior NCO/officer shop supervisors were in disbelief... and the shithitthefan. Several Jets in-work and few flying jets were affected. What a hot mess: Someone had dutifully stocked the same bins with [2] very different alloy rivets. This is just one of MY many-similar experiences.

Thanks to one young airman asking me a simple and puzzling question that was bugging him... we identified a serious issue and were able to take 'swift/painful' corrective actions.

My conclusion: well-meaning mechanics often aren't experienced-enough to notice fine details and spot when 'something isn't kosher'... and speak up.

Are You using the NAS523 FASTENER CODE, or corporate equivalent? IF SO, then You are committed to the NAS523 'system' which clearly include GL in the code.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov