2024 "DD" alloy rivet heat treat... and other 2

[RoarkS]

Howdy,

Okay so I'm teaching a structures class for A&P mechanics and I'm falling down the rabbit hole on more than a few things. One of them is 2024 "DD" alloy solid rivets.

AC43.13-1B 4-57 d talks all day about heat treating and quenching and dry ice and such but has no specifics. I am looking for FAA acceptable information.
I have a VERY complex old coffee stained Boeing standard for doing this... but want a general industry standard. If I have to buy it so be it. One of the labs is heat treating rivets and I want it done correctly because I don't think we have been doing it correctly in the past. Previously we would heat treat them, then put them in a freezer at 0°F for up to 3 weeks before we would use them... They had already hardened.

Also to my shock looking at the specs for 426DD and 470DD rivets it says "inactive for new design"... what is the new design?

Thanks!
-Roark

 

[KSshopper]

Trick Question!
The FAA acceptable information for mechanics doing structural repairs will always be the Structural Repair Manual (SRM). If the SRM doesn't refer you to a more in depth fastener installation/processing spec (i.e. BAC5004), and doesn't include any information on how to heat treat the rivets, you should be able to use MIL-H-6088 as an acceptable reference for heat treating and quenching the rivets. MIL-H-6088 is on everyspec look at page 26 Table 2 for 2024 rivets.

For your second question, MS20426 and MS20470 rivets were redesignated NASM20426 and NASM20470. Basically the military stopped maintaining many military standards and offloaded them to industry groups. AIA now owns the spec and you have to purchase it rather than get it free like MS20426.

A couple of comments... DD rivets are a pain to install even if properly heat treated. Your lab may be doing it right, but they are still a pain. A good reference for processing icebox rivets is Air Force TO 1-1A-9, but it doesn't fit your FAA acceptable requirement. It's on everyspec too.

 

[RoarkS]

Some aircraft don't have an SRMs, kind of like this lab project... MIL-H-6088 looks to be pretty close. It mentions the heat soak temp, but says little about quench and "icebox" process.

I have BAC5602 page 31 and 32. If that procedure reflects what really needs to be done, our lab has most certainly not been doing it right. sub zero quench fluids!!! I wish I could find something else to confirm I really need to do all the BAC steps. I would be afraid to present that one because I think BAC are proprietary and it was written in December of 1941!

Air Force TO 1-1A-9... 3-36 and 3-37 I think meets my FAA requirement perfectly... I've got stuff flying on standards not nearly official as that. Reading what I have in it already again reflects that we've been doing it wrong. It specifically says that -10°F is required for long term storage. 32°F presumably down to -10°F says it's only good for 24 hours. Our freezer is at 0°F. Again they sit for weeks.

On a side note I'm shocked TO1-1a-9 still calls and cast alloy 142...142. It was changed to alloy 242 a long time ago.

NASM... got it. Yeah not paying for that... Clinton...

Thanks for the info!

 

[Sparweb]

A few points that hopefully will keep things in perspective...

I have designed repairs literally on dozens of aircraft and only once or twice spec'ed a "DD" rivet. Very old ships (Bell 204B is the only specific one I can remember) that hail from the 1950's. I can ask around among the old-hands in structures at work, if any of them remember ever driving a DD in their careers, and if so where. My point is that the industry has moved on from these fussy things and I doubt your students will ever need to know this, and if so probably won't be responsible for the process of procuring and storing them even if they do.

Another thing: What exactly do you expect to have been changed in the intervening 20 years since the MS20470 became NASM202470? I am lucky, I suppose, to work where I have access to the printed sets of NAS binders, and I'm at liberty to download the old free copies from ASSIST any time I want. Upon comparison of these and many others, I have found extremely little evolution, except for record-keeping changes like the "QQ-A-225" becoming "AMS-QQ-A-225" and so on. It's a bit of a protection-racket.

In the past 10 years or so, I have noticed a new trend. There are designers now specifying close-tolerance, 180ksi, A286, high-temp, unobtainium bolts and screws, for extraordinarily mundane purposes like holding curtain rods and magazine pockets.

If you want to leave your students with any kind of valuable lesson, let's hope it's one that provides some judgement when the right fastener is the common and simple one to use and install, versus when the only way to do it right is to match the fastener carefully to the application.

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

 

[WKTaylor]

TA4L...

AN426 SSB MS20246 which has been SSB NASM20426 [Part numbers remain MS20426]
AN470 SSB MS202470 which has been SSB NASM20470 [Part numbers remain MS20470]

SSB= superseded by

Also 'ancient' alloys numbering systems for aluminum and magnesium alloys were replaced by the modern alloy numbering systems... See... FAA ANM-112N-06-04 FORMER ALLOY AND TEMPER DESIGNATIONS, ALUMINUM AND MAGNESIUM ALLOYS for everything You need-to-know on this subject...

Also 'cold driven' D [2017-T41] rivets tend to have been used by manufacturers ILO DD [2024-T41] simply because they can be driven cold and have shear strength just under DDs properly installed.

Also... there are 7050-T731 [E or KE] rivets which can develop shear strengths in-excess-of DD rivets... and can be driven cold... with experience.

NOTE. DD [ice-box] or AD [2117-T4] rivets are used/preferred [better ductility when bucking the tail] when reverse-flush riveting or double-flush riveting ['NACA Method', bucked-tail into 100-Deg or 82-Deg countersinks].

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]

 

[RoarkS]

@SparWeb

I understand that it is outdated. The Mechanic school curriculum is "approved" by the FAA and I'm stuck doing it until it is changed. I'm currently testing the waters getting HiShear replaced with HiLok in another project... Mainly because aircraft spruce wanted nearly $300 for 25 HiShears! I'm still new there... with a bunch of old guys and don't want to rock the boat too much.

What is useful is that we typically do this lab when we are talking about metals... Steel, aluminum, alloys thereof, heat treat. I use a bucking bar dropped in a tube of a specific height to comparatively show the difference heat treating makes... works pretty good to drive home the concept. Also gives them an appreciation for work hardening for those who set the rivet gun so light it takes 100 hits to drive a -3. I honestly don't grade them on how well they install the DD's and usually have them drill them out and install AD.

I miss having ASSIST. I started at a new company last year that is 100% GA. I know that the NASM are probably identical. Trust me I get it... I've had customers spec out inconel hardware on... a curtain rod. That's funny.

But I figure if we're going to do it, I would like a "real heat" treat procedure.

 

[Sparweb]

I miss having ASSIST

Why do you miss ASSIST?

The URL moves around every few years or so, and sometimes goes dark for a while, but so far, it has always popped up again.
My current workplace has security protocols on the web browsers that throw up alerts and "are you sure?" messages before connecting to ASSIST, but it does still work for those who are persistent.

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

 

[RoarkS]

LOL! Truth. It had all the old stuff... usually in better scans than what I can find on everyspec and they often had the new current mil specs I needed. I was the registered data custodian for the company, so I got to ask why a lot and find out a lot of stuff.

 

[edmeister]

Before this topic slides of the original mark ..
I would like to point out a 'still valid & effective use' for DD Rivets.
Granted other varieties of fasteners have made the DD rivet obsolete in most situations ..
- but given a MRO that has the facilities to 'cook' and correctly use these fasteners .. some great savings can be achieved.
In situations where holes are to be plugged in difficult locations - nothing beats a DD rivet. Say we have a 1/4" dia hole to be plugged in a frame with difficult access - what are the options? 1/ AD/KE/DN O/S rivet - nope .. too difficult to install at large Diameter. 2/ Hilok .. nope .. large Dia in thin material 3/ machine & finish a freeze plug .. expensive & time consuming 4/ Cutout & patch frame. 5/ DD rivet !! because it will be more ductile than an AD rivet to install. Easily squeezed with a pneumatic tooling.
- I use this fastener in most cases where a freeze plug is used. It can be procured in variety of diameters & comes in Prot & Csk head. It is already surface finished (anodized) & eliminates the need to outsource the finishing (WRT freeze plug). Because it swells in the hole (at installation) - the final hole size is not too critical - thus a good candidate for difficult situations.
- Take the the case where a critical fitting is to be installed against a worn-out hole in a flange? O/S the hole on the fitting or replace the the flange? How about O/S hole in the flange & trimming DD CSK rivet head to 1/3 Flange thk & install. Drill thru DD rivet. Easy! Instant Interference fit bushing.
I have seen countless $$ and hours wasted machining & sending out for finish interference fit bushings that end up 'spinning' at the end .. or additional damage inflicted in the structure attempting to 'smash in' a O/S AD (or KE / DN) rivets.
- The 'cooking' takes place in a mini-oven (temp - regulated) & immediately immersed in dry ice upon removal. (similar to interference fit bushing installation)
- as someone previously said .. right fastener for the right job! .. my 2 cents.

 

[RoarkS]

That's interesting application. Thanks for sharing.

Just a follow up, I did the lab... I followed the old Boeing procedure exactly.

I had a extreme temp thermocouple for mt fluke, I got an ultra low temp thermometer to double check.

I chilled alcohol on dry ice, got it to -45°F and immediately quenched the rivets within the time limits. (I had 2 students on stop watches) it was great fun!

Everyone's project was waiting for the final 8x 470DD5-4 rivets... 192 rivets were driven in less than 15 minutes and every one of them were perfect! I couldn't have been more excited!

I drove one... I was stunned! It was MUCH easier to drive than even a 3-3. It's obvious to me now that the way the lab had been conducted before was completely wrong. They kept the rivets in a freezer that averages 7°F for up to a week... even at that temp they had obviously hardened to the point they were not feasible to drive.

That was the most fun I've had in a while teaching a class.

Thanks for all the input!

 

[Sparweb]

Thank you Edmeister for the tip about plugging errant/oversize/worn-out holes with DD's.


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

 

[WKTaylor]

Ed... a few added comments...

You have good points regarding DDs as repair for oddly OS/shaped holes... however odd/OS holes often harbor other damage that drilling/reaming/deburring eliminates... validated by [usually] EC-NDI... before plugging.

There is also a chance that HT goes sideways for a small batch of rivets and they are accidentally annealed... and never spontaneously age-harden to -T42... which is a structural crisis.

Large HT DDs 'shoot [bucked or squeezed] like butter' and are useful for plugging tooling holes and crack-tip stop-drill holes with minimal distortion.

Anodized D rivets may also be 'shot' as HT/ice-boxed or may be 'shot' in the fully HT [-T42] condition with effort... and almost no defects. In either case bucked D rivets are marginally lower in strength relative to DDs... but may save time/effort.

NOTE. There are actually code differences in NAS523 for D rivets to be bucked 'cold' or HT/quenched/bucked/aged 'cold'.
MS20426D BD 2017-T31 Rivet, 100° CSK Head (Driven Ice-Boxed)
MS20426D BE 2017-T4 Rivet, 100° CSK Head (Driven Hard)
MS20470D BL 2017-T31 Rivet, Universal Head (Driven Ice-Boxed)
MS20470D BM 2017-T4 Rivet, Universal Head (Driven Hard)
NAS1097D MB 2117-T4 Rivet, 100° CSK Shear Head (Driven Hard)
NAS1097D MK 2117-T4 Rivet, 100° CSK Shear Head (Driven Ice-Boxed)

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]

 

[RoarkS]

@WKTaylor

"There is also a chance that HT goes sideways for a small batch of rivets and they are accidentally annealed... and never spontaneously age-harden to -T42... which is a structural crisis."

Anything more you could share on that!? or possibly a current spec I could read up on?

 

[edmeister]

Mr .. Taylor ..
I never intend to install the DD-rivet into the original hole that has become damaged, cracked, etc.. without rework.
The damage in the hole would be removed by increasing the diameter to the next size, usually -7, -8, 10 Dia.
if hole was 'worn out' universally EC-NDT would be performed around the perimeter of the hole.
Bore would be deburred & primed per SOP.

I would just like to emphasize that this DD rivet application has some benefits that are not realized with other fasteners.
1/ Ease of installation - exceptionally malleable when removed from interm cold storage.
2/ Surface Finish (anodized) benefit inherent .. 'out of the bag'.
3/ Available in variety of flavours; different diameters + Prot & csk heads.
4/ Can be used as a quasi - interference fit bushing - whereas Hilocks or jobolts cannot. Redrill correct centre bore allowing for correct wall thk.
5/ Bore does not have to be a high tolerance diameter.
6/ Fairly 'smart' installation in thin skins - unlike a hilok with spacer washers on either side + the nut.
7/ Unlike a interference fit bushing - a minor csk on the opposite side of the flange will 'lock-in' the rivet post bucking.
8/ WRT edge distance to the flange - rivet can be 'pulled' so that hole center is shifted towards the 'meaty' section of flange & the original hole dia. redrilled eccentrically WRT wall thk.
9/ Installation fairly expedient - does not require high skill level. Possibly some minor lathe work to trim the head & shank to length. Rivet ends to be touched up with primer.

Agree .. that some possibility may exist for small batches to go 'sideways' .. that is 'remain annealed' ..
- but this is why an inhouse process should be developed where 1 or 2 rivets are sent to QA from every HT batch, QA will perform a hardness test to quantify the HT result post couple days @ room temp.
if fail - job to be redone .. just as everything else can fail sometime ..

 

[WKTaylor]

Ed...

You have listed good/valid Points... indicates how installing solid rivets is not necessarily straight forward... and may diverse aspects.

As You suggested... a rigorous QA/QC program MUST be instituted to ensure processes are followed to the letter... including follow-on hardness testing of samples.

RoarKs

I have often heard the term 'annealing DD rivets'... which would kill spontaneous hardenability. These rivets would install like ice-box rivets... but would never age-harden to -T4 to MIL-HDBK-5 strength.

One of the greatest problems with HT of small parts is the 'quench-delay' time being exceeded... which can be very short! 'Quench-delay' is the time between removal from solution HT [salt] bath or [air] oven... and when parts actually contact the water for shock cooling. Anything longer which allows parts to cool below a certain temperature [that is very close to SHT temp] will result in rapid annealing instead of shock-quench to the 'W' temper.

Anytime SHT/quench/ice-box is done by non-heat-certified techs... IE: typical shop mechanics... things can go wrong. FAA AC43-18 Fabrication of Aircraft Parts by Maintenance Personnel... explains critical nature of fabrication issues... which aren't far removed from critical shop processes like HT rivets.

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]

 

[RoarkS]

@WKTaylor

Thanks for the info!

I followed the Boeing procedure precisely, I had 2 students be "inspectors" run a stop watch to make sure furnace removal to water quench was less than 15 seconds. I didn't fully understand the implications of the 15 seconds now I do!

I also preformed what I would call a comparative harness test to demonstrate the effects of heat treating. I had off the shelf presumably full hardness, dropped a weight in a track from a specific height (round bucking bar in some unistrut) and recorded before and after deformation of the rivets. I reliably got a delta of .025" pre-heat treat and .037" after heat treat. After the weekend we smashed some again and they were right back to the mid .020". So I inadvertently did some hardness testing! lol.

I will have a look at AC43-18.

-Roark

 

[edmeister]

Since i cannot seem to get off this DD rivet train ..
Interesting story ...
In stores we usually have a bags of various diameters & types (Prot, CSK, NAS1241/NAS1242, etc ..)
-usually the longest we can obtain - trim down to the length when required ... otherwise difficult to locate+purchase on AOG Basis.
As so happens .. these rivets are sold in 1 LB bags (same as all other rivets) & there are approx (guess) 200 units per bag .. -8, -10 Dia ..
When a rivet (or several) are required - the few items are removed from the bag .. the bag & remainder remain on the shelf ..
In most cases the inventory does not get changed - because the 1 Lb bag is still there ..
.. Our every - observant Bean-counters review the inventory stock levels & the quantities that 'move' (get used) annually.
Since this 1 Lb bag is always on the shelf & only 4 - 10 individual items are ever removed per year .. The quantity of '1 Lb bag' remains as 1 Lb for years ..
The occasional removal of items from the bag does not get accounted for in inventory .. just the 'traceability' reference goes along with the removed items.
- As would happen; in the middle of a 'do or die' situation - one day .. none of these fasteners can be found anywhere in stores. All the 1 Lb bags were gone!
Lo & behold our ever - attentive Bean-counters had disposed of the 'Dead' stock in the name of efficiency ! Long story short .. details (shortcuts) will always bite u at the end !

 

[WKTaylor]

Ed...

"...bean-counters had disposed of the 'Dead' stock in the name of efficiency!..."

It sounds like you are talking about DoD/DLA loggies on critical long-lead items that are consumed sparingly... never knowing how incredibly valuable something is when its absolutely needed. To bean-counter-loggies 'what-ever-it-is' is just another 'item on the shelf' taking up space... not a diamond in the rough.

Where I work, each major shop area has formal 'hidey-holes' for items like these rivets... and for other hard-to-find 'residue' [oversizes, etc] from modification kits... along with their certification paperwork. Every 4-to-5-years, or-so, loggies get in a lather to 'clean-up/clean-out' these 'hidey-holes' and have been mostly unsuccessful because shop leads raise the red-flags and validate 'real-world' value in these parts. The few times the loggies have been 'successful', work-delays/chaos due-to awaiting these hard-to-get parts/items, has re-taught the lesson.

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]

 

[bnrg]

Another interesting story, and it isn't always the bean counters!

I worked a few years at a small commercial machinery equipment manufacturing plant, that generally had a pretty good reputation. Once during a lull period in sales an unexpected OEM RFQ came in for some assemblies needed quickly that had not been made for maybe a decade. A quick check showed that amazingly we had the needed machined castings in stock so the RFQ was quoted with the requested delivery and the resulting order placed. This plant was located in a small town where nearly all manufacturing personnel were related in one way or another, and at one time inventory control fell under the manufacturing group but that task had been taken over by the 'out of towner' computer people some time before. When the box of castings was sent to assembly everything fell apart. The castings were junk, all machined wrong and could not be salvaged. Turns out those particular parts had been in there forever, but that box of junk had been machined by so and so's nephews brother in laws cousin and he would have been fired if the bad case of castings had been discovered so they were hid at the back. 'Everyone' knew that when an order came in for those assemblies to order an extra set of castings to 'be ready' in case a future rush order came in, but the extra set was actually for the manufacturing run set to be run to cover for the inventoried defect parts in the box marked in a special way and rotated to the back of the shelf. But since 'everyone' had moved on or up or out that piece of information was lost. No name ever came out that I heard and even this explanation was only divulged over beers well after the fact.