Doubloe dimpled shear strength

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,427
Location
North Carolina
Attempting to design a joint with double dimpled rivets and I cannot for the love of google find a value for yield shear using AD rivets and 6061-T6. Best I've found is 2024-T3 sheet and ADs. Anyone have some numbers?
 

BBerson

Well-Known Member
HBA Supporter
Joined
Dec 16, 2007
Messages
13,119
Location
Port Townsend WA
Are you talking about the sheet rivet bearing strength by thickness?
Should be in mechanics Airframe manual, or AC43.13
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,427
Location
North Carolina
I know bearing and rivet shear strength for undimpled protuding rivets, dimpled rivet joints are a bit stronger. At least they are with 2024 and 7075 sheet...
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,536
Location
NJ
Attempting to design a joint with double dimpled rivets and I cannot for the love of google find a value for yield shear using AD rivets and 6061-T6. Best I've found is 2024-T3 sheet and ADs. Anyone have some numbers?
I would think the numbers would be proportional to the difference of strength between 6061 and 2024....I am assuming by "double dimpled" you mean joining two dimpled sheets with AD rivets.
 

Victor Bravo

Well-Known Member
Joined
Jul 30, 2014
Messages
7,289
Location
KWHP, Los Angeles CA, USA
You can probably increase the shear strength by some measurable amount (perhaps significant) by dimpling the two sheets, holding them in place, and running a final size reamer through the hole. The reason for this is that the edges of a dimpled hole are sharp and jagged, creating far more of a sharp rivet-cutting tool than if it were a parallel sided "clean" hole. All the drawings in the sheet metal books show this jagged hole very clearly.

Yes of course the rivet fills up all the spaces in the hole, but you now have a definite stress riser (or two) created in the fastener.

Now doing this with a reamed hole would require a custom made or modified dimpling tool to be machined on a lathe. Not difficult, but you just can't buy it from Aircraft Spruce off the shelf.

The reason is that the dimpling tools have a pilot that is the same size as the rivet shank. So if you drilled a #30 hole, inserted the #30 dimple tool and dimpled the hole, you cannot make the correct final hole by running a #30 reamer through it afterward.

Or to put it another way, if you drilled #40 and used a #40 dimple tool, then tried to run a #30 reamer through it... the rivet shank would be a proper fit but the size of the dimple depression in the sheets would be too small for the #30 rivet head and it would not sit down flush.

So make a custom tool. A dimple die set that creates the proper size dimple for the AN426AD4 rivet but that has a #40 pilot and pilot alignment hole (instead of the normal #30).

Now you would drill a #40 size pilot hole, diimple it with this custom tool, then go back through the dimpled sheets with the #30 reamer to yield a perfect smooth dimpled hole for an AN426AD4 size rivet.

I would expect this new riveted joint to have the full shear strength of a regular AN470 "universal head" rivet, and I would expect the old textbook "jagged hole" method to have some amount less shear strength.

Yes it is extra work to create this tool, but I will bet that there is a measurable improvement in shear strength of the finished joint. You only have to make the tool once, and standard drills, reamers, Clecos, etc. can be used.
 
Last edited:

Mad MAC

Well-Known Member
Joined
Dec 9, 2004
Messages
612
Location
Hamilton New Zealand
Data for double dimple is not common. Had a dig through my documents all I found was the data in MMPDS-04 looks pretty much the same as ANC-5 and the statement below.

From the Lockheed Stress Memo 65
For flush rivets in dimpled joints the shear allowable shall be the same as the shear allowable of similar protruding head fasteners and the tension allowable shall be the same as the tension allowable of the countersunk fastener, in the same thickness of material.

The failure mechanism of a double dimple would seem to involve the rivet being under some tension due to the dimples disengaging which makes scaling by Fby from the test data a bit iffy .
 
  • Like
Reactions: BJC

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,851
Location
Memphis, TN
It’s in AC43-13 to be considered as equal to regular AD rivets of the same size. Now if you were designing an airliner with a rows of 40,000 rivets, they probably have special engineering data for that but not to be used for general usage.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,536
Location
NJ
You can probably increase the shear strength by some measurable amount (perhaps significant) by dimpling the two sheets, holding them in place, and running a final size reamer through the hole.
The standard way (as I understand it) is to de-burr the hole before you dimple it......That removes the sharp edges . I think I also have ran some fine sandpaper over the bottom of the hole after I dimple it to accomplish the same thing.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,536
Location
NJ
I think the OP was asking about "sheet strength" as was stated the rivet strength remains the same. The rivet does not care what sheet you drive it into.

The attachment shows the numbers for other than 6061-t6 and as I stated before I believe the 6061 strength should be proportional.

What is also interesting,..... reading the notes...the data is for a dimpled sheet into a machine counter-sunk plate. They are not measuring two dimpled sheets riveted together.
 

Attachments

Last edited:

Victor Bravo

Well-Known Member
Joined
Jul 30, 2014
Messages
7,289
Location
KWHP, Los Angeles CA, USA
The standard way (as I understand it) is to de-burr the hole before you dimple it......That removes the sharp edges . I think I also have ran some fine sandpaper over the bottom of the hole after I dimple it to accomplish the same thing.
I was not talking about burrs being sharp. I was talking about the fact that when you dimple the sheets you get a hole that has one or two narrow spots that are smaller diameter than other parts of the hole. After diimpling you get a hole where the upper part of the hole is significantly smaller than the bottom part of the hole. Even if the smallest part of the hole is still .1285 (for a #30), the hole itself does not have parallel vertical sides, and it seems to me that there has to be some stress riser involved in that.
 

Mad MAC

Well-Known Member
Joined
Dec 9, 2004
Messages
612
Location
Hamilton New Zealand
The attachment shows the numbers for other than 6061-t6 and as I stated before I believe the 6061 strength should be proportional.

What is also interesting,..... reading the notes...the data is for a dimpled sheet into a machine counter-sunk plate. They are not measuring two dimpled sheets riveted together.
Just be aware that if one checks that scaling between the various materials on the test data in MIL-HDK-5H doesn't appear particularly reliable*. However down is always more conservative so sanity checking against the protruding heads rivets to ensure the numbers arn't excessively conservative should work. Note that dimpled fasteners are really prone to skin cracking on the edge of the dimple so allow for this in fatigue critical locations.

Does any one have access to the copies of MMPDS later than -06, this data was dropped out MMPDS-05 so if its been put back in, it will include new test data.


*Some of the original data appears to have been scaled from one test data set. The more i look at that data, the more I am tempted to spend a couple of hours scaling the data to work out just what they did to get those tables.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,427
Location
North Carolina
I have a copy of ANC-5. All the dimpled values given are higher than protruding rivet values. It does vary with sheet thickness and grade, though. I can safely use protuding rivet values. But, if I have a borderline joint, an extra 10% strength may mean one less rivet in an awkward spot.

NACA tn-854 has the specs for dimple dies. It recommends a no35 drill for 24S-T sheet, which produces mostly parallel holes. Going small resulted in cracking. But that will be alloy dependant. Testing is never a bad idea...
 

wktaylor

Well-Known Member
Joined
Sep 5, 2003
Messages
184
Location
Midwest USA
Per ANC-5, MIL-HDBK-5 and MMPDS-[all] the only published allowables for rivets in double-dimple installation were for temper variations of clad sheet 2024-T3, -T42, -T62, -T81 [cold dimpled] and 7075-T6 [hot-dimpled] for MS20426AD, D and DD rivets.

6061-T6 has substantially inferior mechanical properties relative to 2024-T3, -T42, -T6, -T62... so dimpling 'allowables' for 2024-T3 and 6061-T6 are NOT comparable.

Regards
 
Top