What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Dimple / Countersink Sandwich

skelrad

Well Known Member
Friend
I am finishing up my horizontal stabilizer and riveting on the rear spar. When I countersunk the spar I basically did it to the depth of a rivet head. I checked a few times with a dimpled piece of sheet and the interface seemed pretty tight. Now that I'm riveting it together, the space is bigger. I'm guessing this is fairly normal? Should I countersink deeper in general? I'm not worried about the strength of this (right?:eek:) , just wondering if I should change my approach to measuring countersink depths in the future.
 

Attachments

  • 16338221976222703750252008525205.jpg
    16338221976222703750252008525205.jpg
    131.7 KB · Views: 319
"depth of a rivet head" isn't going to do it. Make yourself test coupons and do not machine countersink so much that the back side becomes a knife edge.
 
Countersink

Vans recommends .007" below flush. I use test coupons as well. I also double check so I don't over countersink past Vans recommendation.
 
Justme

Also the countersink leaves a sharp edge where it transitions from flat to the angled countersink. The skin dimpling could never be this sharp. So i broke this edge of the countersink using a piece of sandpaper and my finger or a piece of scotchbrite to let everything to nest better.
 
Vans recommends .007" below flush. I use test coupons as well. I also double check so I don't over countersink past Vans recommendation.

I did test it against a dimpled skin and could have sworn in nested perfectly, but obviously it was not so. Welcome to the education of homebuilding I guess. I will try .007" below flush next time and see what kind of difference that makes. Aside from a very slight wave between some rivets if you get the light juuuuust right, my current setup doesn't seem bad enough to redo. Guess I just lost my Grand Champion prospects, but I'll go cry in the corner in private and move on. :)
 
FYI- The actual Mil-Spec for driven rivets is that you shouldn't be able to slide a .002 feeler gage into the gap far enough to touch the rivet shank.

Obviously this doesn't work quite the same way on dimpled stock because the dimple gets in the way, but hopefully it will give you some idea of what you're looking for in regard to acceptable gaps.

FYI- in certain places, like countersinking the aft elevator spar on the RV7 where the trim tab hinge goes, it's impossible to countersink .007 below without knife edging stuff. in that case, it's just a balancing act to try to get it "good enough"
 
Also the countersink leaves a sharp edge where it transitions from flat to the angled countersink. The skin dimpling could never be this sharp. So i broke this edge of the countersink using a piece of sandpaper and my finger or a piece of scotchbrite to let everything to nest better.

That is next level commitment! I can't wait to see your airplane some day.
 
Thank you for all of the replies. I know just about every question that can be asked, has been. The absolute volume of information on VAF can sometimes make it harder to find answers, so I appreciate people taking the time to compile thoughts again for some of us guys who are just beginning to wrap our heads around things. That previous post with the examples shown was super helpful in my understanding. A picture is definitely worth 1000 words.

Thinking back on things, I think my problem arose because I probably tested a dimple against the countersink with a different thickness of dimpled metal. I didn't connect the dots that the dimple would be slightly different size depending on the thickness. A "well duh" moment when I think back on it. Reading through all of the info from Van's, it sounds like the strength of my joint is just fine at least, so I'll just put this in the hopper for next time and do better.
 
Last edited:
Vans recommends .007" below flush. I use test coupons as well. I also double check so I don't over countersink past Vans recommendation.

I don't understand this recommendation from Vans. I'm trying to get a good fit for HS710 and HS714 to the front HS spar. I kept inching deeper until the thinner dimpled spar wouldn't rock in the countersink but not so deep that it slides around. Now if I put a rivet in the countersunk substructure (HS710 and HS714) it is MUCH deeper than .007" from flush.

So the question is, stick with the .007" or keep cutting until the dimpled piece doesn't rock?

Also, wouldn't the depth of the rivet depend on the thickness of the dimpled material? Or am I not understanding Vans recommendation?

Thanks!
 
Countersink

I don't understand this recommendation from Vans. I'm trying to get a good fit for HS710 and HS714 to the front HS spar. I kept inching deeper until the thinner dimpled spar wouldn't rock in the countersink but not so deep that it slides around. Now if I put a rivet in the countersunk substructure (HS710 and HS714) it is MUCH deeper than .007" from flush.

So the question is, stick with the .007" or keep cutting until the dimpled piece doesn't rock?

Also, wouldn't the depth of the rivet depend on the thickness of the dimpled material? Or am I not understanding Vans recommendation?

Thanks!

1. Vans says .007". Stick with that. They have the engineering. Do not countersink so deep the edge of the hole is knife edge unless Vans says it's ok.

2. Yes. Drives me nuts. I actually have a full set of dimple tokens to check countersink/dimple fit. However, doesn't matter. Refer to #1.
 
The countersink has a sharp edge, it goes from the 100* angle to flat with the sheet top. The dimple has slightly rounded edges. The dimpled top sheet will never sit flush until countersunk sheet is too deep. It will all come together when riveting.

Van's recommends .007" deep, stick with that.
 
This video from Cleaveland Tools says to cut until the top piece is flush with the substructure (i.e. not rocking). Start watching at the 1:30 time mark.

https://youtu.be/IQyipOcCxgc

Using a fixed number like .007" regardless of the thickness of the material makes no sense at all. Also if you rivet an under-countersunk piece, then the two pieces will have a gap between them which is bad.
 
Having to replace an expensive structural part like a wing spar because of over counter sinking makes no sense either.
I suggest builders look at the sticky thread <countersink diameter for dimples> at the top of this forum page and check out the photos of the tests that we did that ended up resolving to the .007 inch depth recommendation.
 
Countersinks

I have a question. This issue of countersinks comes up often.
Has a Vans RV ever crashed attributed to a wing spar hole over countersinked?

It's kinda creepy to think about for sure.
I'm not saying it's ok to go past .007". I'm just curious, because I suspect there are a few flying with over countersinked holes. For whatever reason. Maybe they built with little experience, no outside or Vans influence.
 
My take

Just because a builder has made a mistake, and has eaten into the design margin, doesn't mean what he did was OK for the rest of us.

I prefer to follow Van's recommendations and preserve as much margin as possible.
JMHO
 
I have a question. This issue of countersinks comes up often.
Has a Vans RV ever crashed attributed to a wing spar hole over countersinked?

It's kinda creepy to think about for sure.
I'm not saying it's ok to go past .007". I'm just curious, because I suspect there are a few flying with over countersinked holes. For whatever reason. Maybe they built with little experience, no outside or Vans influence.

A crash is not the only negative result of bad workmanship (but, no, I am not aware of any accident that could be attributed to excessively deep countersinks).

What overly deep countersinks are likely to cause is a progressive degradation in a structure that would be first detected by smoking rivets.

Even if that is the only negative result, that is a reason that a builder will be advised (what they do with that advice is up to them) to replace a major structural member that has been severely over countersunk....... If they are told it is fine, and then 500 hrs of flight time later there is a lot of smoking rivets, the typical response now a days is for the builder to look for someone else to blame. Who do you suppose that will be?

Diving a bit deeper......
Keep in mind how a riveted joint works.
The primary load orientation of a rivet fastener, as used in aircraft design, is a single shear joint.
The simplest example of this would be two overlapping aluminum sheets joined with an AN470 rivet. Any load applied that would try and move one sheet away from the other is trying to "shear" the rivet across its diameter via the edges of the hole in each sheet, baring against the rivet shank.

With that in mind, look at the 3rd photo in THIS POST I made previously, were the countersink is only .004" over sized beyond what is recommended. Note that there is nothing for the dimpled sheet to bare directly against, until it moves enough that the lower face of the dimple comes into contact with the wall of the countersink. The rivet would take some load prior to that, but is purely in bending so it wont resist movement very well.

Detecting a .004" difference by evaluating the fit of a test patch dimple is very difficult. Especially since there is a large variation in dimple quality because of different dimple dies, tools used, and technique in using them.

There are already many posts in the forums that have pointed out that bottom shape of even the best dimple doesn't match the shape of a countersink. For this reason, making a test sample and countersinking for it to fit perfection flush with the surface that is countersunk will result in a dimple that fits loosely in the countersink (and usually worse than the .004 over sized example in the photo).

Wrapping up -
It is frustrating to us, to see single builders pushing specific build advice (based on their personal build experience, often on an airplane that hasn't yet accumulated very many flight hrs) that is contrary to recommendations that we make that are based on actual in depth testing (such as the rivet samples shown in the photos of the post I linked too. This is home building after all, so every one is free to choose. My suggestion is to do your research, and make sure you understand all of the factors and variables involved with a particular detail, before you choose.
 
Countersinks

A crash is not the only negative result of bad workmanship (but, no, I am not aware of any accident that could be attributed to excessively deep countersinks).

What overly deep countersinks are likely to cause is a progressive degradation in a structure that would be first detected by smoking rivets.

Even if that is the only negative result, that is a reason that a builder will be advised (what they do with that advice is up to them) to replace a major structural member that has been severely over countersunk....... If they are told it is fine, and then 500 hrs of flight time later there is a lot of smoking rivets, the typical response now a days is for the builder to look for someone else to blame. Who do you suppose that will be?

Diving a bit deeper......
Keep in mind how a riveted joint works.
The primary load orientation of a rivet fastener, as used in aircraft design, is a single shear joint.
The simplest example of this would be two overlapping aluminum sheets joined with an AN470 rivet. Any load applied that would try and move one sheet away from the other is trying to "shear" the rivet across its diameter via the edges of the hole in each sheet, baring against the rivet shank.

With that in mind, look at the 3rd photo in THIS POST I made previously, were the countersink is only .004" over sized beyond what is recommended. Note that there is nothing for the dimpled sheet to bare directly against, until it moves enough that the lower face of the dimple comes into contact with the wall of the countersink. The rivet would take some load prior to that, but is purely in bending so it wont resist movement very well.

Detecting a .004" difference by evaluating the fit of a test patch dimple is very difficult. Especially since there is a large variation in dimple quality because of different dimple dies, tools used, and technique in using them.

There are already many posts in the forums that have pointed out that bottom shape of even the best dimple doesn't match the shape of a countersink. For this reason, making a test sample and countersinking for it to fit perfection flush with the surface that is countersunk will result in a dimple that fits loosely in the countersink (and usually worse than the .004 over sized example in the photo).

Wrapping up -
It is frustrating to us, to see single builders pushing specific build advice (based on their personal build experience, often on an airplane that hasn't yet accumulated very many flight hrs) that is contrary to recommendations that we make that are based on actual in depth testing (such as the rivet samples shown in the photos of the post I linked too. This is home building after all, so every one is free to choose. My suggestion is to do your research, and make sure you understand all of the factors and variables involved with a particular detail, before you choose.

Thanks Scott.
I have no engineering knowledge. It was just curiosity and intrest to learn.
 
Wrapping up -
It is frustrating to us, to see single builders pushing specific build advice (based on their personal build experience, often on an airplane that hasn't yet accumulated very many flight hrs) that is contrary to recommendations that we make that are based on actual in depth testing (such as the rivet samples shown in the photos of the post I linked too. This is home building after all, so every one is free to choose. My suggestion is to do your research, and make sure you understand all of the factors and variables involved with a particular detail, before you choose.

It's frustrating for me too, because I was innocently taught the incorrect method when I first started building. End result is I'm out $3300 for new wing spars. I'm learning this lesson the expensive way, yet people continue to make my mistake. I shameless share my story so others don't do the same thing.

Do what Scott, the rest of the Van's team and the builder's manual says. It might just save your wing spars.
 
Back
Top