Countersink Diameter for Dimples..

[TASEsq]

Hi all,

Thanks again for all your advice. Here is a follow up post. Sorry for the links in lieu of embedding the photos - this is more likely to work.

I used the vans practice project and deriveted the angle from the 0.025" sheet. I drilled 6 x #30 and 6 x #40 holes, and dimpled the skin.

I then (starting with the #30) countersunk in a few stages so that a rivet sat flush:
https://photos.app.goo.gl/nhXBHK6x9DfiFLUW2

I then noted the position of the arrow on my countersink cage, and moved it 7 "clicks" deeper.
https://photos.app.goo.gl/HUxy1Ap6mTdEok282

I hit the hole again, and the rivet looks like this now:
https://photos.app.goo.gl/a0L1Y6N7okhSQ66g1
https://photos.app.goo.gl/731Q2kmkgR7qlHxl2
Is this roughly what others are seeing when they are countersinking for dimples?
I measured the depth of the rivet in the hole (it is a bit hard to measure accurately), but it was 0.007" or thereabouts.

I then clecod the token to the hole. As you can see there is a gap, and this is where i think i may have gone wrong (?) the first time. I assumed this was not right, so kept countersinking until this gap was gone. This resulted in almost knife edge in the angle (which is 0.060" not 0.080" as previously mentioned).
https://photos.app.goo.gl/w2pKkE1QETOxekLT2
https://photos.app.goo.gl/0feq8E7Uc1SQKYjE2
https://photos.app.goo.gl/YoAih4J36RDZ7ygB2
https://photos.app.goo.gl/0u6tjZrGtdLJK1kG2

I then countersunk all the other holes on the angle as above, obviously resetting the cutter again to flush, then 7 clicks more when i swapped to the #40 cutter. As you can see the gap is still there with all the celcos in:
https://photos.app.goo.gl/jaJ43uyJpyVymSqp1
https://photos.app.goo.gl/xTb6XheAXIXjjBC93
https://photos.app.goo.gl/EC5xUjRMM6srJ8sC2

Thinking that the gap may go away when i riveted them together, i got out the back rivet plate and finished the 12 rivets (gee i like backriveting! - helps my counting skills!)
https://photos.app.goo.gl/X28y0b7jc0vsuqKz1

Alas, the gap is still there after riveting:
https://photos.app.goo.gl/fm3HbRtIQ6k1HDHQ2
https://photos.app.goo.gl/HXOWAZT2J8npNv8y2
https://photos.app.goo.gl/uLcE5hf3RphQJVth1

So here is the question:
1. Is this gap acceptable?

Obviously if this were a real piece, i would have broken the edge and the gap would have likely not been there visually. But is this correct strength wise?

Can i go a little deeper, if the material was a little thicker to make for a nicer fit?

Thanks all - sorry for the long post! Appreciate all replies and assistance in the learning curve i am on before the emp arrives.

[PilotjohnS]

What nobody has mentioned is one needs to make sure the dimple is very crisp, to minimize the radius trying to contact the countersink top edge. For critical items, like the tank dimples to the spar, i broke the edge of the countersink with some 220 grit sanpaper to help this meshing of the dimple and countersink; but this is not required.

[StressedOut]

Originally Posted by PilotjohnS

What nobody has mentioned is one needs to make sure the dimple is very crisp, to minimize the radius trying to contact the countersink top edge. For critical items, like the tank dimples to the spar, i broke the edge of the countersink with some 220 grit sanpaper to help this meshing of the dimple and countersink; but this is not required.

I would be leery of thinning the material at its highest stress point. A few dollars invested in a high quality set of springback dies is money well spent. The thickness is not reduced and the fit is better.

3/32 in. skin dies (rivet sits flush with skin)
3/32 in. substructure dies (0.011 in. deeper than skin die - use on interior parts that mate to exterior parts e.g. skin)
3/32 in. fuel tank skin dies (0.007 in. deeper than skin die - allows for sealant under rivet so you don't have to shave the head flush)

There is a video on these pages that go into detail about each of these dies. I haven't used the tank dies yet, but the substructure dies do result in a better fit.

[bobnoffs]

if you made a countersink that is just too deep i know you can dimple a thin material [.016-.020 ] and then cut, beltsand around that dimple until you have a ''cup'' that will nestle between the countersunk hole and the bottom of the dimple. a touch of superglue holds it in place until the rivet is set. works well and is easy. i would never be so inept as to over countersink a hole, i just heard about this somewhere.

[TASEsq]

Originally Posted by StressedOut

A few dollars invested in a high quality set of springback dies is money well spent. The thickness is not reduced and the fit is better.

Thanks Art - those are the ones I?ve been using, and I agree, you can tell the difference when you use the SS dies on the ribs.

One other poster on another forum made a comment that no dimple should be wider than the female dimple die! Makes sense to me, but I would still like some
guidance on how much gap is acceptable around the skin either side of the dimple. It?s surprising to a newbie - so much of this is scientific, then something like this is like using an uncalibrated hammer. I get where vans is going with the 0.007? but section 5 doesn?t really give any indications on what gap is appropriate either.

[rvbuilder2002]

Originally Posted by TASEsq

Makes sense to me, but I would still like some
guidance on how much gap is acceptable around the skin either side of the dimple. It’s surprising to a newbie - so much of this is scientific, then something like this is like using an uncalibrated hammer. I get where vans is going with the 0.007” but section 5 doesn’t really give any indications on what gap is appropriate either.

Originally Posted by rvbuilder2002

That is typical of all inexperienced RV builders and why Section 5 makes a very specific recommendation.

This recommendation is based on testing by machining cross sections of rivet joints so they can be viewed and analyzed under high power magnification.

This is from a post about a year ago......

Edit: Here is another thread that discusses the subject (which it appears you (the OP) has commented in so you likely have read all of this already. http://www.vansairforce.com/communit...d.php?t=157662

Use caution......

One of the more common critical mistakes that we have to help builders through is when they countersink too deep for dimpled skins.

If you go too deep, you will have no gap between the skin and the sub structure but the strength is likely compromised, With countersinks that are slightly too shallow for the skin to lay entirely flush (as in the attached photo) the strength is no different, but there is assurance that it is not too deep. That is the reason for specific recommendations in Section 5 of the manual. The depth was derived from doing samples of different depths and then machining off 50% to allow magnified inspection of the rivet joints.
Countersinks only deep enough for a rivet head still produced totally acceptable riveted joints (though not quite as nice ascetically). Countersinks any deep than about .009 " beyond what would be needed for a flush rivet began to produce visible gaps in the riveted joint.

In a nut shell, the skin adjacent to a dimple can not lay entirely net to the surface below without making the countersink too deep. This is because the outer edge of the countersink has a sharp corner on it, but the angle at the outer edge of the dimple does not.

With proper technique used to form the dimple and install the rivets, from the outside, a rivet where the counter sink was done only 7 clicks deep vs one that was deep enough for the skin to lay 100% flush, the finished appearance will be indistinguishable but the over countersunk one will have a much higher possibility of problems in the future and is not as strong of a joint.

Edit: Here is another thread that discusses the subject which it appears you (the OP) has commented in so you likely have read all of this already. http://www.vansairforce.com/communit...d.php?t=157662

Repeating my previous post...... it is still a valid answer to your question. I have highlighted a few key points. Bottom line is you can not have the skin lay flush with no gap, and have the sides of the dimple fit tightly within the countersink. The tight fit is much more important. The slight gap you see will not have any effect on structural strength or appearance quality. Van's doesn't specify an allowable gap because if you follow the recommendation of Section 5, whatever gap there is, is considered acceptable.

[StressedOut]

Originally Posted by TASEsq

Thanks Art - those are the ones I?ve been using, and I agree, you can tell the difference when you use the SS dies on the ribs.

One other poster on another forum made a comment that no dimple should be wider than the female dimple die! Makes sense to me, but I would still like some
guidance on how much gap is acceptable around the skin either side of the dimple. It?s surprising to a newbie - so much of this is scientific, then something like this is like using an uncalibrated hammer. I get where vans is going with the 0.007? but section 5 doesn?t really give any indications on what gap is appropriate either.

Trent,

I know what you're thinking and feeling regarding the dimples. I went through exactly the same thing a few months ago when I started my empennage kit. I was worried about every little thing and concerned that if I screwed up I'd eventually have a massive in-flight break up and end up a statistic. Eh, I got over it.

The bottom line is that there is plenty of safety margin built into these planes, and if you follow the guidelines in section 5 you will end up with a very good airplane.

Remember, "perfect" is the enemy of "acceptable". Unless you're building a show plane you should be shooting for "acceptable".

P.S. There are a couple of entries in my construction log on this very topic. Feel free to have a look.

[TASEsq]

Thanks Gents for both your feedback.

Seems sticking with the Vans recommendations is the way most people go of course, it just doesn?t stick well with me that 0.007? is correct for all skin thicknesses. Anyway..

Originally Posted by StressedOut

Remember, "perfect" is the enemy of "acceptable". Unless you're building a show plane you should be shooting for "acceptable".

P.S. There are a couple of entries in my construction log on this very topic. Feel free to have a look.

Fair enough! Thanks. I?ve looked at your log a few times - especially your paint booth. Re the CSK, have you tried an electric drill? I?ve tried both, and it is easy to get too fast with the air drill and it starts to screech. I?ve found the electric drill is better, and is mostly what I?ve read people use. Heaps easier to go lightly and slowly than the air drill, but harder to keep normal to the surface than an air drill as it?s heavier.

Originally Posted by rvbuilder2002

Repeating my previous post...... it is still a valid answer to your question. I have highlighted a few key points. Bottom line is you can not have the skin lay flush with no gap, and have the sides of the dimple fit tightly within the countersink. The tight fit is much more important. The slight gap you see will not have any effect on structural strength or appearance quality. Van's doesn't specify an allowable gap because if you follow the recommendation of Section 5, whatever gap there is, is considered acceptable.

Thank you Scott. I appreciate the time you?ve taken and I do take your points and understand the reason why vans uses the 0.007?. Thanks again. I?m just a ?why does it work that way?, or ?why do they ask us to do it that way? type of guy. Better to have these questions answered now on the practice kit than go too deep on a real part - the shipping down here is a bloody killer.

I think overall this is a good experience and something I could do. Almost got the minister for war and finances convinced we should have an emp send to Aus!

[StressedOut]

Originally Posted by TASEsq

Thanks Gents for both your feedback.

Seems sticking with the Vans recommendations is the way most people go of course, it just doesn’t stick well with me that 0.007” is correct for all skin thicknesses. Anyway..

...

Your instincts are correct. Using a 0.007 in. additional depth is not perfect for all skin thicknesses. If you were to model the dimples using Catia or SolidWorks you could create an assembly showing this.

However, the differences are very small for the range of skin thickness used in RV's. It's a little like picking the fly poop out of the pepper. Recommending various depths for different combinations of thicknesses will get very complicated very quickly. In the end it's better to use the recommended 0.007 in. additional depth because it's simpler and gives acceptable structural performance.

Just for fun, here's a link to a WW2 training film about riveting. Go to the 4:13 mark to see how they do dimpling in two thin sheets. Their approach is a little different than what we use today.

[TASEsq]

Thanks very much guys.. I had a response from Van's (which is impressive given Sun and Fun is on!)..

You guys were spot on:
Use the directions in Section 5 to avoid countersinking too deep.
There may well be a gap ? the idea is to make sure the dimpled skin and countersunk part are locked together. A countersink that is to deep will prevent that, and though there is no gap between the parts, the joint is not so strong.