What's new
Van's Air Force

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

Basic building skill question

P

PipeDreamer

Member
I have a question that's bugging me a lot and I tried googling it and searching to no avail perhaps because the right terminology is eluding me.
The questionis : when the skin is riveted and its surface looks warped or distorted around the rivet whats is it called ?
I want to learn how to achieve commercial quality finished surface. A lot of good examples here, so obviously it is doable but what is it called?
Would appreciate references, links, etc.
 
I don't know of a specific name for it, but assume like a concave area centered on the rivet? If there is some picture of one it will help all identify the issue and propose corrective actions.

Sometimes it is due to the underlying part as it is below grade and may need a shim. It may be an over driven rivet, air pressure too high or driven too long with too little bucking. There are little circles created with dimpling tools, but mostly cosmetic.

Post a good picture and stand back, the information flow will be unleashed.
 
If it's actual damage to the skin, from the rivet set, for instance, most people call the damage a 'smiley', because the mushroom set will leave a 'smile' shaped arc if you let the gun get too far from perpendicular.

If you're talking about a properly set rivet (in a dimpled hole), but you can still see what appears as a slight concave appearance around the rivet, I don't have a name for you, but it's caused by not creating a crisp dimple before riveting. Many people fear damaging the skin, and under-dimple. Several will no doubt recommend a multi-hundred dollar tool to cure it (which works great if you don't mind spending the money), but you can get the same results with a simple C-frame & a rubber mallet. Just hit the die holder harder. To expand your comfort zone, play with some scrap & hit the die holder with a fat rubber mallet like you were driving a big framing nail. (I usually strike one medium force blow, which forms the dimple maybe 3/4 of the way, then hit it again, hard.) I'll bet you can't actually damage the skin using a rubber mallet on the die holder. Goal is a sharp transition from dimple to surrounding skin. If properly dimpled, you shouldn't be able to catch a fingernail on the edge of the head after riveting. If under-dimpled, you can.

I think Cleveland Tool has a youtube video showing how to detect under-dimpling.

Hope that helps,

Charlie
 
Deflection

I believe "deflection" is the term. It can also be from under dimpling
the skin. A good wack on a solid surface with the C frame dimpler and a good quality dimple die will make for flat skins. 35lbs with a 3 x gun is enough pressure for 3/32 rivets. Go lightly with the rivet gun. It doesn't take much to get the correct shop head.. If you dimple with the pneumatic squeezer, it often fails to make crisp, non-deflected dimples in the skin.
 
None of it seems to be relevant to riveting itself, it is more of distortion of skin where it is riveted to whats under, warping sized from a penny to silver dollar size or more. It is what makes home build look like a home build. Perhaps it is what can be covered with bondo, but it sure will take a lot of bondo to fill all of it when its bad.
 
Example

2re5bq0.jpg


This is a minor case obviously. Is there a definition for it?
 
A friend of mine who is helping me finish my 9A has a beautiful 4 and 6. His wings look like glass. He used 3m epoxy to glue the skin to the ribs, cleco them together, let it dry, then rivet. I used the same technique on the access panels on the foreskin of my plane and it worked well.

If I was building again, I would epoxy the skins on the tops of the wings to keep them smooth. I may add a few ounces, but not more than skipping breakfast before you fly. I would not do the underside of the wings - if anyone crawls under the wings to check I am OK with a few dimples there.
 
I don't know what it is called either, but it is caused by being too gentle when dimpling. My first rudder skin was like that. I posted it here and someone told me what the problem is and what the proper results look like, so I redid the part and all is good.
 
PipeDreamer said:
2re5bq0.jpg


This is a minor case obviously. Is there a definition for it?
Click to expand...

Hard to tell what your picture is showing - longeron? If so, that should be fairly easy to get them all looking like the rivets to the left of the ugly ones that are circled. Like the others have said - either under dimpling or a shim needed. My guess is under dimpling. I've seen examples of folks that had no clue they were under dimpling - looked like they riveted using a ball-peen hammer. Not pretty. I used a c-frame for my entire airframe - even though I own a DRDT - and I wasn't very gentle with it. Most of my rivet lines are very flat. I'm a fan of the c-frame and rubber mushroom set for solo riveting.
 
Last edited:
I think it can also be caused if you pound the rivet with too much pressure on rivet gun, or not having the gun centered on the rivet
 
Shimming is good for an odd case or two. There is fundamental failure of some sorts in the last picture. Where does one starts getting ideas what is right? How do you know that skin will lay flat on these multiple planar surfaces?
 
Repost- these look like puckers from not having the skin perfectly flat on the spar before final drilling and then riveting as the spar should have been quite flat.

On the contoured depressions, likely just and underlying gap. Some areas are unavoidable.

Hopefully some repeat offenders will weigh in.

I had no experience with under dimlping - sorry. That could be it on flat surfaces too. However, I often used dimple dies in finished/dimple assemblies to ensure they parts are fitting together. I actually squeezed every hole with dimple dies before every rivet on my firewall to get it tight and flat. There were 4 layers of material that had to match. firewall,skin, spacer, skyloc.
 
Last edited:
PipeDreamer said:
Shimming is good for an odd case or two. There is fundamental failure of some sorts in the last picture. Where does one starts getting ideas what is right? How do you know that skin will lay flat on these multiple planar surfaces?
Click to expand...

Clecos in every hole, and pay attention to order of installing them. No set rule for that. How good are you at drilling out rivets?
 
Last edited:
rockwoodrv9 said:
A friend of mine who is helping me finish my 9A has a beautiful 4 and 6. His wings look like glass. He used 3m epoxy to glue the skin to the ribs, cleco them together, let it dry, then rivet. I used the same technique on the access panels on the foreskin of my plane and it worked well.
.
Click to expand...

I wonder if there is a specification on this method, is it reliable?
 
If done wrong (easier than doing right), it can actually weaken the joint. Reason is that the epoxy can hold the skins apart, which loads the rivets in bending instead of shear.

The pics look like simple under-dimpling, to me. Simple test is to dimple a couple of scraps using the same technique, rivet, and compare to the problem piece. Then do the same process, but *try* to over-dimple, and compare.
 
Please reconsider this procedure!!!

rockwoodrv9 said:
A friend of mine who is helping me finish my 9A has a beautiful 4 and 6. His wings look like glass. He used 3m epoxy to glue the skin to the ribs, cleco them together, let it dry, then rivet. I used the same technique on the access panels on the foreskin of my plane and it worked well.
If I was building again, I would epoxy the skins on the tops of the wings to keep them smooth. I may add a few ounces, but not more than skipping breakfast before you fly. I would not do the underside of the wings - if anyone crawls under the wings to check I am OK with a few dimples there.
Click to expand...

I would NEVER recommend this procedure! Aluminum skin should be riveted directly to the superstructure. What happens when the brittle epoxy cracks and falls out leaving the skin loose from the ribs/spar/etc.?
 
Mel said:
I would NEVER recommend this procedure! Aluminum skin should be riveted directly to the superstructure. What happens when the brittle epoxy cracks and falls out leaving the skin loose from the ribs/spar/etc.?
Click to expand...

There is no space on his rib - skin connection. He is a perfectionist. His ribs are exact and the 3m epoxy is structural. His 4 is many years old and so is his 6. They both have the best looking wings of any I have seen - no smoking rivets, no paint issues, just nice smooth wings.

I can see if you are using the epoxy to fill gaps or fix ribs that are not straight, you could surly have an issue. It is not a way to fix poor building skills.
 
Under dimpling

rv7charlie said:
If done wrong (easier than doing right), it can actually weaken the joint. Reason is that the epoxy can hold the skins apart, which loads the rivets in bending instead of shear.

The pics look like simple under-dimpling, to me. Simple test is to dimple a couple of scraps using the same technique, rivet, and compare to the problem piece. Then do the same process, but *try* to over-dimple, and compare.
Click to expand...

+1
Looks under dimpled.
 
PipeDreamer said:
I wonder if there is a specification on this method, is it reliable?
Click to expand...

Yes there is; a friend of mine wrote one. His airplanes, both the one he built earlier and now flies and the one he's building, have absolutely smooth skins and the rivets are exactly flush with the skins. They are perfect. It's also a very time-consuming way to build.

His process rather long and detailed (and I don't have permission to share it, sorry). There are a lot of intermediate steps which he learned about through experience. And before we embark on that, the underlying structure must be at the same height so that the skin lies smooth on it with no gaps. On my RV-3B fuselage, I'm gluing down shims before I even think about drilling or riveting or gluing and riveting. He limits the use of the glue as a shim to .015 inches gap, by the way. Anything more is shimmed with aluminum.

The use of glue to hold shims in place is relatively straightforward. Once I've determined how thick the shim has to be and made it, I clean and abrade the aluminum by scrubbing it with Ajax cleanser in water, using a maroon Scotchbrite pad, until the aluminum is a consistent gray. This is followed by a water rinse and within two hours, the bond. The two hour limit is to avoid the effects of aluminum oxide build-up. He uses 3M 2216 and fills it with aluminum powder. I'm using the thickened version of West System's G-flex and haven't been using the aluminum powder for shim bonding (I will if I need to use it for anything else).

Dave
RV-3B, now skinning the fuselage
 
Regarding the depressions at the rivets, on my first horizontal stabilizer, it ended up looking like that. I'd used a too-narrow rivet set at too high a pressure. Before driving the rivets, that one had really looked good, with properly-dimpled rivet holes.

I scrapped that stabilizer, by the way, after finding a better riveting gun, better sets, and learning a bit.

jion5i.jpg


Look at all the rivets, especially at the top middle of the image.

You can read about it here.

And you can read about the tools in this thread.

Dave
RV-3B, still starting to skin the fuselage
 
Pucker

It takes a little time and a lot of practice to get smooth flat finishes of the skin and joining peace under it. Be it dimple or counter sink. Mel is correct and the small parts of the tail or a practice kit are the place to start. Just start slow, make gages and match your dimples to the ones above in the skin. It will get easier as you get further along and have more practice. If you need it get a tech to look over your shoulder. There are around 15,000 rivets in a two place RV, by the time you drive 1 or 2 thousand you should get it.
Hope this helps, Yours, R.E.A. III #80888
 
rockwoodrv9 said:
A friend of mine who is helping me finish my 9A has a beautiful 4 and 6. His wings look like glass. He used 3m epoxy to glue the skin to the ribs, cleco them together, let it dry, then rivet. I used the same technique on the access panels on the foreskin of my plane and it worked well.
Click to expand...

How is your friend going to remove the skins from the ribs in the event that there is some wing damage to be repaired in the future. As it stands, even quite minor damage might necessitate the complete replacement of the wing if the skin is adhered to the ribs.

The best way for builders to get superior riveting results is to practice every technique on scrap, or on a practice kit, before hopping into the plane. NEVER teach yourself to rivet on the actual aircraft. For example, if you're halfway through the build and you decide to try some back-riveting for the first time....hone your skills on some scrap.

Plus, I agree that insufficient dimpling produces B grade flush rivets. The squeezing tools do not produce a properly defined cross section. To achieve satisfactory definition you need to form the dimple with a dynamic force (impact) rather than a static force (squeezing).
 
Captain Avgas said:
How is your friend going to remove the skins from the ribs in the event that there is some wing damage to be repaired in the future. As it stands, even quite minor damage might necessitate the complete replacement of the wing if the skin is adhered to the ribs.
Click to expand...

Probably would.
 
PipeDreamer said:
This a "good" example
34znc48.jpg
Click to expand...


This picture looks like the same problem I had when first starting. The pressure on the gun was too high.

It was driving the surface and underlying structure down leaving a dent no matter how hard I pressed up on the bucking bar. With the gun at the right pressure, you should be able to hold the bucking bar with steady almost gentle force to the rivet bottom with the gun firmly held on top.

I was also using a set that had a hemispherical face on it, thinking it would better set the rivet head into the dimple. It didn't leave a smiley, but a dent. I changed to a flat faced set and turned the gun pressure down, and the problem went away.
 
Bob
I have seen some great planes finished and flying built with drd2s or squeezers used to dimple. I'm a newbie and looked at all the different types. Seen some older planes with this issue that was done with a c-frame.
I think it comes down to builder error in not setting up their tools properly.
Be it the dimpling process or how they rivet.

Always great info on VAF
 
I concur with the other posters about proper tool set up and technique. IMO there should be no need to shim, glue, or use any other esoteric method in order to avoid dents, divots, deflections or whatever you want to call it. I know for a fact you can achieve glass-like surfaces without having to epoxy components together because I've done it and I'm a first time builder. Here's a pic of my top left wing getting prepped for paint--see what I mean?
34178115566_9c5a764505_c.jpg
 
Last edited:
I don't think the problem is caused by a dimpling issue but rather by applying too much pressure to the rivet gun. The pressure I am referring to is not "air pressure" but the amount of push pressure exerted on the rivet gun and thereby denting the the skin and underlying structure when setting the rivet.
Too much air pressure generally results in over driven flat shop heads and inadequate dimpling either leaves a gap between the skins or an elevated rivet head.
Rather than looking at "exotic" and time consuming builder practices such as epoxying the skins or spending hundreds of dollars on fancy dimpling equipment perhaps you simply need to improve a few simple steps to riveting
as outlined in Van's red instruction booklet that came with your kit.
A nearby builder who can point you in the right direction would be my preferred choice in addition to reviewing the dimple/rivet process.
Lastly, any row of rivets that can accessed with a hand squeezer such as those along the trailing edge of the rudder should be squeezed. I would be impossiple to create rivets like those in the picture by using a hand squeezer:)
 
Neal Trombley said:
Bob
I have seen some great planes finished and flying built with drd2s or squeezers used to dimple. I'm a newbie and looked at all the different types. Seen some older planes with this issue that was done with a c-frame.
I think it comes down to builder error in not setting up their tools properly.
Be it the dimpling process or how they rivet.

Always great info on VAF
Click to expand...

Hi Neal, here is a URL to a 2012 thread on VansAirforce that dealt with this very problem:
http://www.vansairforce.com/community/showthread.php?t=91158&highlight=forming+dimples
Check out my post #11 in that thread that explains my thoughts on the matter.
I was a metal aircraft judge for the Sport Aircraft Association of Australia for 7 years straight and I have inspected literally 100s of RVs. I think I am probably in a good position to know what results in excellent deep-seated flush rivets and what does not.

It really depends on the final quality you wish to produce. Squeezed dimples are OK for most purposes and will produce a result that is OK to go, but not necessarily to show (as in a show plane). Squeezed dimples always have rounded shoulders as opposed to impact-formed dimples which should exhibit crisp changes of angle (well defined discontinuities). It is the well defined dimple that allows the rivet to seat properly. I recommend that you simply do some comparative testing on scrap and you will quickly see my point.
 
Last edited:
You guys are missing a technique to fix a "factor " only a few pointed out.

You have to get some movement of the skin in order to drive the rivet. This will often lead to a depression like witnessed in many of the photos.

First thing with panel distortion are using rivets that are too hard. Forming the shop head increases the shaft diameter to the extent that it distorts the skin.

But you can still get the depressions when using softer rivets, The fix - taught to my by "old timers" from Lockheed, Northrup and North American who started during WWII, was to take a "fresh" or soft rubber mallet, back the shop head with a bucking bar, and starting light, working up in force, tap the outside surface of the skin and "draw" up the depression from driving the rivet. As always, keep things clean!

Boeing Engineer, Will Taylor, son of Don Taylor, wrote extensively about gluing for structure. First thing is the assembly HAS to be designed for this approach, almost all homebuilt designs are not, and, Second, the quality control/methods and equipment necessary to obtain the benefit are not obtainable at the typical homebuilder's budget. Without either of these points, the structure will be less than can be obtained by standard assembly techniques - i.e., riveting w/o adhesive.

Before you register your objection, there are many areas of a design that are "overbuilt" for expected air loads because experience has taught the designer that the 'risky" loads come from ground handling, etc. so, riveted trailing edges and the like may not see loads that bring the compromised strength into issue.
 
I also think this is from to much hand pressure on the rivet gun, I riveted my tanks and found a heavy pressure on the gun worked great but then moved to the leading edges and still using a heavier than normal pressure on the gun I got the dimples pictured for the first time. They were not as bad as the pictures for sure so the dimple could definitely play a part also.
 
Couple more thoughts, triggered by recent posts.

1st. Have you done test samples, as described earlier? We're all wasting our time if you haven't.

If you're actually denting the skins, you should be able to see the dents in the underlying structure, especially on a trailing edge flange. Do you?

Assuming that you're actually denting the skins, on what surface are you placing the assembly while riveting? The gun is going to deflect whatever it can. If you're putting it on a solid work bench, then it's likely you will bend something when the set hits the skin because the structure can't move with the blow. If you've been riveting on a solid workbench, try putting some type of pad under the structure. An old scrap of rubber carpet pad works well for me, because it allows vertical movement and helps prevent skidding sideways. Any kind of padding will help, though; folded towels, blankets, etc.

Last, have you done those test samples yet?
 
You must log in or register to reply here.
Back
Top