How to swage a hole for welding in an aluminum fitting on a fuel tank?

Dear All,
I am building an aluminum fuel tank out of 5052 H32 .090" that has aluminum fittings welded into swaged holes for finger strainers,external fuel gauge
and filler neck. The holes will have a finished inside diameter of 1-7/8", 2-5/16" and 3" . There is a total of six swaged holes I need to make. I have received one method on rec.aviation.homebuilt but would like to have a back-up in the event that method doesn't work. Looking for a suggested method with handtools or inexpensive set-up equipment. Last resort would be to find a company to do the swaging but there I am at a lack of knowledge too. Anyone out there had experience with this? Please advise.
Ebby
--
John "Ebby" Ebensperger
Hatz Classic s/n37
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I do not understand the need for swaging. Why not cut a hole with holesaw slightly smaller than the bung or plate or fitting and weld it onto the surface? If the fitting must protrude inside the tank for minimum external exposure, just drill the hole slightly undersize and ream or file to a snug fit. I do just that on aluminum fuel tanks all the time.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
The plans show a swaged hole with the swage towards the outside of the tank. The bung has a flat surface on one side which is supposed to be inside the tank flush with the tank bottom. The flange on the bung is outside the tank intended to be edge welded on the outside. If installed the other way there would be about a half in of fuel in the tank that is unusable.
The tank will go into the center section of a wing with finish covers of spun aluminum over the bung. The fuel lines run from the bung to the gascolator. I can scan a section of the plan where this detail is drawn as my description may be unclear. It sure would be less work to just punch hole and weld in the bungs so long as it was as good as the swaged method. I have a good amount of the aluminum to experiment with and I am not in a rush.
Ebby.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I have always seen these swedges made using a circular die on the outside, and a drawbar that goes through the hole in the tank with a polished ball on the end. The drawbar is pulled through the die with either a screwjack or a hydraulic ram.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Ernie,
Is the ball end pressed through all the way or does it stop and then the hole is machined leaving the swedged flange. I am thinking of what a piece of metal looks like after a bullet passes through. There is a "round" hole and a ragged flange on the exit side. I'm guessing the ball end does not pass all the way through the metal.
Ebby

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I have always seen it pulled completely through. Then the drawn edge is cleaned up. This is exactly how a lot of brewing tanks are made.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
In your experience is there any advantage to a swedged hole versus simply making the hole slightly smaller than the fitting and welding from the outside. I will send you a diagram of what I mean directly to you.
Ebby

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Better flow and a weld that is easier to back purge when welding.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
This is a test. Having trouble posting.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I know of no single-engine aircraft that use swedged holes for fittings, which causes me to wonder why it is called for in your plans? Also the .090" thick material is way heavier than "normal", say .040-.050"? I hope you have a good sump with drain in your tank for contaminants, as bad fuel causes most engine failures here in Alaska. I do aircraft welding, ie. tube frame, fuel, and exhaust systems, and have experience with this topic. Still just my $.02.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
If anyone is interested, I worked out a solution to the swage problem. Thanks to input from three different newsgroups I successfully am now able to swage holes for my aluminum fuel tank. To view the results go to the metalworking.com dropbox and look for the files called "SwageDie". I posted them earlier today.
Thanks for the help.
Ebby

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
It's normally done with a male and female dieset. Drill or hole saw a hole 1/8" to 1/2" smaller than the desired finished hole, exact undersize is a function of the swedge depth. The longer the swedge, the more chance of splitting the edge. For just a few holes, the dies can be made of hardwood or aluminum. For a few dozen use steel, hardened tool steel is best. The female die has a rounded edge, start of the edge should be the finished inside hole diameter plus 2x or 3x the material thickness (each side) A little expeimentation is needed to get the right combination of clearance and shape, depends on the hardness or softness of material. Male die is tapered. If the male die is not steel, it can be a bit troublesome to get the die removed from the finished piece. For one off, I'd just use some green ash turned in a wood lathe, sanded down smooth, and pounded through the base metal.
Ebby wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
First of all now I know it's swedge not swage. Your method is just about what I thought when I was trying to engineer this in my head. I have an old blacksmith book which refers to swedge blocks with tapered make dies. Your mention of a tapered male end reminded me of what I researched. Also the size of the hole is a function of the swedge depth. That's very helpful. I have a 4' x 10' sheet and will only need about half of it for the tank so I've got lots of material for R&D. Thanks.
If anyone has other tips or techniques for this process chime in. This is all very educational.
Ebby

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
It would be helpful for you to really understand what minimum dimensions you really need, how accurate it has to be, what radius you need, etc. If you are just trying to get a small 1/8" deep recess, something as simple as a block of wood with suitable hole and a ballpein hammer might be close enough. Do a small lip, hit the ragged edge with a belt sander, insert the fitting, weld it up. It sounds crude but light tapping moves the metal nicely. If you needed a full 1/2", I suspect a full set of dies is in your future. Ditto for the quality and size of the radius.
Ebby wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
RoyJ wrote:

Lautard's "The Machinists Third Bedside Reader". Page 80 has a good description of a jig used to pull tees out of copper pipe.
Stuart
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
The EAA states that building your own airplane is educational. Well I'll say!! I spent the day down in my shop working on the swedging problem. I am pretty sure I have the answer to what needs to be done. As the alloy is 5052 and .090" in thickness, it seems to hve good workability.
I ended up making a device very similar to a regular Greenlee chassis punch except instead of cutting a sharp edged hole, the male portion of the die has relieved edges and has a diameter equal to the slug I want to weld into the aluminum. Before I use the device, I cut a hole in the aluminum with a bi-metallic hole saw slightly smaller than the finished hole size. It is smaller by twice the length of the swedge flange. A 1/2 inch bolt goes through the female die (oak plywood block nailed to another plywood block which has been drilled to finished hole size plus twice the thickness of the metal plus twice the bend radius for .090" aluminum) then through the aluminun sheet then the male die. A large washer is placed over the male die then the nut goes on. I spray WD40 on the threads and start to torque the nut. I was surprised at how nicely the aluminum deflects downward. I continue to wind the nut down until the male die is pulled through the piece of aluminum. Originally I was using oak as the male die but it didn't hold up. Then I switched to UHDPE. The PE went through fine but compresses slightly so the hole ends up undersized. I ordered a piece of 3" steel bar which I will turn down on a lathe to make the male dies. At this point I think I have the problem solved. A lot of work for five holes. But as I started my message it's all educational and other builders may benefit.
Ebby

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.