All MetalShaping

Go Back   All MetalShaping > General Metal Shaping Discussion > Welding Sheet Metal
  Today's Posts Posts for Last 7 Days Posts for Last 14 Days  

Reply
 
Thread Tools Search this Thread Display Modes
  #11  
Old 12-30-2019, 11:38 AM
BTromblay BTromblay is offline
MetalShaper of the Month August 2018
 
Join Date: Oct 2015
Location: Mukwonago, Wi
Posts: 356
Default

Quote:
Originally Posted by Jerry Jackson View Post
Sean

Let us know how the baffles work out. I tried to install baffles using 5356 filler on .05 inch 5052 tank material. I was very disappointed in the amount of distortion caused by welding on the flat tank surface. I welcome any suggestions on how to control that distortion.

Jerry Jackson
Hi,

I made a smoke oil tank for a North American F-86, this may held answer some questions, here is a link. http://allmetalshaping.com/showthread.php?t=18066

To control distortion from welding in a flat section of the tank can be done with some pre-planing. A flat sheet of material has relative low stress from the cold rolling process. The stress is uniform and that is why the panel stays flat. Welding of the edges or rosette welds to hold in bulkheads causes higher stress concentrations from several factors including, material shrinkage from the weld, re-crystallization from exceeding the anneal temp and internal material changes to crystallized structure from the high heat of welding.

It is important to isolate the high stress area of the weld, from the low stress area of the flat panel. This is done by tipping all edges several degrees 3/16" - 1/4" in from the edge. Dimple all holes that is the same size or slightly larger than your weld. The broke edge creates a mechanical lock and defines the edge from high stress to low stress. The weld is large enough that it covers the broke edge so it is not seen once welded. Because a steel weld is smaller than an aluminum weld, plan the broke edge accordingly.

Flange welded tanks have been around for many years with millions o flight hours flown on this seam design. With that being said, it is possible to have them fail due, more times than not due to bad welding. Lack of penetration, oxide development and filler rod selection all play a roll in the failure. Pressure testing should always done, once the tank is made. I fill all of my tanks with water, then use compressed air to 5 psi.

I'm welding up a tank right now, will see i I can post some pictures in the next few days.

Bill
__________________
Bill Tromblay

"A sign of a good machinist, is one who can fix his F$@& Ups" My mentor and friend, Gil Zietz Micro Metric Machine.
Reply With Quote
  #12  
Old 12-30-2019, 01:22 PM
sblack sblack is offline
Member
 
Join Date: Sep 2011
Location: Montreal
Posts: 371
Default

I recall kne of the tanks on an old classic sports car had a lot of detail embossed on the flatter section to stiffen it. Sounds similar to what you are talking about although this tank was riveted and soldered. It was steel. But large flat sections are probably not desirable.

https://www.tinmantech.com/assets/im...rdFuelTank.JPG
__________________
Scott in Montreal

Last edited by sblack; 12-30-2019 at 01:26 PM.
Reply With Quote
  #13  
Old 12-30-2019, 01:52 PM
Marc Bourget Marc Bourget is offline
Member
 
Join Date: May 2013
Location: North Ca
Posts: 417
Default

I suspect it wasn't the flanges, but the welding that resulted in the problem. There doesn't seem to be enough load on the seams to result in failure when running up on the ground.


John Thorp was the Lockheed connection I mentioned previously. His tank designs were butt welded. Another friend, a top level auto engineer with extensive tank design experience felt flange welds were fine. all else being equal.


FWIW
__________________
Marc
Reply With Quote
  #14  
Old 12-30-2019, 03:15 PM
crystallographic crystallographic is online now
MetalShaper of the Month October '14 & April '16
 
Join Date: Dec 2010
Location: Western Sierra Nevadas, Badger Hill, CA
Posts: 3,446
Default

Quote:
Originally Posted by sblack View Post
A friend bought a fuel tank for his bushby mustang II homebuilt airplane from the designer. It was 6061t6, probably .040 or 0.050. It was made in the traditional way with the ends pounded over a form block and the skins wrapped around and flange welded. On that airplane the tank sits on the fwd fuselage behind the panel, so basically in your lap. He finished the airplane and was doing initial engine runs. He had the tank full. After a few minutes of running the tank seams opened up completely, essentially the tank burst and 30 gal of av gas was dumped on his lap, with the engine running and electric system powered. The fuel started draining out the drain holes in the belly, right by the hot exhaust. He pulled the mixture and jumped out, waiting for 12 yrs of work to go up in flames. Luckily it did not. It was a huge mess.

He had a new tank locally made and the welder said that flange welds are no good for tanks and did butt joints instead. I have read all of Kentís articles on tank construction and I know that flange welds have been widely used for decades on all sorts of containers. So I suspect that there was a problem with the welding or the tank design i.e. not stiff enough etc. Strange as the bushby mustang has been around for eons so you would think the design was proven.

So this us a bit off topic but an interesting story and it reminds us that you really want a fuel tank to be well built.

Hi Scott,
The problem with that tank is not the design. That design is proven by the tens of thousands of airplanes carrying fuel inside it, for decades, in rough environments. Had there been an issue with that design it would have been subject to an Airworthiness Directive, and the entire industry would have changed that design.
The problem is in the execution of that design, by the craftsman building that tank.
Example from real world aviation:

I have two wing tanks here that were made by a "pro fab/welder" shop.
.040 5052 with flanged ends that were fusion welded with a tig/TIG.
Made to "copy" an original 1940's existing pair of tanks.

The cracks in these end seams are not longer than 14 inches, and the fuel also splashed all over a hot engine and pilot, as he got down as fast as possible, and got it all shut off in time to exit safely and the plane was saved.
Problem: no baffles to support the long distances between end plates, so under positive and negative G's the seams failed, opposite each other and at the same end panel.

P1030144 copy.jpg
P1030145 copy.jpg
It does help to have some familiarity with the various aviation tank designs so as to make clear and accurate conclusions.
__________________
Kent

http://www.tinmantech.com

"All it takes is a little practical experience to blow the he!! out of a perfectly good theory." --- Lloyd Rosenquist, charter member AWS, 1919.
Reply With Quote
  #15  
Old 12-30-2019, 03:31 PM
crystallographic crystallographic is online now
MetalShaper of the Month October '14 & April '16
 
Join Date: Dec 2010
Location: Western Sierra Nevadas, Badger Hill, CA
Posts: 3,446
Default

Quote:
Originally Posted by Sean in CT View Post
I ended up with some oil canning on the bottom after i welded the filler "neck" on the side. I am going to flange the baffles to make it all more rigid. Thinking of drilling holes and plug welding them.


I think i would have needed to roll some beads or other shape into the bottom to reduce deformation. I used aluminum blocks as heat sinks whic i think helped. Also used the Fireball Tool Mega Squares to fixture - they work great!

Hi Sean,
Here's one proven baffle install design, no distortion, no big heat sinks, no fancy clamping ....

:
IMG_0111.jpg

70_brazing the rivet heads.jpg
The olde bead roller makes twin beads parallel to each other, downwards, trapping the baffles in place, allowing for easy riveting, and then distortion-free brazing of the rivet heads. (pssst - brazing does not melt the parent metals to join them together - so there is zero "solidus" or "shrinking" as the molten metal cools back down).
Brazing the aluminum rivet heads onto aluminum tanks is very common in aviation tank methodology, and has been since about 1939, according to ALCOA.
__________________
Kent

http://www.tinmantech.com

"All it takes is a little practical experience to blow the he!! out of a perfectly good theory." --- Lloyd Rosenquist, charter member AWS, 1919.
Reply With Quote
  #16  
Old 12-30-2019, 04:57 PM
Sean in CT Sean in CT is offline
Member
 
Join Date: Jan 2017
Location: North Stonington, CT
Posts: 25
Default

Wow Kent. Thanks! Have to make a supercharger coolant tank for under the hood. Iíll definitely copy that design.
__________________
Sean
Reply With Quote
  #17  
Old 01-04-2020, 08:55 AM
sandmanred sandmanred is offline
MetalShaper of the Month July 2017
 
Join Date: Nov 2015
Location: Saint Paul, MN
Posts: 175
Default

Here's a table of weld properties based on the two base alloys and the weld properties you most value. It's best to print it out and tape it together. You pick your base alloys off the edges and the intersection of the base metals provide codes for weld properties for most common fillers. For a tank I'd usually pick 5356 for 6061/6061 due to the high ductility of the weld. I would expect 1100 to be very ductile too but not sure about the rest of the weld properties.
Attached Files
File Type: pdf Aluminum_Filler_Alloy_Selection_Chart.pdf (20.9 KB, 42 views)
__________________
Jeff
Reply With Quote
  #18  
Old 01-04-2020, 01:50 PM
crystallographic crystallographic is online now
MetalShaper of the Month October '14 & April '16
 
Join Date: Dec 2010
Location: Western Sierra Nevadas, Badger Hill, CA
Posts: 3,446
Default

Quote:
Originally Posted by sandmanred View Post
Here's a table of weld properties based on the two base alloys and the weld properties you most value. It's best to print it out and tape it together. You pick your base alloys off the edges and the intersection of the base metals provide codes for weld properties for most common fillers. For a tank I'd usually pick 5356 for 6061/6061 due to the high ductility of the weld. I would expect 1100 to be very ductile too but not sure about the rest of the weld properties.

Global selection, from a global supplier.
__________________
Kent

http://www.tinmantech.com

"All it takes is a little practical experience to blow the he!! out of a perfectly good theory." --- Lloyd Rosenquist, charter member AWS, 1919.
Reply With Quote
  #19  
Old 01-06-2020, 11:18 AM
sblack sblack is offline
Member
 
Join Date: Sep 2011
Location: Montreal
Posts: 371
Default

Quote:
Originally Posted by crystallographic View Post
Hi Scott,
The problem with that tank is not the design. That design is proven by the tens of thousands of airplanes carrying fuel inside it, for decades, in rough environments. Had there been an issue with that design it would have been subject to an Airworthiness Directive, and the entire industry would have changed that design.
The problem is in the execution of that design, by the craftsman building that tank.
Example from real world aviation:

I have two wing tanks here that were made by a "pro fab/welder" shop.
.040 5052 with flanged ends that were fusion welded with a tig/TIG.
Made to "copy" an original 1940's existing pair of tanks.

The cracks in these end seams are not longer than 14 inches, and the fuel also splashed all over a hot engine and pilot, as he got down as fast as possible, and got it all shut off in time to exit safely and the plane was saved.
Problem: no baffles to support the long distances between end plates, so under positive and negative G's the seams failed, opposite each other and at the same end panel.

Attachment 54670
Attachment 54671
It does help to have some familiarity with the various aviation tank designs so as to make clear and accurate conclusions.
Thanks Kent. I was sure based on your writings that the flange weld design was not the issue. Given that there are many bushby mustangs flying without any problem and that the tank was supplied by the designer so presumably built to plan, i suspect poor metal prep. But I never saw the tank myself so I guess we will never know.
__________________
Scott in Montreal
Reply With Quote
  #20  
Old 02-17-2020, 07:27 PM
Sean in CT Sean in CT is offline
Member
 
Join Date: Jan 2017
Location: North Stonington, CT
Posts: 25
Default

So.....i wasnt happy with my 6061 0.060" tank. Kept getting little cracks > rewelded = shrinkage......scrapped it

Bought some 3003 0.125" Cut it up with table saw. Got the fits really nice, including baffles. Welded it up in no time, no cracks, no leaks. Its a little overweight i supposed, but hey, i could lose a few lbs too.

cdbd7NCu7l10WVu9KJEwxl5FwQYNh3YoIGFktWc542qeAgvqi3BKbLB_5nXOG-5RoTFfvxBtRDlPn41kF3wS0bYLalACHfIW.jpg
__________________
Sean

Last edited by galooph; 02-18-2020 at 08:42 AM.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

All times are GMT -5. The time now is 12:08 PM.

Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.