Try number two on the cooler mount!
With the last one a beautiful failure, I tried again, making it larger. My reduced 35 degree angle turned out to be much closer to the original 45, though.
http://www.darts-page.com/images/Nieuport/cooler006.jpg
And done.
http://www.darts-page.com/images/Nieuport/cooler008.jpg
Of note is four things here:
1) Steel is doubled up. I have no idea if this is going to be an improvement or not; my thinking is that it's going to half my chance of a catastrophic failure before detection.
2) Rubber grommet at the engine mounting points, with washers on both sides larger than the holes. This is to hopefully reduce the vibration on the piece, and if the rubber fails, at least the mount won't fall off the engine.
3) One can't see it, but there is a steel plate under the oil cooler itself holding things together. My first installation had the nuts going just to the cooler itself and it's soft metal. Now it will have to think of a different way to fail. I also put in two more bolts to even out the load on everything.
4) The baffling material on the cooler between the mount and the strip underneath isn't there to prevent vibration as much as it is to reduce the friction between the steel mount and the steel strip against the cooler.
Oddly enough, when I enlarged my mount measurements and bent them, I came pretty close to what was there originally.
Everything ran great and no leaks.
For those wondering where I get my engineering inspiration, note the decal.
http://www.darts-page.com/images/Nieuport/cooler007.jpg