Thanks for your input, gentlemen.
Yes, I absolutely agree that 100 lm is too much for the intended use of the light as it was on the original. I mean, a light made to read instruments, maps, flight plans inside a capsule on an
emergency. But again, being absolutely loyal to the original, I should not use an LED and a driver, but a bulb. I even thought the possibility of using a direct drive instead of a driver -which is something I havent discarded- as it will be even more reliable.
But reading or working under such low light conditions is not recommended and we know that this is a
real concern of NASA. So the question is: Was 15-20 lm the flux they wanted or just what was possible to achieve?
What I'm trying to get is a balance of the old style outlook and ambience with nowadays technology for the intended use. Sticking as much as possible to the original design is possible, but at the end they are not going to be the original, because they are not. They are made now, in our time. So this is a give and take. I'm using AA batteries which are technologically dated, but I'm keeping the shape as much as possible. I'm using a warm white LED, and a floody lens with high CRI, which is what you need to read or find a cable into a circuit, for example. I'm making the light waterproof, which will make it virtually ATEX certified, which the original was not, giving it more performance, and skipping weak points of the original design. As I said, is a balance, but I don't discard making a lower output one, with a crazy runtime.
Javier