I have a large supply of NiMh rechargeable batteries that I use for all my gadgets. I have several small very bright LED flashlights that use these. My question is how can I construct, from parts, a strong bike headlamp that uses such batteries? Based on review of many product recommendations offered, I have not found a 300+ lumen light, so I'm looking for a way to construct such a light. It would need to be easily detachable so I could remove it when locking the bike in public.
I am interested in solving the problem of getting such a light, either buying or constructing such a bike light, by someone who has done this and tested it successfully, not a "which brand is best" product recommendation. I would think that pointing me to a general approach and/or a general place (links, marketplaces, etc.) where I could find such items would also answer the question without violating the norms, as long as they are fact based (e.g. successfully used by someone to solve this specific problem) and avoid naming specific brands or models. I don't know or care about electronics because it's beyond me how to put together circuits or LED's, I just want to understand how to get components that can be combined to attach and mount something very effectively and waterproof to a bike. I'm not looking for instructional material or training on bicycle repair or how to upgrade a specific kind of bike. Just how to solve a problem.
I'm just kind of a clueless dude here trying to ask how to go about this, looking for different approaches, some involving DIY and some involving using framework approaches, and letting people vote on which approaches worked best for them. (See https://stackoverflow.com/questions/9329446/for-each-over-an-array-in-javascript/9329476#9329476 for an example of this kind of a question/answer dynamic on a different StackExchange type site.) If that's not the intent of this site I guess I get it, keep the question closed. I put up a question in meta about whether this question is appropriate and so hopefully, as this is closed, the discussion on the appropriateness of this kind of question can be contained there.
One other thing is I'm looking for inexpensive way to go about this; $200 is out of my range.
I want to just point out that I am approaching this as an engineering problem with a bunch of requirements and looking for empirical evidence to support a solution. I am hoping the fact that some components involve electronics or that there are cost constraints does not take this out of the realm of ordinary bike problems.