User talk:Colin

Hi Colin, sorry for not seeing your question before. You should write on my talk page so I will see it, rather than the discussion for a content page. Design fields like mechanical and electrical engineering consider specific parts of a project. Systems Engineering considers the entire system. This includes how the various parts interact with each other, people, maintenance, and other factors beyond what you can put on a CAD drawing. Optimizing a system as a whole gives a different result than optimizing each part in isolation. The field has been around for a number of decades now, and has fairly well developed processes. It isn't cheap to apply the methodology if you are doing it right. So you have to consider if the benefits are worth the effort for a given project. Generally the more complex and higher value a project, the more useful systems engineering methods are. I used to work for Boeing (now retired), and their products are both complex and expensive. So the Systems Engineering methods were very much used there.

The OSE project is both complex, and "expensive" in the sense that if you replicate it to many villages, that adds up to a lot of total equipment and labor. Thus optimizing it is worthwhile. So far the OSE project has focused on a bunch of machines, but not so much on how they would work together and with people. So I decided to help with the latter part. I'm happy to teach what I know, and I am writing the pages in the wiki to self-explain things as I go along. If you have direct questions, you can contact me at danielravennest (at) gmail (dot) com. That way we don't clutter up the wiki with side conversations. If the answers are worth sharing, I will likely post them to one of the pages.

The basic principles are not that hard to explain. Humans have finite intelligence, so we have to break down a complex project into simpler pieces (functions) to the point you can actually design a solution. When you break it into pieces, you now create connections between the pieces. For example an electric motor now needs power input. That has to come from somewhere, and the "no loose ends" rule says you need to add that to the system diagram. Otherwise you have a flow that appears or vanishes from nothing. When you have a system broken down, you can now apply mathematical models to understand, optimize, etc. For example, an input/output spreadsheet can track if the total flows from all the components add up. For example, a power supply needs to feed sufficient electricity to all the components it runs. When you change the values for one component, it then propagates to the rest of the system. For example, engine weight for the tractor affects tread size for ground load, which in turn affects row crop spacing, which affects yield per acre. By seeing the effects at the system level, you then can make the right choices at the detail level.

-

> I'd like to compare notes with you, as I have only a basic understanding of its chemistry.

Well, I have to admit that my chemistry knowledge isn't all that strong, thought it was part of my engineering training (a long time ago). My interest is somewhat personal, some what project related. Personally, I want to be able to generate electrical power from a variety of alternative sources (solar, micro-hydro, and the steam engine - naturally). I will need a way to store that power and NiFe batteries seem like a good possibility for that. That's why I was tracking your inputs and when I saw a little bit to contribute - I did so.

> I've got a decent design figured out, using sheet steel and nickel hydroxide (build the battery discharged, then charge it). Turns out that a 1kWh cell is quite feasible.

Excellent news, Colin.

> Anyway, I hope to learn a thing or two from you and I hope we can flesh this battery out as well.

I'd be happy to contribute if I can, but I fear that I don't really know all that much about it. The steam engine is my priority for OSE at this point in time, but batteries are part of the over all power strategy, so I want to understand it. Mark J Nortonn