Hello from the Microship lab....

This issue is going to about 20 new project volunteers as well as the 20 usual recipients (major sponsors, project participants, UCSD faculty, and close friends). As you new folks move from having innocently expressed interest to becoming drawn into direct participation, you'll be moved to the list that routinely receives these updates.... that is, whenever I actually WRITE them. I'll also put you on the huge Nomadness alias (over 3,000 people) for the occasional full-scale chapters, but those are much less frequent than these, um, "daily" reports.

The past week (!) has been crammed with managerial work and a few noteworthy developments. In the former category, quite a bit of time was involved in the creation of the "Microship Projects Catalog" which you should have already received. Although that was exhausting, it was an extremely useful formalization of the whole system and has already proven its value as a framework for project planning. It's also giving project volunteers something more tangible to think about than the vague notion of "helping that guy build a boat."

Today I had a very enthusiastic meeting with four people from Nelson/Marek (the marine architects who are designing the physical boat for us). This ranks as major support -- without them, my options would consist of bumbling through the process of inadequately reinventing the wheel, or trying to retrofit a set of unique requirements onto a commercial boat designed for different purposes. Nelson/Marek's contribution of design expertise is significantly increasing the probability that this will be a successful system.
In the meeting, after the requisite BEHEMOTH demo, I talked through all the requirements of the Microship -- then they gave me some homework to do. Step one: define the weights and general locations of all known gear. This will help determine waterline, fore-aft bouyancy balance, locations of bulkheads and outriggers, and so on. Actually, this has been discussed before, but now I have a spreadsheet and the befefit of some solid discussion on the level of detail required (and its rationale).

Robb had an interesting idea today... I don't know if I've spoken in these reports of the crossbeam angle (dihedral) issue, but it's a sticky one. In short, where relative to center-hull waterline should the outriggers be positioned? When under sail, it's good to have them high so I can reduce wetted surface by flying a hull. On flat water under solar propulsion -- or at anchor -- this would drive me crazy, tipping back and forth with every move. We've had some discussions about clever ways to tweak this, most involving too much mechanical complexity.

"Why not just sink the center hull?" Robb asked -- responding to my panicked look with the explanation that flooding or pumping out a ballast chamber would let us have the best of both worlds. At rest, the tank is full, lowering the boat until the outriggers are carrying their share of the load. Under sail, we lighten it and fly. Clever. Since we can define the hull shape any way we want, including the addition of above-waterline flare, we can define the waterline-flotation transfer function up front and not have to kluge it into something that wasn't intended to behave this way.

For purposes of calculation, we've established reference planes at the forward bulkhead and deck. Stations will be defined in terms of feet forward (-) or aft (+) of the bulkhead separating the forward two hull segments, and vertical CG of components will be expressed relative to the average plane defined by the hull-deck joint in the center segment (this can always be mapped to the more conventional DWL later).
There is some chance that we'll be building our own outriggers as well, and not depending on commercial kayaks. While this seems like an instance of the Not-Invented-Here syndrome, it is appearing that solving the problem right will require either some agressive kayak hacking or hulls designed from the keel up to suit the Microship. Among other issues, they have to be bulkheaded at 10-foot intervals to suit the placement of crossbeams, they will require extremely beefy decks to accommodate the stresses of attachment, and all hatches and cockpits will have to be fitted with covers that won't wash off in rushing water when I'm burying a hull at 20 knots. These are things that kayakers don't have to think about... yet the amas must still be operable as kayak-like craft (albeit pedal powered) when detached from the mothership. Should be interesting. At least, this way, we can make all the hulls visually match...
* * *
I spent some time this week on the definition of the Microship's network architecture. This takes a CAD system or drawing package to do right and I'm not much of a fan of ASCII art, but, hey, this is email....

AppleTalk Network ----------------------------------------------------------> ext hooks | | | | | PB 180C PB 170 file Wireless | console crew server gateway | (MAC) | | MAC|DOS | RF | ---|--- email external-------| gateway gateway laptop | | RF | | flotilla-------/ | boats | |Ethernet /-------------------> ext hooks NMEA 0183 Bus | | -------------------------------------Compaq /------Ampro | | | | | | 486 | PC104 (DOS) GPS Depth Kts Steer radar | Nav | MCS | | | \---------------/ | | | Power management processor--------| | Power distribution processor------| (PIC) | Propulsion control system---------| | Communication manager-------------| | Multidrop Internal data collection----------| (RS485?) | Microship Environmental data collection-----| Control | Network Video system controller-----------| | Pneumatic and packaging control---| | Audio crosspoint server-----------| | Security system processor---------| | Diagnostic/self-test system-------| | I/O expansion, general------------| o | | | Comm----------------Packet access------------| | \--> ext hooks sat/cell, laptop

There. Notice that there are three regions, labeled MAC, DOS, and PIC. I want to emphasize the point of this architecture without getting too mired in detail...

Basically, we're dealing with different computer worlds here. When I'm on board and working, I live in the Mac world and want AppleTalk. It will be extended via Digital Ocean's wireless boxes to the manpack laptop and the machines on neighboring boats, but basically it should feel like a cozy office network of Macs (complete with hooks to cellular and satellite-delivered email).

A network gateway ties my "personal productivity" machines to the boat's DOS world, consisting of two machines -- the Compaq 486 for navigation, charting, and marine software... and the Ampro PC/104 system that is the diskless, low-power hub for the Microship control system. Both of these get to see the NMEA-standard bus for communication with shipboard nav and off-the-shelf maritime instrumentation. The concept here is that marine functions will pretty much take over a machine, and I don't want to have to keep context switching my Mac between productivity and survival. This also gives me two major platforms /for available software, minimizes single-point failure potential, and doubles visual display bandwidth. The keyboard and pointing device will be seamlessly switchable between the two worlds.

The Ampro machine (MCS) is in charge of a multidrop network of extremely low power, cheap, distributed servers and controllers -- each dedicated to a small and manageable task. If you recall the project breakdown in the mega-document I just sent you, then you'll recognize these... each is owned by a project group and localizes failure potential to a single functional region.

Because of low power and the need to have peer-to-peer access between controllers, this is also the interface point for external packet-radio or laptop access to the kayacht's control system. High level logging in (via telnet or email daemons, PPP, or whatever) will take place in the Mac world -- responding to a power-control command from one of the tiny machines that's always on and listening.
Sailing just ain't what it used to be........

Enough for now. I'm having a succession of meetings this week with new project volunteers, then planning a group meeting sometime soon to start nailing down specifics. The hottest human-resource issue right now is management: I need a team of 3 people who are collectively cognizant of ALL parts of the project and want to help manage the focused workgroups. This will be time-consuming, but very exciting since you'll get to play with everything at once. If this is for you, please talk to me NOW!

Cheers from a thrilling Friday night on campus.... <grin>

Steve