The Microship Status Reports

Microship Status 12/21/93 by Steven K. Roberts

In This Issue:
Mcs Hub Board Alive!
San Diego Supercomputer Center
Kayak Order, Structure, Thrusters
Landing Gear Akas
Tadpole On The Net

Miscellaneous Updates

Solar Still IEE Keyboards Autodesk donation Specification document Media watch Lab quest
Hello from the eerily silent world of a campus on holiday. The few denizens still slaving away in scattered labs pass on deserted sidewalks, pause, and exchange a few words... "sure is dead around here, huh?" "Yeah... Is anything open over at Price Center?"
I've been here all day, every day, and sometimes all night as well. I will be here working on the ship even as families across America are huddled in cozy cocoons of consumerism. It seems so far away, bizarre yet poignant: music in the lab ranges from Vivaldi to Depeche Mode, and, mercifully, I haven't heard a Christmas melody yet this year (though "Greensleeves" is forever tainted, even while Rampal is playing variations on the traditional theme sandwiched incongruously between Franck and Prokofiev...). But when I glimpse a decorated tree still alight on my 2AM bicycle commute back to condo hell it suddenly seems an echo of childhood, almost confusing...
No matter. The tasks before us continue to occupy center stage, and although I have been slow on updates lately since over half the mailing list is offline for vacation, things have been happening.

Mcs Hub Board Alive!

I mentioned in the last update that New Micros was about to ship an NMIT-0020 as the first in our suite of 16 FORTH 68HC11 boards for the control system. It arrived about a week ago, and I spent much of this past weekend packaging it on a development chassis (the lid of an old TNC case) along with a small 5-V power supply and a 2-line LCD behind a sloping lexan panel. About half the job involved the usual screwing around with RS-232 (can you imagine how many man-millennia have been wasted hacking serial ports since the dawn of the computer age?), but now it works with both Mac and DOS machines here. I even hand-entered a page or so of LCD driver routines and had fun sending characters to the screen until a pilot error sent FORTH careening into a stack crash that resulted in amnesia.
Enroute now are the 5001 serial board (one serial port, configured with drivers for RS-422 that will host the multidrop network) and the 9003 real-time-clock. This should come close to the complete configuration of the hub, and will give us a FORTH development environment replete with enough I/O and resources to provide the requisite visceral joy of actually seeing a computer DO something. If you're contemplating getting involved in one of the control system teams, this is your invitation to begin the FORTH learning curve. It's here, it works, and you can use Mac or DOS to talk to it.

San Diego Supercomputer Center

TJ Tyler and I now have access to SDSC, with accounts on the Cray and SGI machines in progress. The Cray will be used for extensive finite element analysis of hull, frame, landing gear, outrigger, and other structures; the Silicon Graphics systems in the Visualization Lab are already being used for renderings of the ship based upon Robb Walker's sketches, my comments, pictures of kayaks, and any other input we can provide. My sincere thanks to Len Wanger, Mike Bailey, and Mike Kelley for making this happen and working on these magnificent images! T-shirts soon...
We took a tour there last Friday evening and were, frankly, blown away. This is a remarkable facility, staffed by some equally remarkable people. This will be a powerful tool for helping us design this system, and Robb is eyeing it for its potential in working with more traditional clients (imagine handing a prospective yacht buyer a picture of her yet-imaginary dreamboat floating in some dramatic cove...).
I'm sure we'll have much more to report here as artists' renderings and mechanical designs proceed...

Kayak Order, Structure, Thrusters

Well, I did it -- I just sent $750 as a 50% deposit on the custom hull and deck order from Current Designs in Sidney, BC, Canada. They're being very nice to us here, basically charging only for materials. What we should see by the end of January is a pair of flanged hulls with extra-robust layup, a matching pair of flanged decks with hatches but no cockpits or rudders, H-extrusion for the whole lot, four bulkheads, and miscellaneous components. This will give us a good head start on the amas, and our task will be to create smooth cowlings for the pedaling envelope, beefed-up structure and detachable outrigger interface, pedal drive assemblies, and hatch covers that can go away in use yet withstand the stresses of rushing water. Should be amusing.
I've moved the thrusters to the inboard sides of the kayaks, in the form of flip-down leeboard-like assemblies pivoting around the pedaling axis. They will now be direct drive, but linked by a clutch to a motor-generator. In fact, there are three clutches, allowing any of four operating modes (in addition to retracted and passive, of course):
o Direct low-loss mechanical pedal-drive o Motor drive from solar or ship power bus o Pedal-charging batteries without involving thruster o Wind-charging batteries by lowering drive units while under sail

Landing Gear Akas

Speaking of interesting mode changes, I had a thought the other day while working with TJ on the mechanical structure. Since Robb's initial weight distribution and trailer-mode calculations suggest that we can in fact locate the aft outrigger and landing-gear attachments on the same transverse plane as the aft bulkhead (between cockpit and crew module), the obvious refinement is to use the aka arms as struts. By relocating one pivot point in the Farrier-like folding system (see the cardboard hinged model in my lab), we can detach the kayaks, attach wheels, and swing the arms down to form landing gear. Tongue weight should be about 10% of the 3,000-pound estimated total, and this constraint will provide feedback to the weight analysis currently being undertaken on Excel by Scott Poorman.

Tadpole On The Net

I'm going to ask John Studarus to write something for us detailing this when he gets back from vacation -- he's the unix/net wiz in the group, not me. But the short form is that my Tadpole SPARCbook laptop is now at Scripps, continuously on the Internet. We've even taken on the $12/month fee for domain name service at UCSD, so this is official.

The machine is a Gopher server, already accepting dozens of connections a day from all over the world as people snoop around and grab my nomadness reports and GIFs. It's also set up as a mail host for Microship project newsgroups, and will be the archive site for all these reports. You'll be hearing a lot more about this once John writes a summary of the available services.
The plan is to integrate this into the ship, linked by ethernet to the Macs and the Ampro PC and acting as file and SMTP server. The "Microship Control System" drawing, available now in hardcopy for anyone interested, shows the overall architecture.

Miscellaneous Updates

Solar Still: Dan Perry has acquired much of the hardware for the folding still project (to make fresh water from seawater). The top and bottom frame is made from teak... and this is a good lesson in another reason, besides rainforest conservation, NOT to use rare tropical hardwoods as building material! A 3/4" plank, 8" wide and 4' long, was $40. <shudder> Anyway, I'm looking for a volunteer or two with good experience in woodworking and fabrication to work with Dan on this -- hopefully someone who has a table or radial arm saw to rip this heirloom into the sizes we need. If you want to help, contact Dan directly: <dperry@UCSD.EDU> Can't wait to pour in a bucket of seawater, sit in the sun a while, then drink the results! Sounds like a way to justify a beach day as research...

IEE Keyboards: I think I mentioned that Paravant is out as far as waterproof keyboards are concerned -- since then, I've found a more robust source with a HUGE product line of keyboards and displays for harsh industrial and military environments. We've swapped literature and the vibes are good -- I should know in a couple of weeks. In the meantime, I connected them with Silicon Valley Bus Company, whose Keystone product designed by Jay Hamlin accepts DOS keyboard and mouse inputs, producing ADB output for the Macintosh.

Autodesk donation: Thanks again to Autodesk for donating a second copy of AutoCAD Release 12 for Macintosh -- this time to Nelson/Marek Yacht Design for use on the Microship project! We'll use this for the whole system design, as well as input to the visualization software at SDSC.
Specification document: I'm trying, really I am, to finish the full spec document during the "holidays." I've finished four of the 16 sections in the control part, and haven't even thought about the high-level applications yet. The intent is to have initial specs ready for all student groups beginning FORTH-based control projects in January. Ready now: Hub, Audio Crossbar, Serial Crossbar, and Video.
Media watch: The Jan/Feb issue of _Internet World_ has a 5-page article by yours truly, with lots of pretty BEHEMOTH photos. The subject is "Technomadness and the Internet," focusing on the methods of decoupling geographically while maintaining a virtual home online. There's a small article about this project in the December 3 issue of the local rag _ComputorEdge_, the next issue of UCSD's _Perspectives_ will carry a piece on the Microship project on campus, and the whole Microship overview along with control system and rendering artwork will run in the February issue of _IEEE Potentials_. I also interviewed with _Popular Science_ a week or so ago and am negotiating a French documentary film deal. I need something to keep busy, you know...

Lab quest: Finally, I'm not sure how much good this will do, but I might as well mention it here. By March 11, I have to be out of my current borrowed lab in the Engineering Building, and I'm hoping that the next move will be to a place where we can complete the project without further disruption (a panic last week ensued when we discovered a conflict with a planned course... the professor graciously agreed to share space in another lab, but it's still only a winter quarter solution). If ANYONE on this list knows of 1,200 or more available square feet on campus with ground-floor access, wide doors, a net connection, clean environment, and reasonable security, please let me know ASAP! Things could get very difficult if we don't find a place to swarm around the hull and spend a year in system integration...
With that, I wish you all happy holidays -- I'm eagerly looking forward to the resumption of campus normalcy, easy access to services, and the energy of many people working together on a thoroughly mad project. Cheers!

Steve