Build hope?

A project log for TrillSat

Texting Without Cell Towers, Power Grid, or the Internet: IoT meets AX.25 Aboard an Arboreal Space Elevator

Lee DjavaherianLee Djavaherian 07/25/2018 at 02:433 Comments

The twenty 2018 Hackaday Prize Power Harvesting Challenge winners were announced today, and again I did not win a single thing, not even a non-monetary achievement.  This is the 3rd loss for TrillSat in a row, and I see a trend here, unfortunately.

You'd think that its unique mass-tilt/catenary spiral-axis solar tracker (both elevating the craft for line-of-sight radio communication and approaching dual-axis efficiency) using a single $9 screwdriver motor (with a custom H-Bridge built into the handle) would win at least something, right?  If so, you would be wrong.

You'd think that the only tether/winchbot in this year's Hackaday Prize competition might have a chance to win something.  Heck, at the time of this writing, tether or winchbots are relatively rare, and TrillSat uses an onboard capstan winch for aerial locomotion, which is very rare (I've never even seen one used in this way before) and uses complex PWM drive mechanics.  And it even has a gyrofan, using mass x velocity for experimental inertial stabilization built out of a repurposed DVD spindle motor, and the BLDC is driven via programming of a single ATtiny.  But alas, it didn't win the robotics round--not even a non-monetary achievement.  The capstan was kicked to the curb.

You'd think that a tiny audio/data/PTT interface custom-built to connect to a $26 ham radio (which was dismantled to eliminate weight and allow variable power from both series cells and boost converters) with a custom program that uses 16-byte chunks to quickly program the clone-mode flash to allow it to mimic the features of more expensive dual VFO's for both APRS/Packet functions during a single session would at least win something, even a non-monetary achievement, right?  Sorry to disappoint you again.

You'd think that a custom Packet BBS in Python on a Pi Zero using Unix standard streams instead of the C library, connected to a custom XMPP server in Lua/NodeMCU on an ESP8266, things that had never been programmed on those architectures before (to the best of my knowledge) would at least earn the smallest of awards--again, nope.  I even had to build a software simulation in order to test it, since I couldn't put it on the air.  Heck, I even added waterproof, inductive Qi charging...  Cool, right?  The judges don't seem to think so.

And you'd think that showing proficiency with 4 interconnecting CPUs and 4 languages ATtinys (in C), ESP8266 (in Lua), Pi Zero (in Python/BASH) on a single project might garner some cred from the judges, right?  Nah!  It doesn't appear they look at such trivial things...

Oh, and the care taken to design all parts in OpenSCAD so the majority of the unit could be printed on a Prusa i3 printbed in PETG (built from kit, no less), using pronsole on a Raspberry Pi 3 would garner at least a tiny bit of modern maker cred, right?  That would be a big NO.

Oh yeah, and the wooden Ark and Test Frame support structure that I had to build to test and calibrate the unit indoors--it's another project in itself: it folds, uses counterbalanced tethers, collapses for storage, and assembles and operates in several different modes.  Neat, right?  Ha!  No prize for you!

And finally, I built an actual weatherproof, temperature-controlled prototype (it's not just vaporware like some of the projects), strong enough for testing, and I published all of the source code, schematic, and video.  I also documented it in great detail (almost 200 pages), including an itemized Bill of Materials.  But I also went through the trouble of writing up separate project text for Hackaday (until I ran out of room).  Did I get recognized for such meticulous work?  Of course not.

I don't see this trend changing and have therefore decided not to enter my haptic, tethered Morse code system into the 4th challenge (Human-Computer Interface).  The competition is over for me.  All future updates to the TrillSat project will take place where it first began at

Thank you Hackaday followers/likers for taking the time to acknowledge this project.  I often work on my projects in a social vacuum, and your simple gesture means a lot.


Jan wrote 07/25/2018 at 14:34 point

Ah come on. I fully understand your disappointment, but you sound like someone who really really needs confirmation from others. Why? Your projects (like you said yourself) are big ones with lots of details and knowledge in it. It is a CONTEST. Not everyone can always win. Look at your numbers. Many many people saw/rated/klicked your projects. The idea is. In the wild, just keep hanging on! 

  Are you sure? yes | no

K.C. Lee wrote 07/25/2018 at 12:28 point

I hear you.  That's part of the reasons why I am moving all my projects on to a different host ASAP.  Seeing that no one appreciates harder projects, there are no reasons for me to be here.

  Are you sure? yes | no

Lee Djavaherian wrote 07/27/2018 at 03:14 point

Good luck K.C. Lee.  I came across a few of your project pages weeks ago (many very interesting), and I will keep an eye on your new site.

  Are you sure? yes | no