Tuesday, June 25, 2013

SLO County Power Outage

 Sunday night in San Luis Obispo, we experienced a very odd brown-out / black-out condition for several hours. I was walking over to a friend's apartment, when suddenly all of the street lights went out. This by itself isn't a very odd failure, but then immediately Marcel (who has been crashing on my couch) calls me on the W6BHZ repeater and asks if it's normal for my apartment's wall power to drop to 75VAC.

It very much is not normal. My apartment's wiring is bad, and often sags several volts (120->115VAC), but 75 is ridiculous. I look around for the nearest lighted window, and yeah, SLO power is grossly missing volts. Other amateur radio operators jump on and confirm with their multimeters that voltages across SLO are varying from 70-80VAC.

The county-wide SLOECC repeater was having the same sort of conversation, but wasn't scaling anywhere near as well as on our local campus repeater. We clearly needed a net control operator to get people to stop talking over each other and only pass useful information on such a long-range channel.  Sadly, it didn't sound like any of the other ECCs were activating, so over the W6BHZ repeater we organized a carload of hams to get to the club's ECC-16 on campus.
While they were en-route to campus, I talked Marcel through how to operate one of my latest projects, which is an emergency 12VDC system for my apartment. All of my routers, my radio, my USB power strip, and some LED lighting are all run off of a 17Ah lead acid battery in case of lost power. I had actually just done some of the major wiring for this project Sunday afternoon, so all Marcel had to do was by-pass the unfinished change-over controller from wall to battery power.

This meant that my apartment had Internet, WiFi, emergency LED lighting, and USB power (which came in handy when more hams showed up at my apartment with dead cell phones). Marcel was then able to activate the SLOECC net until we were able to spool up the campus ECC on generator power.
The project isn't completely finished yet. I'm having some noise issues with the long 12VDC runs throughout my apartment (which is piggy-backed on some of my 100bT Ethernet links using passive PoE injectors), and the 17Ah battery really isn't big enough. Between the routers and lighting, the system draws two amps before you factor in transmitting on the radio and >10W of cell phone charging. I have a 400Ah flooded pack ready to be added to the system, but need to route wire between my patio and my bedroom distribution point. My landlord wouldn't mind me drilling some holes, right?

The biggest lesson to take away from this event was probably the USB power issue. We have backup power for the radios and lights, but clearly the next generation of ECCs need to put more thought into how to keep everyone's mobile devices charged. We were lucky that this was only a single-evening event, since my system was down to 11.7V by the end of the evening, which was just above my 11.5V cut-off voltage, and our ECC didn't have any additional gasoline for the generator after what was in the tank. The extra 15W of cell phone charging load on top of the 20W of routers and lights was higher than my original power budget had planned for, and made a huge difference.

I'm now designing a high efficiency and high power/port count USB charging station, as an improvement to my USB power strip, which is only powerful enough to fast charge a single smartphone at a time. It's going to be based around one of CUI's quarter brick DC-DC converters, which they were kind enough to give me last month.

Monday, June 24, 2013

CPARC Spring Work Week

Just in case any of you have been getting worried, I am in fact still alive. School this quarter did nothing short of kick my ass. My amateur radio club duties consumed literally every weekend of this quarter, and teaching (and more so grading) a lab section in the EE department was a stimulating experience that didn't yield much blog post fodder. 
Once finals week wrapped up last Friday, my buddy Marcel showed up and has been crashing on my couch (which is almost always open for any friends or hacker types who need a place in SLO) for the last week while we had another whirlwind work week.  The club turn-out was great, and we made great progress on building the club's infrastructure, maintaining what we already have built, and just generally having a good time playing with radios.

The first major project of the week was tearing down the old University Police Department's radio tower. UPD recently moved to a new facility, and CPARC volunteered to assist in the demolition of the old building.  This doubtless saved the university a few thousand dollars on their quote from the demolition company, and we get a lightly-used 40 foot Rohn 45 tower for club use.

On day two, we turned our focus back to the club shack, and finally hanging the club's 80m dipole on our new tower. Due to the slightly different location of our new tower, and how inconvenient it is to deal with an 80m dipole on a telescoping tower, we spent quite a bit of club member engineering talent designing a clever auto-retraction system to allow us to raise or lower the tower as a one-man job.
The problem is that when you lower a 72' tower, the ends of the dipole fall on the roof or people below it. This means you normally need two people at the ends to spool up the dipole, and the antenna becomes useless when the tower is down (This is bad since 80m NVIS is an important capability for our shack in storm conditions, since we are an Emergency Communications Center for SLO county). To overcome this issue, we came up with a very clever and clean counter-weight design based on railroad catenary systems, where you have a pulley and weight at the end of a line to keep constant tension under varying lengths.

After working out the trigonometry, we figured the end of the dipole needed to move 15 feet as the tower went from fully retracted to fully extended. We then purchased two 20 foot pieces of 3" conduit, mounted a pulley on the end, and used 3' heavy-duty tent stakes as the weights. The weights are inside the conduit, and move from hanging two feet from the bottom of the conduit up to just below the pulley as the tower is extended. A very clean solution that works great and looks much simpler than it actually is. We were particularly impressed that once it was all built, our math for how long the lines and weights needed to be were off by less than six inches on both sides.

This was the biggest job of the week, so thanks again to everyone who helped out, and a huge thanks to our EE department head, Dr. Derickson, for allowing us to erect even more masts and towers on the EE building.

The rest of the week was spent working on countless projects big and small around the shack and EE department. Since the amateur radio club maintains quite a bit of expertise in battery management (due to our ECC battery bank and general mobile battery ops), we are the department's de-facto golf cart care-takers. We are happy to do this, since we are by far the biggest users of the department's cart for moving our equipment between buildings on campus.  The cart's last battery pack had finally given up the ghost this quarter, so we generated for the department an $800 quote for the new pack of batteries we would need to return the cart to service. The department gladly fulfilled this for us, and we enjoyed spending a day on a fun battery project.
After replacing the pack, we identified the bad battery in the old pack, and cleaned up the rest of them for possible future application replacing the club's currently non-operational battery backup (a very old flooded NiCad pack from a hospital operating room).

After a long week of being on campus from 8am until well past dinner, we were looking forward to finishing off the week participating in the ARRL Field Day contest. We decided to activate our shack using the club vanity call N6CP as an emergency operations center with two active operating stations (meaning we were class 2F). I sent out a lot of the pre-weekend emails to drum up enough operators to keep us active most of the 24 hours (we were down for only 6.5hrs throughout the weekend), and helped Marcel with much of the final testing before the contest kicked off at 11am Saturday.
My contribution that I'm most proud of was setting up WriteLog on all the shack computers so that we had tab auto-completion throughout the contest, and live log updates on the shack projector for everyone else to watch without shoulder-surfing the two operators.

Thanks to all the club members and alumni who showed up and helped us have such a good time. Our rough contact count rounded out at 871, which we were really happy with. Here is our soapbox submission for the event:
The Cal Poly Amateur Radio Club W6BHZ/N6CP operated ARRL Field Day from the club shack as a 2F EOC taking advantage of the brand new 72ft antenna tower with a TH-7DX Tribander, XM-240 40m beam, and 80m inverted-V. Almost a dozen students and alumni activated the shack in the most active contest the club has seen in several years. Many brand new hams got to try out HF contesting and everyone did a wonderful job handling the pileups on both voice and digital modes. As a 2F station, we had the Yaesu FTDX-9000D and the FT-920 operating concurrently into the early hours of the morning. With only a four hour break at 5am, the station was on the air for a record 19 hours. Be sure to look for the Cal Poly station in upcoming contests and for Field Day 2014!