September VHF QSO Party

September was not a calm contest. For a variety of reasons, we didn't get up to the mountain until early Friday morning - a full day behind. My personal goal of a full suite of bulkhead connectors for 10, 24, 47, tower power and control was satisfied. I tested every one before the contest by pulling out the gear, hooking it up and running it. It all worked, both in testing and on Mt Greylock. But I'm ahead of myself.

Friday was clear and hot: sweltering hot. We set to work unloading the tower trailer. All the towers were re-taped with appropriate colour codes, and the colour chart was lovingly removed from Wilfrid and put on the back of Roy so that anybody could tell which tower was which. We got all the tower sections off the trailer, pushed the trailer into the circle and started unloading the antenna trailer (which was already in the circle.) We self-organised (woo hoo! shades of eXtreme Programming!) into crews that assembled towers and antennas. Don't forget the guys tying tree ropes and uncoiling guy wires...

By lunch time, we'd assembled all the towers and got 3 pushed up. Took a break for lunch and went at it again. By nightfall, had all the towers up and most of the antennas, too. In my case, I had the 440 yagi, 10GHz wideband, 47GHZ and 24GHZ wideband on the tower. Note to self: 47GHz goes all the way up, onto the top section below the curve!

Saturday dawned early, bright and hot again. I'm the default IT guy, although poorly suited to the task. I got the computers running the newest version of RoverLog and started working on the fiber runs between trucks. Of course there's always some fiddling, but after swaping connectors inside the hub (Roy) the fiber came up.

Had to get my own station set up, but that went pretty fast due to the bulkheads. The 222/432 truck was having issues with their rigs. 222 in particular didn't get on the air until after the contest started. In a tizzy over the network (no network, no contest!) I waited until late to haul up the 6 foot dish and TWT. I didn't get on the air until after the contest started, but I had a backup operator so I wasn't completely worried.

Eventually, it all came up except the 2m computer. It turned out that there was an IP address conflict, which meant we had to change the IP address. Of course we also had to change the IP address in all the RoverLog config files, and a restart of RoverLog lost all the skeds. Not good.

By and large, it's a simple matter to pass a calling station among the lower bands, because most of the lower band Qs are made on voice. That means the lower band stations (on average) are busy for less than 30 seconds. On microwaves, many, many Qs are made with CW. Some of those Qs are very weak stations, and it can take upwards of 10 minutes to dig the Q out. Unline the lower bands, microwave stations drift in frequency; especially rovers.

When a pass comes in for the microwaves, there's a very good chance that the 902/1296 station is currently engaged in a QSO. That means the incoming station has to wait - in effect, to take a number. Depending on how busy things are, that can be a deep stack. Because it's impossible to tell if any given QSO will be done on SSB in 30 seconds or on CW in 10 minutes, we have to schedule 10 minute blocks for inbound microwaves. This can result in a situation where we work someone quickly and have a 9 minute gap before the next sked. The same situation holds true for the higher microwaves as well.

From the perspective of a rover waiting to work us, it must seem horribly frustrating to have to wait so long for us to turn the dish and work him, but it's an artefact of how our microwave stations are set up. The fact is that a typical rover is much more agile than a multi-op station can be. We don't have all our microwaves on one tower, with a single operator to handle them all. The microwave duties are spread among 4 different operators and 4 towers. If one of the 2304 stations is transmitting, the other has to wait (ARRL rules.) So we sometimes do goofy things like sked you on 3456 before 2304... to squeeze those extra minutes into the schedule.

The rain moved in on Saturday, and completely wiped out my 24GHz Qs. Between rain and fog, the duraation of the contest was wet. Made only a few 24GHz QSOs, but did make a bunch of 10GHz Qs. Wanted more, but between scheduling problems with RoverLog and the lack of success on the lower uW bands, I didn't get too many stations who worked their way up the bands to me.

When stations called in on 432 and asked to go to 10GHz, it takes time to turn the tower - over a minute sometimes. And I don't get to work you if the dish isn't pointed your way!

The end of the contest was disappointing. We realised late Sunday that the scoring was incorrect. It looks like the logs are OK, it's only the scoring that's the problem. So we really don't know our score yet...

Monday went fantastically well. The weather spit only a little bit and we got everything taken down and loaded up in what seemed like record time.

And so another contest comes to a close.