Catching up in Rocky Mount

We are at the Elks lodge in Rocky Mount, NC (map). Our guide said this lodge had parking with electric power for $10, but the price has gone up to $15. Still, that's $10 less than we were paying over in Raleigh, and we have the whole place to ourselves. We've got a pair of 20-amp outlets, and there is water and even a dump on the property.

We ran the air conditioner when we arrived, but the last couple of days have been cool enough that we've been OK just with the fans, so paying for power seems a bit excessive. However once we were settled in with the scooters out, we didn't want to go hunting for other digs, and besides, this is a nice private spot to get some work done up on the roof, which is not the sort of thing I would do at, say, Wal-Mart.

Speaking of fans, Friday's roof-top project was getting the bedroom FanTastic fan working. It quit entirely on Thursday, and no amount of fiddling with the three switches would get it running. I could tell it had power, because the separate motor which raises and lowers the lid worked, but the fan itself would not spin. My guess was a corroded pin switch, which stops the fan from spinning when the lid is closed, and that's a roof-top repair.

The bedroom fan is on the one part of the roof that's hard to access, necessitating a precarious shuffle between the solar panels and the roof edge on a slippery slope. I took the inside grille off the unit so Louise could hand me tools up between the fan blades as I worked. The pin switch turned out to be fine, and poking around with the meter revealed that the power was stopping at a small thermal cutout device just under a small grille in one corner of the unit. That turned out to be an inside repair, and so back down I went after poking all the tools back down through the fan.

As it turns out, FanTastic implements the "three speed" feature of these fans by using a DC motor and a pair of open wire-wound resistors for the two intermediate speeds. The fan itself is supposed to blow some air over the resistors, and the small grille in the corner is for the heat to escape. Centered in that grille immediately above the resistors is a small sacrificial thermal fuse. I think what happened here is that something dropped into the fan from outside, keeping the blades from spinning, which caused the resistor to overheat -- we almost always run the fan on its lowest setting. That blew the thermal fuse.



I've got a note in to FanTastic about getting a replacement. In the meantime, I replaced the thermal fuse assembly, which had spade lugs, with a 10-amp ATO automotive fuse I had lying around, as the blades on these are about the same size as spades. At least we'll have some ventilation in the bedroom while I wait for a replacement. With five mammals on board, working vents are mandatory.

Yesterday's rooftop project involved the long-dormant satellite dish repair. Regular readers know that the automatic aiming function has not worked now for several months, the problem initially starting when HughesNet moved us to a different satellite transponder. They have since moved us yet again, but alas this did not magically fix the problem. As I wrote here back in June, I had purchased some used dish parts on-line in order to continue the troubleshooting process, wherein we had been told our hardware was too outdated to be supported.

Since leaving Minneapolis, it's just been too hot outside to be working on the roof, and we had a manual work-around. The very small handful of times it has been both cool and dry enough to be disassembling waveguides, we've been busy with other things or unable to go without Internet access for the roughly four hours or so the project might take. Yesterday was the magic confluence of perfect weather (I finished just before the rain started) and no other obligations; Louise was out running errands, so she would not miss the network access either.

I'm sorry to say that replacing the entire business end of the dish as a set did not fix the problem. I did not think it would, but it was the mandatory next step in the troubleshooting process. After the testing, I had to put the original electronics back on, because my feed arm does not have the correct mounting hardware for the newer electronics. Had the test worked, I'd be back on eBay looking for the right style arm. I'm back to square one at this point, with my next project being to put a sniffer on the net while the modem and positioner are doing their dance, then combing through the resulting capture to see where things are going wrong. What a pain.

While I was up there I also took the shroud off the bottom of the mount. For the last couple months the dish has made horrendous noises during the "stow" process, where the positioner brings the dish up to full elevation, and I could see the support arms starting to crack the shroud. Once I had it open, the problem was painfully obvious.

The dish has three low-power DC motors that move it into position along its three axes. Each axis has a fixed range of motion, but there are no limit switches or sensors to inform the positioner that the end of travel has been reached. Instead, the positioner is monitoring the motor current, and when it sees the current rise above a set threshold as the motor stalls out at the end of the travel, it stops. Inside the coach, you can hear the motors bog down each time they hit their stops.

This tends to put a lot of stress on various parts near the travel limits. Long-time readers may remember that five years ago a tooth sheared off the azimuth ring gear due to the stress of hitting the zero-limit, and we had to bring the whole dish back to MotoSat in Salt Lake to get it fixed (they will not sell parts, such as the ring gear). Fortunately at that time, they deemed the plastic ring gear to be a design problem, and replaced it with a metal one under warranty.

The current problem, pardon the pun, is much the same. There is a pot-metal casting mounted to the swiveling part of the elevation bearing that is supposed to act as a stop, hitting another pot metal casting at what is supposed to be the top of the elevation travel. While the casting is more than an inch thick, there is (stupidly) a screw pocket right at the point of maximum stress that leaves only a quarter inch of low-grade metal to take the strain. After seven years, the pot metal had enough and gave way. You can even see the screw threads visible through the crack in the metal.



I know from our experience five years ago, and again when we knocked the feed-arm off on a low overpass (long story, here) that MotoSat will not sell us this part, although I will call to ask tomorrow. Perhaps one of their installing dealers might be willing to order the part for me. I suspect, however, that what I will need to do instead is to find a machine shop someplace who can mill me one out of aluminum, and spend a night there while I disassemble the mount far enough to get the piece out -- it's sort of a donut and the elevation axle goes right through it.

In the meantime, I will now need to use manual motor controls to stow the dish as well as deploy it, since we are likely to do even more damage if we continue to let the motor stall out with the elevation supports stopping out on, as it turns out, the motor mount. Considering our dish is one of the original pre-production beta-test units, I suppose I can't complain too loudly that pieces are failing now after six years of hard use.

We both had a pile of unanswered emails and forum posts that had accumulated after we were deployed to Earl, and I think we are now mostly caught up. I also got my monthly column for Bus Conversions Magazine sent in just under the wire for the October issue. Yesterday afternoon blog reader Jay dropped by on his Gold Wing. He participates in a motorcycle drill team, and they apparently had an exhibition here in Rocky Mount yesterday morning. I'm sorry we did not learn about it ahead of time, as we would have gone to see it.

We are paid up through tomorrow, and we should find out in the morning if we will be heading up I-95 to Richmond. While Hurricane Igor, Tropical Depression 12, and two investigation areas are all spinning out in the Atlantic, none is threatening land in the next few days.