Saturday, I went out for a day on the bike. I parked the car at the start of the run, got the bike out, Got it ready, and set off. After a few hundred yards, I had a problem. When I gave the motor some throttle, it acted as a brake.
This can't be "wrong connections", I've been riding it with the connections as they are, for quite a while. But last week, I went through some serious water (three feet deep, we've been having a bit of weather in England, although we're trying not to let the foreigners find out, or they might not come to sunny England for their holidays), and although I waded through holding the motor wheel out of the water, it's possible it got wet inside. Or something? Anyone got any ideas?
So I went back home, put the bike in the garage, and put two fan heaters on it, one on the motor and one on the controller. My idea was to get them dry, without needing to dismantle either. I ran the two fan heaters (2kw and 3kw) off a long spool of extension cable. You can guess what's coming.
5 kw, 240 volts, so 20 amps, and the cable is rated for 13. Why did I do that? I was being stupid, it just didn't occur to me that I was overloading the cable. The cable heated up (when I felt it afterwards, it was very hot indeed), and that blew a contact breaker. Why didn't the 13 amp fuse in the extension cable blow? I don't know, possibly it realised it would do more damage if it didn't blow.
I didn't mention, I have about 30-40 computers running there. They're on UPSes, but I didn't know I had a power-out until the UPSes had exhausted.
So I had a burned-out extension cable (now retired) and 20 downed computers. I realised immediately what had happened, unplugged the extension cable and reset the contact breaker. And then one of the two UPSes cut out.
More investigation revealed that over the years, I've gradually been unplugging load from one of the UPSes, and plugging it into the other. That was working as long as the computers kept running, but they pull more power at start-up than when running, and that's why the UPS couldn't take it. So I rebalanced the loads, and started up the computers.
One of the computers wouldn't start up.
There's a bug in the Seagate firmware. http://www.msfn.org/board/topic/143880-seagate-barracuda-720011-read-me-first/
From what I understand, there's a small (0.5%) chance each time you start up a drive, that it will report zero capacity. And that's what happened here. And although I tried the usual remedy for this, I couldn't make it work. So I replaced that hard drive, and I'm reloading the data onto it (I do have a backup
) But that's about three terabytes, and it's going to take a couple of weeks.
Meanwhile, I put the fan heaters on again, but this time each one on a different mains supply, and not using a coiled extension, and with an inline ammeter so I know how many amps was running through. After some hours of this, I switched off the fans, and tried the motor wheel out with a battery. And it worked! But remember, when it failed before, it failed after a few hundred yards. So, I took the bike out for a test run, a mile or so. It performed perfectly.
So I took the bike back to the garage, put that battery on charge (it's two 4s hard cases in series, giving me 29.6 volts nominal to drive my 24 volt motor, and I charge it as 2 4s batteries in parallel), and replaced the casing that I use to waterproof the wiring just before it goes into the controller (I use a cut-down plastic oilcan, and duct tape). And when I had it taped up, I thought I should test the motor, so I connected my wiring harness (that puts the two 4s in series, puts a fuse inline, and leads out to a kettle plug that's used to connect to the bike). I connected the two 4s in series without bothering to unplug the paralleled balance leads, and just as I did it ...
If you've been following this, you know what comes next.
I have never seen so much blue smoke.
Blue smoke is, of course, intimately connected with electricity. Ineeded, electricity isn't, as many people think, a stream of electrons quantum-jumping their way along the copper. It is actually a stream of blue smoke, and when something goes badly wrong, the blue smoke trickles out ... or, as in this case, pours out. Gushes out.
I wish I could say that I had the presence of mind to disconnect immediately. But I didn't. I just stared stupidly at all the blue smoke wondering what had happened until plastic dripped, and copper melted, and it disconnected itself.
I opened up the garage to let all the blue smoke out into the wide world (if you live near London, you probably saw the column of smoke), and counted up the damage. I got lucky; one parallel balance cable is history, value $5, but the battery, the charger and everything else is fine.
So that's three mistakes I won't be making again.
This can't be "wrong connections", I've been riding it with the connections as they are, for quite a while. But last week, I went through some serious water (three feet deep, we've been having a bit of weather in England, although we're trying not to let the foreigners find out, or they might not come to sunny England for their holidays), and although I waded through holding the motor wheel out of the water, it's possible it got wet inside. Or something? Anyone got any ideas?
So I went back home, put the bike in the garage, and put two fan heaters on it, one on the motor and one on the controller. My idea was to get them dry, without needing to dismantle either. I ran the two fan heaters (2kw and 3kw) off a long spool of extension cable. You can guess what's coming.
5 kw, 240 volts, so 20 amps, and the cable is rated for 13. Why did I do that? I was being stupid, it just didn't occur to me that I was overloading the cable. The cable heated up (when I felt it afterwards, it was very hot indeed), and that blew a contact breaker. Why didn't the 13 amp fuse in the extension cable blow? I don't know, possibly it realised it would do more damage if it didn't blow.
I didn't mention, I have about 30-40 computers running there. They're on UPSes, but I didn't know I had a power-out until the UPSes had exhausted.
So I had a burned-out extension cable (now retired) and 20 downed computers. I realised immediately what had happened, unplugged the extension cable and reset the contact breaker. And then one of the two UPSes cut out.
More investigation revealed that over the years, I've gradually been unplugging load from one of the UPSes, and plugging it into the other. That was working as long as the computers kept running, but they pull more power at start-up than when running, and that's why the UPS couldn't take it. So I rebalanced the loads, and started up the computers.
One of the computers wouldn't start up.
There's a bug in the Seagate firmware. http://www.msfn.org/board/topic/143880-seagate-barracuda-720011-read-me-first/
From what I understand, there's a small (0.5%) chance each time you start up a drive, that it will report zero capacity. And that's what happened here. And although I tried the usual remedy for this, I couldn't make it work. So I replaced that hard drive, and I'm reloading the data onto it (I do have a backup
) But that's about three terabytes, and it's going to take a couple of weeks.Meanwhile, I put the fan heaters on again, but this time each one on a different mains supply, and not using a coiled extension, and with an inline ammeter so I know how many amps was running through. After some hours of this, I switched off the fans, and tried the motor wheel out with a battery. And it worked! But remember, when it failed before, it failed after a few hundred yards. So, I took the bike out for a test run, a mile or so. It performed perfectly.
So I took the bike back to the garage, put that battery on charge (it's two 4s hard cases in series, giving me 29.6 volts nominal to drive my 24 volt motor, and I charge it as 2 4s batteries in parallel), and replaced the casing that I use to waterproof the wiring just before it goes into the controller (I use a cut-down plastic oilcan, and duct tape). And when I had it taped up, I thought I should test the motor, so I connected my wiring harness (that puts the two 4s in series, puts a fuse inline, and leads out to a kettle plug that's used to connect to the bike). I connected the two 4s in series without bothering to unplug the paralleled balance leads, and just as I did it ...
If you've been following this, you know what comes next.
I have never seen so much blue smoke.
Blue smoke is, of course, intimately connected with electricity. Ineeded, electricity isn't, as many people think, a stream of electrons quantum-jumping their way along the copper. It is actually a stream of blue smoke, and when something goes badly wrong, the blue smoke trickles out ... or, as in this case, pours out. Gushes out.
I wish I could say that I had the presence of mind to disconnect immediately. But I didn't. I just stared stupidly at all the blue smoke wondering what had happened until plastic dripped, and copper melted, and it disconnected itself.
I opened up the garage to let all the blue smoke out into the wide world (if you live near London, you probably saw the column of smoke), and counted up the damage. I got lucky; one parallel balance cable is history, value $5, but the battery, the charger and everything else is fine.
So that's three mistakes I won't be making again.
)
