random reboots

[peppersass]

You may have read my other posts on this site and HB about slow pressure ramp-up and low flow rate problems on my first GS/3. After many unsuccessful attempts to resolve the problems, Chris Coffee replaced that machine with a brand-new GS/3 (my kudos and thanks to Chris.)

The water system of the new machine, serial number 613 manufactured 05/08, operates perfectly in every respect. I'm very pleased with the shots and the steaming. However, during initial setup and testing, I noticed that every now and then the machine spontaneously reboots. First the button lights go out, then a couple of seconds later the initialization message appears on the display, same as what you see when you power on the machine with the switch in the back. Maybe a second or two after the initialization message appears, the blue button lights come back on and all is normal. The delay is short when the boilers are already up to temperature, of course. In that sense, the reboot is harmless except for a few seconds of delay before I can use the machine (if it occurs during a shot, which hasn't happened yet, that would be another story.) One time, when testing the tea water functions, the tea water button flickered several times, then all the buttons went off, and the GS/3 rebooted. I've only seen that once. Usually, what I've seen is all the lights going out followed by the initialization message. Although the sequence looks like a power off/on, it's not exactly the same: I'm pretty sure the main display light never goes out. I say "pretty sure" because I've not actually caught it in the act yet. Anyway, if the main display isn't going off, the reboot probably isn't being caused by a major power interruption.

The machine rebooted at least half-a-dozen times during the initial setup and testing. Then it seemed to settle down and didn't do it again. The next day, after the machine had been on for several hours, I didn't see any reboots at all. Nonetheless, I reported the incident to Chris Coffee in case I need to make a warranty claim. One theory I discussed with Chris Coffee is that a little water dribbled into the control box during transport (the pump input port was left open and there was some water in the pump that may have found its way to the back of the machine.) After the machine had been on for a while, the water dried up and everything was OK.

Unfortunately, since that conversation the machine has rebooted maybe once or twice in five days. I say "maybe" because it may have rebooted without my seeing it. The sequence goes fast enough that it could be happening more often, but I'm missing it. Still, my sense is that it's not doing it as often as it did that first day.

But the problem remains -- I saw a reboot this afternoon. I was practicing steaming using dish soap, and I happened to notice that the button lights were off (the main display was lit.) Then I saw the initialization message. Then everything was back to normal. This was interesting because I hadn't touched any of the buttons, just the steam valve. Earlier I had thought maybe the machine has a defective display/button unit, but this latest reboot seems to argue against that. One thing that occurred to me is that the CPU may have been trying to fill the steam boiler after a low-level signal from the probe. It's conceivable there's a random problem doing that, either due to firmware or data corruption, or maybe a bad relay/solenoid causing the system voltage to drop low enough for the CPU to reboot. The latter seems unlikely because the problem is so intermittent, but it's certainly possible.

One other disturbing symptom: I was using the machine (can't remember what I was doing, but probably drawing tea water or flushing the group), and the pump motor went on and continued to run without any water coming out of the group or hot water wand. It just ran and ran. I didn't see an increase or decrease in steam boiler or brew boiler pressure, so I don't know where the water was going! Hitting the continuous brew or tea water buttons didn't stop it. Finally, I turned off the machine from the front panel and it stopped. Thankfully, I've only seen this once.

Troubleshooting:

When the problem first appeared, I took off the front panel and checked the cable connections. Chris Coffee had removed the panel to install the burn-me steam arm, so I thought maybe a cable hadn't been properly seated. I reseated the cables but that didn't eliminate the problem. After this last incident, I unplugged the machine, pulled out the control box and opened the cover. I didn't see any visible damage or water in there. I reseated all the cables and closed it back up. So far, no reboots, but I'm not convinced they're gone.

I think the next step will be to do a CPU reset (I've copied down all programmable parameters in preparation for that.) If that doesn't solve the problem, it seems to me there are two main branches of troubleshooting to pursue:

One would be to reload the firmware. It's possible that the firmware got corrupted when the machine was loaded with 1.13 (presumably by the US Distributor.) Being a computer/electronics geek, I can handle that. I've got a laptop with a parallel port and the cost of the programming interface is reasonable. Of course, it's always a little unnerving to do a firmware reload because if something goes wrong the machine could be disabled (depending on the bootloader design.)

The other main branch of troubleshooting would be to start replacing components, assuming my dealer and/or LM USA will be willing to supply replacements. My sense is that rather than replacing the CPU board or power supply, it would be best to replace the entire control box. Then if the problem continues, I'd replace the display/button unit. If the problem continues after that, then it's probably a bad/shorting probe, relay, solenoid or heater. That could be tricky to troubleshoot.

It seems to me that trying a firmware reset and reload should be done before attempting to replace hardware components, but a counter-argument is that replacing the control box replaces the firmware, too.

Has anyone seen symptoms like this? Does the proposed troubleshooting sequence sound right? Any suggestions?

[chasr]

Dick:

My GS/3 has never spontaneously rebooted either before or after I performed the firmware upgrade. It sounds like you have tried everything I probably would have tried. I have my doubts as to whether reloading the firmware will help but, at this point, it's worth a try.

I'm not sure what processor they use in the GS/3 but for an error to cause a reboot rather than just going off in the weeds is pretty sophisticated. It usually takes a specific sequence of events to cause a reboot rather than just a lockup.

[networkcrasher]

FWIW, the only time I've seen my GS/3 do a random reboot was when the keypad cable going into the control box wasn't properly seated. As it turns out, it's pretty easy to loosen it up when sliding the control box back away from the GS/3.

I've reloaded the firmware on my machine probably 10x when I was doing the thread on how to perform the upgrade, and if you follow the instructions, you have nothing to worry about.

I originally thought my reboot problems pertained to the fact I left the programming adapter attached and installed under the machine, but after noticing the cable that goes to the key pad being loose, I haven't had a problem since. The keypad ribbon cable is the shorter one closer to where all of the wires exit the control box.

[peppersass]

FWIW, the only time I've seen my GS/3 do a random reboot was when the keypad cable going into the control box wasn't properly seated. As it turns out, it's pretty easy to loosen it up when sliding the control box back away from the GS/3.

Thanks. That's very helpful.

The good news is that I haven't seen a random reboot or any other unusual behavior for at least 10 days, since I disassembled the control box, took out the PCB, swept it with a brush to get rid of little solder balls left over from manufacturing, made sure all the screws in the plugs and solid-state relays were tight, and firmly reseated all of the connectors.

It certainly could have been a loose keypad cable. That would make a lot of sense, as the reboots and odd behavior typically (but not always) followed doing something on the keypad. That said, in my first attempt at solving the problem at the control box, I reseated the PCB connectors of both cables leading to the front panel, and that didn't solve the problem (though it may have reduced the frequency.) It's possible the keypad cable got loosened again when I slid the control box back into the machine. Perhaps I got it reseated correctly the second time, or there was another loose connector in there causing the problem.

I never did a CPU reset, and haven't reloaded the firmware (but I have the loader adapter just in case.) Hopefully those steps won't be necessary. The machine has been working exceptionally well. Please knock on wood for me.