Well 1849 went 15 days with no wdt resets, but today it did. Do you want the the debug info?
Yes ! You bet !
Thanks !
I forgot to record this yesterday when it reset. Today it reads:
Classic 150V (rev 4)
MAC: 60:1D:0F:00:1B:95
Device ID: 16F0 BE2
IP: 192.168.0.223:502
Firmware:
- Classic Rev: 1849
- Network Rev: 1839
Absorb Time:03:30:00
Equalize Time:01:00:00
Last Voc:85.8
DB1: 0x369cebaf
DB2: 0x-d5c61f
DB3: 0x6a4b1472
DB4: 0x-130ce73a
Unfortunately, we still need DB5 added to that list of debug registers which is what addresses the WDT reset issue.
Ill keep an eye out for it next time and save you a complete modbus dump.
Yeah, sorry about that -- my bad. I've got the register in the code, I'm just not displaying it. I've got a Local App update once I've finished release testing.
And I think that Andrew just released a new app/status panel that has this Debug register it it.
Mine just updated.
Thank you, Andrew !!!!!
Your in luck it reset again just now, uptime 3 days.
Absorb Time:03:29:59
Equalize Time:01:00:00
Last Voc:87
DB1: 0x369cebaf
DB2: 0x-d5c61f
DB3: 0x6a4b1472
DB4: 0x-130ce73a
DB5: 0x0
Full regsiter set here (snapshot about 2minutes after reset).
www.zoneblue.org/files/2014-06-13.zip
It also (interestingly) did that thing where the Classic Amps reading (incorrectly) reduces about 2Amps, about an hour before. Rebooted reading right.
That zip includes a spreadsheet of the 1s data.
Closeup thus. Dont know if the two phenomona are related. But hey, Kill two birds.
Quote from: zoneblue on June 12, 2014, 07:41:29 PM
Closeup thus. Dont know if the two phenomona are related. But hey, Kill two birds.
what were you logging here ?
Also from DB regs, looks like data abort and not WDT reset.
Reg 4275:104 tho...
Same as the spreadsheet, classic amps in blue, Vpv in pink.
Just in case anyone thinks this is all acedemic, these two resets occured during one of the worst spells of weather i can remember. Over 4 days of uniformly wet and dull conditions. The bank got down to half... first time ive seen the bank below 24v!
By WDT I am assuming you mean Watch Dog Timer.
Just for those who may not know the acronym.
Tom
Quote from: zoneblue on June 12, 2014, 07:02:40 PM
Your in luck it reset again just now, uptime 3 days.
Absorb Time:03:29:59
Equalize Time:01:00:00
Last Voc:87
DB1: 0x369cebaf
DB2: 0x-d5c61f
DB3: 0x6a4b1472
DB4: 0x-130ce73a
DB5: 0x0
Full regsiter set here (snapshot about 2minutes after reset).
www.zoneblue.org/files/2014-06-13.zip
What is interesting here is that DB5 shows zero but looking at your "register" dump (registers as opposed to addresses)
shows a non-zero and believable address of 0x20bb4 (regs 4387:2996, 4388:2) but unfortunately doesn't really
give much of a clue except that the Classic was just executing regular application code and not stuck in an EEprom
read/write or something interesting.
I do not think that the other DB's are relevant to this particular reset because the fault address lies WAY outside the
flash range of this processor at somewhere around 10 megabytes.
Please keep up the GREAT forensics if you happen upon this one again !
thanks !
Im curious if everyone is experiencing this level of resets, or is my system unusal?
Quote from: zoneblue on June 27, 2014, 06:55:25 PM
Im curious if everyone is experiencing this level of resets, or is my system unusal?
I have noticed an irregular pattern of resets on my c150. With the recent variable weather here, when rapid changes occur jumping from low to high wattage output then sometimes a reset will happen. I often only realize this because the energy count goes back to zero.
When I have been watching the Classic and this reset happens its always when the Classic is working normally then instant reset.
This makes me think its nothing to do with a watchdog reset but is just a fatal bug in the code that crashes the Classic immediately.
I have not bothered reporting this as others, you, are working on tracking down this problem.
dgd
thanks dgd. im trying to rule out what extra load my logging system might be putting on the classic. I cant really get clear in my mind when these resets started but it was definately post 1760. However i also increased the blackbox sample rate around that time.. i think?
I have seen were kwh’s go back to 0 or start over for no rhyme or reason, mostly cloudy days going from low to high output , I've also had a few lockups, by that I mean the network connection quits responding, a power down power up of the C150 fixes it.
ZoneBlue, (and others), does the Classic reset maybe when it is HOT ?
When was your Classic made or its serial number ?
boB
I have not notice it getting hot.
CL02001
This ones not in service yet but do keep the firmware up to date.
CL02392
4037. System here doesnt run hot, 15C ambient, 40A throughput. Fan comes on very little. When 1760 came out the resets were mostly early morning or evening. These days i would say the opposite: mid bulk or so.
Ok, so i looked at it closer, every reset is during bulk charge stage. Time varys, temp varys.