Jump to content

Pylontech batteries went quiet


ChristoSnake

Recommended Posts

Hello all..

We had a very severe thunderstorm yesterday afternoon, and lost a number of electrical goodies as a result (home alarm, AV amplifier, Ethernet switch, WiFi extenders, security camera, etc.).  But the worst of it all is that my 4x Pylontech batteries have decided to stop talking to their beloved ICC Raspberry Pi:

image.thumb.png.412259f4a5dd9d397ca796cc93dc0546.png

I shut the inverter down and started it from the batteries earlier today, but the problem remains.  Judging from the PV generated today they still seem to be functional (typical base load usage with full batteries at the moment, but the PV total suggests there was a lot of charging going on this morning).

Where do I start troubleshooting?  New Pylontech cables are quite expensive, so I'd prefer some cheaper suggestions before going that route!

Thanks in advance...

 

 

Link to comment
Share on other sites

2 hours ago, ChristoSnake said:

(home alarm, AV amplifier, Ethernet switch, WiFi extenders, security camera, etc.).  But the worst of it all is that my 4x Pylontech batteries have decided to stop talking to their beloved ICC Raspberry Pi:

Sorry to here that Snake

IMo you have two things that might be blown. My guess would be that the pi might have taken the hardest knock. Inside your Pylon cable is an Arduino, or if I remember correctly there use to be. With the router gone as well the chances of the spike reaching the PI is good, then it needs to find its way to the Arduino , but I dont think it will pass the arduino as well. 

You don't by any chance have a spare PI that you can test first...

Link to comment
Share on other sites

I was wondering about the bulge in the Pylontech cable, but it may well be the hidden Arduino!  That also explains the cost of the cable.

The Pi seems mostly alive - it's reading threads from the inverter and posting to both emoncms & PVoutput.  But I just realised that the inverter started borrowing from the grid as the PV decreased this afternoon because according to the ICC Pylontech thread there were no batteries available at all 🙄

I ending up disabling the Pylontech thread in ICC so that the inverter reads the battery values directly and it seems to have solved the problem.  Now I'll have to adjust my cutoff values all over again, because the inverter sees a somewhat lower value than what the Pylontech cable was feeding it.

I have some colleagues with similar installations (Pi & Pylontech) - I'll borrow some bits & pieces before I go on a spending spree. 

Edited by ChristoSnake
Clarified post somewhat...
Link to comment
Share on other sites

On 2020/01/04 at 6:02 PM, ChristoSnake said:

I was wondering about the bulge in the Pylontech cable, but it may well be the hidden Arduino!  That also explains the cost of the cable.

I'd hope that they used a stripped-down version of it and not a complete branded "arduino" module. All you need is an ATMega 325 chip (or one of the smaller ATtiny ones), a crystal for the clock frequency (and two ceramic capacitors to load it), and sometimes a 5V regulator (unless you can get 5V directly from USB... then not needed). And you can even get away without the crystal and stuff if an accurate clock speed is not important. Aaaah, and you need a USB-serial chip too if you want to go to USB, but those are dirt cheap now.

Link to comment
Share on other sites

I got a new cable from Riaan this afternoon and he confirmed that there's no Arduino inside.

Much cable swapping, ICC restarting & Pi rebooting later I concluded that the old Pylontech cable was indeed a goner, and that the Comms port on my master Pylontech was also dead.  All the USB ports on the Pi still seem fine.

The Pi still reports a battery error when I connect the new Pylontech cable to the first (master) battery's Comms port.  When I connect it to any of the other (slave) battery Comm ports, it is establishes comms and the error disappears.  But the other batteries are slaves, so they do not have any information on the battery pack.  Due to the very short Pylontech RS485 cables I will have turn all the batteries in the rack off, physically swap the current master with the slave below it in the rack, then restart all the batteries to get the new master appointed.

I'll do that later in the week - Monday has eaten all my energy 😪

Link to comment
Share on other sites

3 hours ago, JacquesVDM said:

The "bulge" in the Pylontech cable is just a normal USB to Rs232 converter. (No Arduino) 

Aaah okay then. Of course, I forgot that that cable snaps into the "console" port instead of the more common CAN-bus connection used by other setups. So RS232 is all you need.

Link to comment
Share on other sites

On 2020/01/04 at 3:24 PM, ChristoSnake said:

Hello all..

We had a very severe thunderstorm yesterday afternoon, and lost a number of electrical goodies as a result (home alarm, AV amplifier, Ethernet switch, WiFi extenders, security camera, etc.).  But the worst of it all is that my 4x Pylontech batteries have decided to stop talking to their beloved ICC Raspberry Pi:

 

Damn...These electrical storms seem to be more vicious this year. A lightning strike very close to our house resulted in similar damage to yours.

Besides some electrical appliances....looks like my monitoring laptop, VE direct cable, serial to usb cable, inverter comms board, BMV702, and one of my 48v 100Ah LiFePO4 has popped.

Link to comment
Share on other sites

13 hours ago, JacquesVDM said:

The "bulge" in the Pylontech cable is just a normal USB to Rs232 converter. (No Arduino) 

Hi Jacques, the first ones were build with a small arduino in a little box filled with some resin like substance, that is why I said It use to be. When He mentioned bulge I thought that they might have found another way of doing it. 

Link to comment
Share on other sites

The Pylontech -> Pi comms cable was bust, but I got a replacement from Riaan for a very fair price.

Sadly, the "Comms" port on the old master battery and the two "Link" ports are also fried.  As a result, the battery refuses to act as master and will not start the slaves when you press the red "SW" button.  When it's connected as a slave, it triggers its error LED and also interferes with the comms of the remainder of the battery group (no readings sent to the Pi.

I've configured the next healthy battery in the rack as a new master, and it talks to the other two batteries and the Pi/inverter as it should.  The broken ex-master battery remains connected to the inverter, but it now works in unmanaged mode.  I presume that its internal BMS is still fully functional (judged by the typical activity of its charging LEDs), so I am still benefiting from its >3kWh until the insurance can decide how to proceed with my claim...

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...