IcePick1957 Posted May 23, 2022 Posted May 23, 2022 (edited) I have a problem with the Pylontech UP2500 batteries. I only linked them up to my Inverter on Friday, 20/05/2022 for the first time. ( I was busy reinstalling all my solar panels, replacing all DC and AC cables as well as redoing\re-arranging my whole setup before installing the Pylons) No communication cable between batteries and Inverter were supplied either with Pylons or with the RCT AXPERT VM3 3KW 24VOLT inverter. I had to make a cable as supplied by members on the Power Forum. (Also tried cable as suggested by PylonTech themselves) Now it gives me an error, I mean a communication error on the Inverter, Error 61. The Pylon batteries also has LED warnings, like RED Protection LED flashing with SOC LED. (With and without comms cable) I have tried to reset them, changed master and back, no difference. Battery voltage is full. BatteryView also cannot connect. Refer to the Google drive link below with video footage and pics of battery info. (Files too big to email.) Batteries are only a temp setup as I'm waiting for the Battery rack to arrive. https://drive.google.com/drive/folders/1iNZy5FPm2uouu-3cs6d2rJk2jLf_eGEq?usp=sharing Anyone that can give me some advice here will be appreciated. Edited May 23, 2022 by IcePick1957 Quote
Coulomb Posted May 23, 2022 Posted May 23, 2022 (edited) 2 hours ago, IcePick1957 said: Now it gives me an error, I mean a communication error on the Inverter, Error 61. My guess is that you still haven't gotten the cable correct. Check for plugging into the right port at both ends, double check the cable pinout and orientation (if not symmetrical), and check DIP switches on the Pylontechs. Finally, Pylontech seem to change the details on various models. Your 3 kW VM III 24 V should be able to communicate with Pylontechs, at least in theory. The firmware definitely checks for VM III versus King, and 48 V versus 24 V all over the place. 24 V models seem to be less popular, and connection to Pylontechs seems less common, do it's possible that there are bugs. But you'd think it would at least connect. Finally, check your display firmware. Early versions had BMS issues. Removable display firmware version 02.66 is available, and 02.70 is available from this post. Unfortunately, you might need updated main firmware to go with the updated display firmware, and I don't offhand know how available that is. After all the above, the next step unfortunately would be to use RS-485 hardware to sniff the data between the inverter and BMS to see what's happening. Just checking for traffic in each direction with a multimeter would be a start. Of course, not everyone is able to do this sort of thing, and should not have to. Edited May 23, 2022 by Coulomb Quote
IcePick1957 Posted May 23, 2022 Author Posted May 23, 2022 @Coulomb I have tried various pin outs and still give the problem. I managed to get Batteryview running. It gives error 256, which is address allocation error on both batteries. Both batteries are totally disconnected from everything and running stand alone. Each one has the address 20. Mean anything? Can I change this? Is their a way to reset it. Quote
Coulomb Posted May 24, 2022 Posted May 24, 2022 9 hours ago, IcePick1957 said: It gives error 256, which is address allocation error on both batteries. That sounds like a big clue. I don't own a Pylontech, so I don't know how to fix this. With just two battery modules, I believe that you need the same address (same group), but maybe it needs to be address 0. Pylontech owners may be able to help. Quote
IcePick1957 Posted May 24, 2022 Author Posted May 24, 2022 6 hours ago, Coulomb said: That sounds like a big clue. I don't own a Pylontech, so I don't know how to fix this. With just two battery modules, I believe that you need the same address (same group), but maybe it needs to be address 0. Pylontech owners may be able to help. I think the firmware software is screwed up. The slave shows drawing 130A from it when master says only 5A. (Gives a Over Current Alarm(Software) but Alarm LED on battery is off. WatchPower shows 12A. (Charging is the same story) Cell temperature is 18 degrees Celsius, cables and batteries are cold on the touch. BTW, if that was really the case, the 125amp fuse/breaker would have blown already. I have Steve looking at this as he sold me the batteries. (Also reported to PYLONTECH) Maybe a re-flash of the firmware will sort this out. Quote
Coulomb Posted May 24, 2022 Posted May 24, 2022 9 minutes ago, IcePick1957 said: The slave shows drawing 130A from it when master says only 5A. That sounds like a problem with the current sharing cables. They're the red and black ones lightly twisted, with the green connectors at each end. Check for continuity and lack of short circuit in these cables. What removable display firmware (U2) are you running? Quote
IcePick1957 Posted May 24, 2022 Author Posted May 24, 2022 1 minute ago, Coulomb said: That sounds like a problem with the current sharing cables. They're the red and black ones lightly twisted, with the green connectors at each end. Check for continuity and lack of short circuit in these cables. What removable display firmware (U2) are you running? Hmmm, I don't have cables like that. Only the orange and black. Parallel cables between them and the main POS and NEG cables to the Inverter. (The link between them is a RS485 Cable via the RS485 Ports.) U2 version is 2.61, Use to be 2.66 but reverted back to see if this sorted it out. (Refer my Post under downloads) I think I will update to 2.66 again tonight. (I don't trust the 2.70 as mentioned in another post. Quote
Coulomb Posted May 24, 2022 Posted May 24, 2022 1 hour ago, IcePick1957 said: Hmmm, I don't have cables like that. Only the orange and black. Not between batteries, I meant between inverters, as part of the paralleling kit. Bottom right in the photo below: These cables enable the master to read the current share from each of the paralleled inverters, and somehow directs them all to share the current about equally. If you don't have these cables, then that's your problem. Quote
IcePick1957 Posted May 24, 2022 Author Posted May 24, 2022 (edited) I only have 1 inverter, so no need for those cables. (One less problem.) I'm just waiting for Steve to contact me as he is on the road back to Cape Town. Batteryview data on the cells of the culprit battery: Only cell 1 is lower than the others, otherwise they are well balanced. Strange thing is, it appears that the UP2500's have Phantom series batteries main boards installed. See this from Batteryview: Device address : 20 Manufacturer : PYLON Device name : UP2500 Board version : PHANTOMSAV10R04 Main Soft version : B66.28 Soft version : V1.1 Boot version : V2.0 Comm version : V2.0 Release Date : 21-07-13 Specification : 24V/111AH Cell Number : 8 Max Dischg Curr : -100000mA Max Charge Curr : 100000mA EPONPort rate : 1200 Console Port rate : 115200 Edited May 24, 2022 by IcePick1957 Quote
Coulomb Posted May 24, 2022 Posted May 24, 2022 (edited) 2 hours ago, IcePick1957 said: The slave shows drawing 130A from it when master says only 5A. 21 minutes ago, IcePick1957 said: I only have 1 inverter, so no need for those cables. (One less problem.) Oh, sorry! With slave and master, I assumed you meant inverters. You meant battery modules. There is nothing that the inverter can do in firmware that can cause the battery current to be unbalanced; that's simply a matter of battery cabling. Are you using the "diagonal takeoff" wiring pattern? Not "daisy chaining"? Edited May 24, 2022 by Coulomb Quote
IcePick1957 Posted May 24, 2022 Author Posted May 24, 2022 (edited) 12 minutes ago, Coulomb said: Are you using the "diagonal takeoff" wiring pattern? Not "daisy chaining"? Like this? Better this way? I don't see anything different in operational sense... ? Edited May 24, 2022 by IcePick1957 Quote
Coulomb Posted May 25, 2022 Posted May 25, 2022 17 hours ago, IcePick1957 said: Better this way? I don't see anything different in operational sense... ? Yes. That way, considering for now only two modules, no one module has a shorter path to the inverter, because the one with the shorter path will take the lion's share of the current, because of the resistance of the cables and connections. With 3 or more modules, perfect paralleling isn't possible using the provided links. But using the terminals at the diagonals is still much better than using terminals at one end. Quote
Beat Posted May 25, 2022 Posted May 25, 2022 5 hours ago, Coulomb said: because the one with the shorter path will take the lion's share of the current, because of the resistance of the cables and connections. I'm still troubled with that idea. I maintain that differences in internal resistances (Ri) might be higher by factors than the resistance of some cm of 25mm² cables. I lately added a 4th 100Ah pack and wired it at the end of the other packs. Although it has the longest leads to the inverter I observe it takes the highest current with heavy load. This might be due to being the newest with lowest Ri and possibly improved manufacturing. Although it carries exactly the same model designation as the others, it has slightly more capacity according to BMS. When it comes to the floating (equalizing) state, the currents are relatively low such that differences in lead lengths become insignificant. Quote
Coulomb Posted May 25, 2022 Posted May 25, 2022 1 hour ago, Beat said: I maintain that differences in internal resistances (Ri) might be higher by factors than the resistance of some cm of 25mm² cables. Well, I'm assuming that the internal resistance of the modules in parallel are about the same. As they would be most of the time. 1 hour ago, Beat said: I lately added a 4th 100Ah pack and wired it at the end of the other packs. Although it has the longest leads to the inverter I observe it takes the highest current with heavy load. This might be due to being the newest with lowest Ri... I'm a little surprised by that finding. But I can't argue with what you found. Quote
IcePick1957 Posted May 25, 2022 Author Posted May 25, 2022 (edited) I have noticed, after using BatteryView, that the slave gets "Over Current Alarms"(as high as 130A). It is also charging and discharging all the time. I assume that it's trying to balance the cells, but the cells differ with about 13 - 15 ma. The temperature stays under 20 deg Celsius which tells me that the high discharge and charge currents are incorrect. Current drawn from AC Power is around 5 Amps. (Should be around 28 Amps AC if it really charging at 130+ Amps It also shows that all voltages and temperatures are normal. I think either the firmware is corrupted or the BMS hardware is faulty. The slave seems to be doing all the work while the master idles most of the time. Maybe the master is also not working 100%. When master running alone the Inverter shutdowns with Low Bus Voltage error. (That's with AC Power switched off ) I get the BatteryView "System error 256" and "ADDR Allocation ERR" on both batteries permanently. And it also appears that the batteries do not communicate. I'm trying to find an explanation and fix for the above error, but nothing comes up 100%. I have logged a call with PylonTech and my supplier. No response from both of them. I need a solution/fix quick, because I need to deal with EKSDOM loadshedding.. Edited May 25, 2022 by IcePick1957 Quote
Beat Posted May 26, 2022 Posted May 26, 2022 (edited) On 2022/05/25 at 1:58 PM, Coulomb said: Well, I'm assuming that the internal resistance of the modules in parallel are about the same. As they would be most of the time. That's a risky assumption. We deal with manufacturing inaccuracies. The variations may be different from one manufacturer to another. My batteries are Loech. Make the following measurements on each pack with the help of the BMS: Prepare a heavy load (i.e. kettle) to be switched on and off. Without PV input in battery mode note pack voltage (U) and current (I). Turn the load on and note again voltage and current. Calculate the difference (d) between each currents (dI) and voltages (dU). Then with Ohms law calculate Ri = dU/dI. Do this on every battery pack. Let us know your findings. You could also do the calculation on an Excel sheet. Edited May 26, 2022 by Beat Coulomb 1 Quote
[email protected] Posted April 26, 2023 Posted April 26, 2023 (edited) On 2022/05/25 at 3:56 PM, IcePick1957 said: I have noticed, after using BatteryView, that the slave gets "Over Current Alarms"(as high as 130A). It is also charging and discharging all the time. I assume that it's trying to balance the cells, but the cells differ with about 13 - 15 ma. The temperature stays under 20 deg Celsius which tells me that the high discharge and charge currents are incorrect. Current drawn from AC Power is around 5 Amps. (Should be around 28 Amps AC if it really charging at 130+ Amps It also shows that all voltages and temperatures are normal. I think either the firmware is corrupted or the BMS hardware is faulty. The slave seems to be doing all the work while the master idles most of the time. Maybe the master is also not working 100%. When master running alone the Inverter shutdowns with Low Bus Voltage error. (That's with AC Power switched off ) I get the BatteryView "System error 256" and "ADDR Allocation ERR" on both batteries permanently. And it also appears that the batteries do not communicate. I'm trying to find an explanation and fix for the above error, but nothing comes up 100%. I have logged a call with PylonTech and my supplier. No response from both of them. I need a solution/fix quick, because I need to deal with EKSDOM loadshedding.. HI, did you find a solution to the "ADDR Allocation ERR", I have the same on my UP5000. Edited April 26, 2023 by [email protected] Quote
[email protected] Posted April 27, 2023 Posted April 27, 2023 17 hours ago, [email protected] said: did you find a solution to the "ADDR Allocation ERR", I have the same on my UP5000. I found the problem, seems like if the DC cables are disconnected from inverter it will give red alarm and the above error on batteryview. Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.