September 13, 20223 yr I have been following The Off Grid Garage since Andy started his channel and now want to put some of his observations to the test myself. I have 2 Narada NPFC100 batteries, they are pretty awesome and deliver the advertised power and specs without fail. Work great with my Sunsynk5.5 with NO battery to inverter coms at all. I will sacrifice the warranty on the one battery to retrofit the JK bms to it. ( I am not too impressed with the shinwa bms's on these batteries.) To document this I will post some pics and MAYBE a few videos when I start with this testing. I have ordered 2 of them, one .6amp and one 2amp (for comparison) Just waiting for the BMS's to be delivered from china, hopefully this week still (1st one is in JHB already)
October 1, 20223 yr Author Quick Update On The JK BMS Retrofit 😀 Will post photos of the actual retrofit later. Have figured out how to get the data out of the JK BMS into my nodered via mqtt and then into my thingsboard portal for visualisation. Still busy hacking the JK BMS modbus records, so not all available for display yet. Enough data to already show how out of wack the cell balancing is.
October 1, 20223 yr Author The new JK BMS retrofit to my 1 Narada Battery is already paying off, cells are starting to pull back towards each other after just 4 hours and no need to disassemble the battery to do a full parallel top balance. 🤩 Edited October 1, 20223 yr by WannabeSolarSparky
October 2, 20223 yr Author Now to slowly start charging the battery to full so that I can get a nice even top-balance.
October 2, 20223 yr Author Ok, Top balancing without parallel connections or disassembly. As all the suggestions on all the different forums I am going with a low 5 amp charge current, so this may take a few days (Cells still out of balance at the top end) to get perfectly top-balanced using just the bms and solar from the inverter for providing the charge current. Let's see how it goes 🙃
October 2, 20223 yr Author Slowly getting there 15 cells started with all totally out of balance (Crap shinwa bms) already 9 of the 15 cells perfectly in balance, remaining 6 slowly working their way up. Charging balancing cycle currently at about 60 minutes, need to adjust the BMS settings to make it cycle a bit quicker, but even at this rate the cells should top balance out perfectly in a few days. And if all the tests Andy at Off Grid Garage did then once it's balanced it should stay that way for a long time Edited October 2, 20223 yr by WannabeSolarSparky
October 2, 20223 yr Author Today's charging/absorption cycle has completed, 4 cells still out of balance by about 100 mV. Will see how it goes tomorrow when the next charge cycle runs.
October 3, 20223 yr Author Sun has finally come out in Cape Town Balance charging for the day has started again. Let's see if this JK BMS can keep the cells balanced all the way to the top. Current deviation less than 10 mV, lets see how that holds when the cells start saturating.
October 3, 20223 yr Author I did not discharge from this battery during the night so the flat part of the charge curve has been reached quickly. According to the JK BMS there is about 22AH left to absorb, most of this will be on the flat side of the charge curve where it is now. Going to be a long day as I have the JK BMS set to only charge at max 2 amps to give it a fighting chance to keep the cells balanced all the way to the top
October 5, 20223 yr Hi WannabeSolarSparky. Looks like this is turning out to be a successful retrofit. I'm very interested to find out about this BMS, where you sourced it from and how much it cost you. The BMS of my Narada NPFC100 of 2 years has failed due to overload 2 weeks back. If I can get an alternate BMS then I'll consider it.
October 30, 20223 yr Author If you look further up in the thread you will have seen some screenshots of the initial data traces after I installed the new JKBMS. Offhand at a glance everything looked ok. But damn I was in for a surprise. No matter how hard I tried to do all the settings correctly and do full charge discharge cycles the graphs just would not balance out. Turns out cell no 8 of 15 was totally messed up. Full pack was only managing 30ah on charge discharge, All this time I thought the 2 Narada's were performing quite well. But is was in fact the healthy one that was nicely working propping up the faulty one to such an extent that I did not really notice the problem. This once again highlights the importance of having FULL ACCESS to the BMS so that you can get a full detailed view of what is happening inside the battery with individual cells. Shame on the battery suppliers who have/give all sorts of excuses why they do not provide the full monitoring software to the end users. Ordered a pair (one extra for spare) of "new grade A" cell's - 100ah lifepo4 cells from china, still waiting, they will most likely only get here in the new year 🤣 Don't be fooled by all the marketing with these pre-built batteries the "new grade A" cells are just sales 99% of them ALL use second life batteries. They may be grade A but they sure as hell most of the time are second life, if they were brand new from e.g. the EVE factory then those battery packs would cost a fortune. So while waiting for the 2 new cells from China I had to source a battery locally, that was a mission as the stock gets bought as quick as they get listed on any of the legit sites around SA. I eventually managed to get two 50AH from takealot even though at a premium of R1199 for the pair Received it on Thursday, well packaged and no bulging at all and a decent shipping voltage of 3.35v and 0.6mΩ I have temporarily added these cells as a 2p replacement for the faulty cell until the new 100AH cells arrive from China. On Saturday morning I switched the battery on again to see what happens and if I can get the full pack to finally balance correctly using just the JKBMS. WELL - So far so good 😃 Pack is busy doing a full balance using only the BMS and it is going surprisingly quick om lifepo4 terms. Started Saturday at about 11am and you can see the progress so far. All the cells are starting to pull up nicely one by one. At the current rate of balancing the pack should be fully balanced by this evening sometime. Then I can do a few full cycle charge/discharges to see how the pack performs. Edited October 30, 20223 yr by WannabeSolarSparky
October 30, 20223 yr Author Balancing completed took 24 hours to 3.52v across all cells with less than 5 mV deviation 😀 Now to give it a few hours to fully absorb at that voltage and then run a full discharge/charge cycle to see if the JKBMS balancing holds nicely as predicted.
October 30, 20223 yr Author Geyser beat me to testing the JKBMS working the balancing function 🤣 Will need to disconnect all loads except the constant test load to do the full discharge/charge cycle.
October 30, 20223 yr Author 1st Discharge/charge cycle started after fully balanced and absorbed. Running a constant 12.9 amp-DC load till lowest cell battery voltage gets to 3v or the inverters cut-off voltage of 46.5v whichever hits 1st. That should then theoretically give the (normal use) usable capacity of the full battery pack. From the Narada spec sheet for this battery model I should get close to 6.5 hours runtime till my set cut-off. By my (Not So Hot Arithmetic) I should get close to 95AH Edited October 30, 20223 yr by WannabeSolarSparky
October 30, 20223 yr Author About halfway through my Not So Scientific Capacity Test 🤣 So far maintaining average DC Amp load 13.2 A and 3 hours and 20 minutes into the test. Maybe during the week I will try a proper constant current load capacity test. But on this current 1st test so far looking good, voltage on the battery has dropped less than half a volt up till now and the delta V has not moved much, the JKBMS does a pretty good job doing discharge balancing as well. Granted, we still in the flat part of the curve, will see closer to the bottom final stages what happens to the voltage deviation. That surge of usage was my wife forgetting I was doing a test and used the microwave for a few minutes. Edited October 30, 20223 yr by WannabeSolarSparky
October 30, 20223 yr Author 60 AH consumed so far (13.12A DC X 4.5 hours) Looking good so far and voltage still sit nice and steady across all the cells at 3.23v and less than 7mv deviation.
October 30, 20223 yr Author 6 hours done on the not so scientific capacity test 79 AH consumed so far (13.29A DC x 6 hours) Voltage deviation has started to creep in as we approach the lower end of the discharge curve, currently at 25mV voltage still holding steady, now down to 3.195v per cell Inverter low voltage warning set at 3.10v Inverter cut-off set at 3.07v so still some way to go
October 30, 20223 yr Author Result of 1st full discharge is in 🤩 Not a perfect capacity test but definitely a good indication of where the pack is with the temporary cell replacement. Also a pretty good test to see how the JK BMS works and behaves. 7 hours 5 minutes, discharge done on the not so scientific capacity test 94 AH consumed by end of discharge test (13.3A DC x 7.1 hours) Voltage deviation at end of discharge test 144mV Voltage at end of discharge at 3.118v average per cell Inverter low voltage warning set at 3.10v not reached I could most likely still have gone down further to about 46V on the pack, would give maybe an extra 2AH but it's been a long day Time to absorb the data observations and get started with a full charge back up to fully balanced and saturated and then compare the results. Also see second image of the live raw mqtt data stream at the point of ending the test Raw data from MQTT Stream direct from the JK BMS Edited October 30, 20223 yr by WannabeSolarSparky
November 1, 20223 yr Author Did some usage calculations for last nights full cycle. Ran my small crypto rig all night to push up the usage and ran both batteries connected parallel as normal. Total runtime from start of discharge to end of discharge came in at 13.5 hours. Total Average DC Amp draw - 11.89 A Total AH consumed - 160 AH Final BANK Voltage reported by inverter at end of discharge - 48.35V SOC Calculated by inverter 36% Battery A (JK BMS Retrofit) Average DC Amp Draw - 5.61 A Total AH Consumed - 75.7 AH Total Balance deviation at end of discharge cycle - 21 mV Voltage at end of discharge - 48.324V SOC reported by bms - 20% Battery B (Narada BMS) Average DC Amp Draw - 6.28 Amps Total AH Consumed - 84.78 AH Total Balance deviation at end of discharge cycle - Unknown Voltage at end of discharge - 48.237V SOC - unknown The results are pretty close to what I expected, considering that Battery B is only a year old with presumed healthy cells and Battery B has a temporary "mc-guyver" bad cell replacement. I will do another full charge and absorb and the do another night time run test but this time I will switch off the solar and let it run all night and into the day till the BMSes shutdown the batteries.
November 11, 20223 yr Hello guys, I have two Narada 48NPFC100 which are connected to the Growatt SPF 5000 ES inverter. I have checked the parameters in the manual for Narada. However I need some help with the following parameters to ensure prolonged life and a healthy battery, would you please provide me with the recommended for the following parameters on the Growatt: Setting voltage point back to utility source when selecting SBU (program 12) Setting voltage point back to battery mode when selecting “SBU priority” (program 13) C.V charging voltage (program 19) Floating charging voltage (program 20) Low DC cut-off voltage (program 21) Thank you in advance!
July 23, 20241 yr What model of JK have you installed btw ? Did you end up using the one with 0.6A or 2A balancer ? Noticed any significant difference? Edited July 23, 20241 yr by zivva
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.