June 27, 20206 yr Hi All, I've wanted to post many times but I don't think I have - so firstly thanks for a great forum, I love reading about how passionate you guys are! Just as an intro - I'm an IT guy with electronics as a hobby. I don't mind tinkering as a rule I'm off the grid, I have about 5k of panels, Axpert 5kVa inverter (stock standard, not flashed). Started with LA 4 years ago but then about 9 months ago installed a Revov 10.2kWh pack with ICC. My home is automated using DIY stuff - Domoticz & sonoffs (on Tasmota) basically. Siri is integrated so even my wife loves it! I don't mind sharing any experience, ask any time. Now my issue - my Revov pack doesn't seem to deliver on it's potential. Especially now in winter, I run out of power early in the evening. I don't think it's because it's due to the reduced hours, I think it's because the pack is not loaded fully when there is sun. So I did a test - I waited until the batteries were "fully charged", ICC showed PV was "cut off", at that stage at 54.2V. Then I disconnected the panels and waited for the power to cut off. ICC shows this was 2.2kWh later. A bit disappointing I think. I called my installer and he suggested I add a Victron device (not sure what it was) as ICC and the Revov BMS seem to not be on the same page. I just don't have any money to spend, so I stopped listening at "Victron", sorry. I obviously want to use my current equipment, the Axpert has been fantastic and I've had no issues at all. So he suggested I raise the float/bulk to 56.1/56.2 V, to see how it performs. Initially they were 54.5/55.5. I don't like the "let's see what happens", hence my seeking advice. Has anybody done a similar install? Or even without that - any suggestions? Not sure what data I should be posting here - I'm sure there's "things" on ICC that I should be providing, please tell me and I'll deliver. Thanks in advance!
June 27, 20206 yr Measuring the capacity of any battery is a challenge. You need an Ah meter like this: https://www.victronenergy.com/battery-monitors/bmv-700 Otherwise you need to provide a constant load. This is also a challenge since as the voltage discharges so the current drops which isn't what you want.. You can build one: https://www.instructables.com/id/DIY-Adjustable-Constant-Load-Current-Power/ or what I built which works fine for my tests..https://www.instructables.com/id/DIY-Constant-current-load/
June 27, 20206 yr 5 hours ago, Coen said: I waited until the batteries were "fully charged", ICC showed PV was "cut off", at that stage at 54.2V. What do you mean by "fully charged"? If you're running ICC with no Victron BMV [ edit: and if ICC can't talk to the battery BMS directly ], then I believe ICC has to rely on the inverter's guess, which is based on lead acid thinking. I'll guess that your low voltage cutoff is set to 44.0V (which is really low); the Axpert calls that voltage -2% SOC, and adds 1% SOC for every tenth of a volt above the setting plus 0.2V, reaching 100% at 44.0 + 0.2 + (100*0.1) = 54.2 V. But that's meaningless; a 16S (I'm guessing) LFP is nowhere near full at that voltage; it needs to get over about 55.0 V; my bulk/absorb charge voltage is set at 55.2 V (and I also have 16S LFP). The premature float bug can't be the problem, because you have your float voltage set just less than your bulk/absorb voltage. That's one way around the bug, but it's not the best for battery life. Does your battery actually reach 55.5 V, and typically stay near 55.4 V for at least an hour to 90 minutes? Raising the already high battery voltage settings will not do any good. What is your low DC cutoff voltage setting (setting 29)? Do you regularly let your battery run until the inverter cuts off? I assume so, since you say you are off-grid, and presumably don't run the generator until you have to. That could be killing your battery. 44.0 V (assuming 16S) is 2.75 VPC, which is really low. I have a rule of thumb (thanks to Weber, distilled from various scientific papers) for low SOC, SOC = (Vbatt - 50.4) * 20%. That puts 0% SOC at 50.4 V. No doubt this rule of thumb breaks down at extremely low SOC, but still, I would not want a 16S LFP battery to ever go below 50.5 V (lightly loaded and rested). Edited June 27, 20206 yr by Coulomb
June 27, 20206 yr 2 hours ago, Coen said: I called my installer and he suggested I add a Victron device (not sure what it was) as ICC and the Revov BMS seem to not be on the same page. He probably suggested a BMV-700 (or 702, or 712). They are around 3k a pop, and a fantastic bit of kit that is easily worth the money BUT.... the BMS of your revov battery should be doing this job already (keeping track of the state of charge), so you really shouldn't need to add this to the setup. I know there are batteries that don't do good SOC tracking at low power values (BYD, Bue Nova) because their current sensors cannot sense accurately low down. For such setups we often advise adding a BMV (which is exceptionally accurate down to at least 100mA), but I don't think this is the problem you have. So it really depends how your SOC tracking is done, and how you measure the charge and discharge energy, and while a good BMS can give you that information, I have no idea what kind of BMS you have fitted or if ICC can read those values from it. I can tell you that a BMV would give you all that information in a heartbeat. It sounds to me that something is up with the battery. It is either not fully charged (though if it reaches 55V and the cells are fully balanced, that's 3.45V per cell which is damn near 100%), or you have a cell that has much lower capacity than the others and you possibly have a battery claim. But you won't know unless you have a reliable way to count energy into- and out of the battery, which means being able to read the BMS. Can ICC even do Revov now?
June 27, 20206 yr Author 8 hours ago, Coulomb said: What do you mean by "fully charged"? If you're running ICC with no Victron BMV [ edit: and if ICC can't talk to the battery BMS directly ], then I believe ICC has to rely on the inverter's guess, which is based on lead acid thinking. I'll guess that your low voltage cutoff is set to 44.0V (which is really low); the Axpert calls that voltage -2% SOC, and adds 1% SOC for every tenth of a volt above the setting plus 0.2V, reaching 100% at 44.0 + 0.2 + (100*0.1) = 54.2 V. But that's meaningless; a 16S (I'm guessing) LFP is nowhere near full at that voltage; it needs to get over about 55.0 V; my bulk/absorb charge voltage is set at 55.2 V (and I also have 16S LFP). The premature float bug can't be the problem, because you have your float voltage set just less than your bulk/absorb voltage. That's one way around the bug, but it's not the best for battery life. Does your battery actually reach 55.5 V, and typically stay near 55.4 V for at least an hour to 90 minutes? Raising the already high battery voltage settings will not do any good. What is your low DC cutoff voltage setting (setting 29)? Do you regularly let your battery run until the inverter cuts off? I assume so, since you say you are off-grid, and presumably don't run the generator until you have to. That could be killing your battery. 44.0 V (assuming 16S) is 2.75 VPC, which is really low. I have a rule of thumb (thanks to Weber, distilled from various scientific papers) for low SOC, SOC = (Vbatt - 50.4) * 20%. That puts 0% SOC at 50.4 V. No doubt this rule of thumb breaks down at extremely low SOC, but still, I would not want a 16S LFP battery to ever go below 50.5 V (lightly loaded and rested). Thanks for the reply. Fully charged - the light on my Axpert stops blinking and stays green My cut-off is set at 47.5, and I basically hit that every night now. At 49.5 the Axpert starts beeping (I leave that on to warn me) - and I go into "Eskom Load Shedding" mode - I switch it off. I have various UPS's to run all the necessities - and a squad of power banks! I'll measure the battery tomorrow when it's "charged", the Axpert/ICC reports the voltage but I'll use Fluke to see.
June 27, 20206 yr 6 hours ago, Coen said: Fully charged - the light on my Axpert stops blinking and stays green OK, that's when the charger goes to float mode. It should not be doing that when the battery voltage is only 54.2 V. Unless 54.2 V was immediately after the charge LED stops flashing; there is a minute or so with zero charge before the charger brings the battery back up to float voltage. 6 hours ago, Coen said: My cut-off is set at 47.5, and I basically hit that every night now. That's still really low, and may well be the source of the problem. Staying at such a low state of charge overnight is not good. 6 hours ago, Coen said: I'll measure the battery tomorrow when it's "charged", the Axpert/ICC reports the voltage but I'll use Fluke to see. Good idea.
June 29, 20206 yr On 2020/06/27 at 11:26 AM, plonkster said: He probably suggested a BMV-700 (or 702, or 712). They are around 3k a pop, and a fantastic bit of kit that is easily worth the money BUT.... the BMS of your revov battery should be doing this job already (keeping track of the state of charge), so you really shouldn't need to add this to the setup. I know there are batteries that don't do good SOC tracking at low power values (BYD, Bue Nova) because their current sensors cannot sense accurately low down. For such setups we often advise adding a BMV (which is exceptionally accurate down to at least 100mA), but I don't think this is the problem you have. So it really depends how your SOC tracking is done, and how you measure the charge and discharge energy, and while a good BMS can give you that information, I have no idea what kind of BMS you have fitted or if ICC can read those values from it. I can tell you that a BMV would give you all that information in a heartbeat. It sounds to me that something is up with the battery. It is either not fully charged (though if it reaches 55V and the cells are fully balanced, that's 3.45V per cell which is damn near 100%), or you have a cell that has much lower capacity than the others and you possibly have a battery claim. But you won't know unless you have a reliable way to count energy into- and out of the battery, which means being able to read the BMS. Can ICC even do Revov now? Yes ICC can do Revov now. ICC website tells you to contact Diversified Solutions for the correct cable and ICC settings.
June 29, 20206 yr Author So I measured the battery using my Fluke. It seems to stay at 54.0V max. ICC and Axpert showed 54.2V but battery at 54. I also increased the cut-off voltage to 48V, that is the highest ICC let me. Upping the voltage made no difference though - at 9PM last night we were in the dark again ;(
June 29, 20206 yr Author 2 minutes ago, Shadders said: Yes ICC can do Revov now. ICC website tells you to contact Diversified Solutions for the correct cable and ICC settings. Thanks!
June 29, 20206 yr 28 minutes ago, Shadders said: Yes ICC can do Revov now. ICC website tells you to contact Diversified Solutions for the correct cable and ICC settings. OK. What can it read? In a case like this, I like to plot the SOC (if you see it discharge at a steady rate, and then suddenly dive to zero, that's evidence of something not being healthy). Also, I like to see the individual cell voltages, or at least the minimum and maximum (the min and max is actually all you need). When it switches off, what's the voltage of the lowest cell? Also, who is doing the switching off. Is it the inverter, or is it the BMS that disconnects the battery from its loads to protect it?
June 30, 20206 yr 17 hours ago, plonkster said: OK. What can it read? In a case like this, I like to plot the SOC (if you see it discharge at a steady rate, and then suddenly dive to zero, that's evidence of something not being healthy). Sadly I'm sitting waiting for my adapter cable to get sent to ZIm from SA, it should hopefully be here within the next 2 weeks and I will update.
July 15, 20206 yr Right so I've been up and running for a couple of days and monitoring the Revov communications. ICC reads the battery temperature, SOC, cycles and current from the Revov BMS. The SOC is exported to the local emoncms so it is logged along with the other standard parameters. Currently no cell voltages are reported. I've not run it down to shut down yet so can't answer what is doing the disconnecting.
July 15, 20206 yr Author Sadly I found out that I'm using the "wrong" ICC software. It seems there are two versions, I have the one supplied by Centurion Solar. They do not cater for Revov, as far as Revov is aware. Some news though - Revov was at my house and drew logs from the BMS. They also swapped the BMS cables to ensure it wasn't that. It seems we all agree now that it is a battery problem. They wanted to take my battery in but since I am off the grid, we are first waiting for a loan battery to become available. Sarel has been a great help and hopefully we will have this sorted out by the end of next week.
July 27, 20206 yr Author Another two weeks gone and I tell you - this is frustrating. No contact, no update. Today someone asked me who they should speak to about going off-grid, they are building a home and they want to do everything up-front. All I could say was "stay away from ZASolar and Revov". What does a 10-year guarantee help you if the supplier is useless? I'm at wits end, after an entire 8 months I am in the dark by 8pm every night. Wow. I hope people search the forums before they take the plunge.
January 17, 20233 yr Hi Coen, Did Revov ever resolve this issue. I'm very interested in their second life batteries, but won't take the plunge if they don't provide good back up. Thanks.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.