Ironman

[sarcasm] Why is it that days with high power consumption correlates with cloudy days - and a serious loss of PV power - battery depleted by 11pm; while days with low power consumption correlates with no cloud cover and perfect PV power?

I am running. Very similar setup, with the 250/100 MPPT, using 6000wp artsolar panels. 20 x 300w pure black panels. 4 x 5s. This was my solar power for yesterday: the MPPt just limits the output current to 100 amp by drawing less current from the panels. No problem.

The Venus controller reads the battery State of Charge from the BMS of the pylontech batteries. So the system gets an accurate battery state of charge, and thus will ensure a much longer discharge time. You can also build your own controller with a raspberry and Venus image. The system will work without the venus, but the MPII will then try to judge the battery charge level by just reading the battery voltage. There are many tales of broken expectations when systems are set up like that - under load, the voltage drop in the batteries and cables cause the MPII to under-estimate the battery level badly, eventually switching off due to low battery voltage, while the batteries still contain a good charge. @plonkster - feel free to elaborate or correct me if I am wrong here.

The rack has space for a third battery if we feel we need more runtime in future.

We needed a UPS for a small office (about 8 PC's, server and network hardware) to be able to continue working through loadshedding (4-6 hours at a time). We are renting our office and did not want a permanent setup. Requirements: - Ultimate long term reliability - high quality brand - Expected lifetime: about 15 years for inverter and 7-10 years for batteries (expecting much more regular loadshedding while eskom repairs its infrastructure) - Backup system that works for many years without maintenance - Standalone system ( rack on wheels) - Ready to be moved to a new office if we have to move, OR be re-purposed into a solar home ESS system (just add MPPT and panels) If we bought an off the rack 3KVA UPS system with 6000 WH runtime, we could probably get it for about R20k. But I have a long history with cheap UPS systems - they never make it to 2 years - then you have to replace the batteries - at huge cost, labour - weekend time etc. At about 4 or 5 years the inverter's cheap capacitors start sweating and it becomes very unreliable - almost worse than not having a UPS. Not one lasts over 6 years - and in the process, the PC's that you are supposed to protect are subjected to many power spikes and failures due to the unreliable UPS. So we went this route - about 4X more expensive but we know it will last, and protect the sensitive equipment properly for many years: - Victron Multiplus 3000 - Victron Venus controller - 2 X Pylontech US3000 batteries - 19" computer rack on wheels - Small DB with input breaker, earth leakage on output, and two circuits on output. - DC Fuses, cables and building the rig (took about 5 hours). @Jaco de Jongh built this rig for me:

Hey @Elbow ...we are on this forum for details details details, not mutual back slapping. Please tell us a bit about your system, the diagrams and documentation needed, and the process. Others want to learn and follow the same path. ... and congrats.

https://www.fin24.com/economy/breaking-eskom-warns-it-will-stop-providing-daytime-power-to-bloemfontein-unless-it-gets-paid-20191125?mobile=true https://twitter.com/Eskom_SA/status/1198952306735288321/photo/1?ref_src=twsrc^tfw|twcamp^embeddedtimeline|twterm^profile%3AEskom_SA&ref_url=http%3A%2F%2Fwww.eskom.co.za%2FPages%2FLanding.aspx If I were a solar installer, I would temporarily move to Bloemfontein - everybody who could afford to add R100-200k to their home loan debt will be clamouring for a solar system. Most will go completely off grid - cutting off more and more of Eskom's income. This is a bit tongue in cheek - but a serious electricity supply situation is developing in one of our large metro's. Eskom wants to cut power every day from 6AM to 8PM. All the citizens that pay their accounts every month will suffer together with the few that has robbed/stolen/mismanaged the city's finances to cause this issue.

So this is a 12V system. According to this you have a 1000W inverter... but the image shows 1500W printed large on the inverter. And you say it is 1200W... These varying specifications indicate a cheap brand that will lie to get you to buy their product. But sometimes they can work ok. Lets give the inverter the benefit of the doubt. Do you know what the low voltage cutoff of the inverter is? When trying to draw 650W from the inverter, it will be drawing about 720W (including inefficiencies) from the battery - and at 12V, that is 60 Amp. I hope you have very thick cables running from the batteries to the inverter. These cables must also be as short as possible, since any voltage drop over the cables is a much more pronounced loss at 12V than a typical 48V system. You can measure the DC input of the inverter with a cheap digital multimeter while drawing a load to see how that voltage drops. Also - Why do you say "inductive load"? Heaters are normally resistive loads. Inductive loads are normally motors, compressors, airconditioners etc. A resistive heater should be an easier load to carry than an inductive load which could have a large startup "surge" current. Can you upload photos of your battery wiring, and heater load so we can get a better idea of your setup?

I have also used mixed panels for many years. The best way to run them: - Get panels that do not differ by more than 15%. - Make one string with each kind of panel. Do not mix different panel ratings in one string. - If possible, keep the strings separate and use separate MPPT's for each string. - If not, and you have to put these strings in parallel on one MPPT: it will hunt around and find the maximum power point for all panels together. It adjusts the voltage and current drawn from the panels in a "searching" algorithm, while testing the power delivered (volts X amps). As the voltage goes up, the current comes down. As the voltage goes down, the current goes up. And all the time it tests - what is the power - should the voltage go up or down? So the MPPT "hunts around" for the maximum power point. - Using mixed panels like this, the overall maximum power point voltage will be at the maximum power point (Vmp) of the weaker panels - the stronger panels will deliver their normal current, but at the (lower) voltage of the max power point of the weaker panels. More or less. So if you use 330 Watt and 375 Watt panels in parallel, you should get about the same as if you only had 330 Watt panels. The 375W panels will be derated to 330W each, as @Jaco de Jongh said.

Not today. Bad clouds - solar production went down to 10W from a 6000W array, over lunchtime! I could not believe that mid day dark clouds could bring down solar output that much! The battery went up to 94%, so I think my problem is solved - but they did not stay there - I don't think they balanced the cells properly.

Thanks for the fix. I did ask two times earlier in this thread why the graphic representation's battery level differed from the Pylontech's indicated level. For anybody that has this issue in future: the setting is under system setup: Mine was set to Multiplus - and somehow the multiplus has a completely different SOC level.

mmm I never saw a SOC higher than 84% today. And even now there is a large difference in the graphic display of the Venus, and the Pylontech data: The one shows the SOC at 68% and the other at 84%... Something really weird happening here.

How do you heat your house?

And now this happens: The Venus display of the Pylontech battery shows 99%: While the overview graphic animation shows the battery at 83%: The voltage is still 52.0V. Maybe the battery voltage "spiked" to 52.5V quickly, and now the battery is in float, which is still at 52V. But then why does the Venus state the mode as "absorption"?