TheOracle

Thanks for all the feedback, really appreciate it. The 0.5% default comes from the NREL field studies, which track what installed panels actually do over decades, and the median lands right around half a percent a year. The datasheet curves on new Tier 1 panels are lower, around 0.4%, but those are warranty floors rather than measured performance, so I kept the field number as the default. That said, it's a slider, not a fixed assumption. You can set anything from 0.3 to 0.8, so new BC or bifacial panels can run at 0.3 or 0.4 and older poly at the top end. I've also updated the note under the slider to explain exactly this, what the default is based on and when to move it. On bifacial gains, the cleanest way to capture those is the yearly output override box. Put in the kWh per kWp your system actually achieves and the gain is baked in, without me having to guess your reflective surface.

Good eye on both. Quick answers: panel ageing is in the off-grid numbers, it's just folded into the "Demand not covered" line (as panels age they make less, which widens exactly that gap) rather than shown separately, so it isn't double-counted. And that "Panel ageing" figure isn't a panel replacement cost, the tool never assumes you replace panels, they're warrantied 25 years plus. It's the savings you slowly lose as ageing panels make a little less power each year, added up over 20 years in future rands, which is why it looks big next to what you paid today. I'm relabelling it to make that clearer.

Thanks, that's a good suggestion and you're spot on about GTI being the more accurate route. I went with a slightly lighter version of the same idea. There's now an optional "Expected yearly output (kWh per kWp)" box under the sun setting. You pop your address into the free PVGIS tool (it already works out your latitude, panel direction and tilt) and drop the yearly figure straight in, and the calculator uses that instead of the regional sun estimate. Same accuracy you were after, just without asking everyone to type coordinates. Give it a go and let me know if it does the trick.

Thanks for the suggestion, Generator fuel cost field added

I have made another update. There's now a proper off-grid mode that shows how much of your usage the system actually covers, so a well-sized setup reads close to 100% instead of being marked down for surplus it curtails on sunny days.

That battery figure was the one that needed fixing first. I've dropped the default to about R2,500 a kWh, which is much closer to what you're actually seeing, and it's still a field you can edit to your own view. The off-grid self-consumption point is next on my list.

Thanks, this is genuinely one of the most useful bits of feedback I've had, and coming from five years off grid it carries real weight. A few quick responses. On battery replacement, the figure is actually a field you can edit, it just defaults high and tracks your bank size, but you're right that the default is steeper than today's pricing, so I'll bring it down closer to what you're seeing. The self-consumption point is fair too. The numbers lean toward a typical grid-tied home, and a properly sized off-grid setup like yours sits much higher, so an off-grid mode with a high utilisation option makes a lot of sense. The Section 12B business toggle and the value of uninterrupted power are both great calls, especially for farmers and anyone running an income off the system. Bi-facial gains and hardware tiers I'll keep on the list, though I want to add them without making the thing a spreadsheet. Really appreciate you taking the time. I'll work through these and post back when the next version is up.

Thanks for the feedback. The range is extended to 6.5. I also added a location dropdown so you can select your location.

Good point, although the first R23,800 interest income per tax year is exempt from income tax, so it will not affect too many people.

Most solar calculators online quote you a payback that never matches reality. They assume you use every kilowatt your panels make, that prices stand still, that your battery lasts forever and that the cash you sank into it could not have earned a cent anywhere else. So I built one that does not pretend. You put in your real numbers and it shows you two figures side by side - the rosy payback a salesman quotes and the realistic one you are actually likely to get - then it breaks down exactly what eats the gap. Wasted daytime generation, panel degradation, the battery you will replace once down the line, financing interest and the higher fixed charge some municipalities slap on solar homes. It is free and there is no signup. Set your own tariff off your bill rather than the presets and be honest about how much you use in daylight and you will get a sensible answer. Anyone interested can have a look here: https://southafricafacts.co.za/solar-savings-calculator-south-africa/ Keen to hear if the numbers line up with what you are actually seeing. Share your own setup and payback if you are up for it.

I have made a cable with the pins as suggested by you, brother, but it is not working. I tested the cable with a cable tester to be sure. The Deye battery type is set to Lithium and the Lithium mode is set to 0 Anything else I should check?

Thanks for the reply. Do you happen to know if the cable from Livestainable has a GND wire?

Maybe I should just try a cable with GRN. Does anyone know where I can find someone to make me a custom cable in Pretoria East?

I have a Deye 6kW off-grid inverter and a Pylontech US3000C battery, and I’m struggling to get reliable communication between the two. According to the manuals: Pylontech: - Pin 4 = CAN-H - Pin 5 = CAN-L - Pin 6 = CAN-GND Deye: - Pin 4 = CAN-H - Pin 5 = CAN-L - Pin 6 = BMS-GND / RS485-GND So on paper, the pinouts seem compatible. I bought the Greenrich/Pylontech to Sunsynk (Deye) CAN cable from GeeWiz, but that cable only uses pins 4 and 5 (CAN-H and CAN-L) and does not include a ground (pin 6). From what I understand, CAN communication should theoretically work with just CAN-H and CAN-L, and some Deye documentation also suggests that only pins 4 and 5 are used. However, I’ve also seen cases where adding a ground connection (pin 6) or using a custom cable resolved communication issues completely My question: Do I actually need the ground (pin 6) for Deye ↔ Pylontech CAN communication, or should pins 4 and 5 be sufficient? If anyone has a confirmed working cable pinout for this exact setup, I’d really appreciate it.

Hi all, I recently replaced my old inverter with a Victron Phoenix 12/250 to power a koi pond pump that runs continuously. I'm monitoring the system in two ways: A plug-in energy monitor at the AC side reports that the pump is drawing 58 W. My BMV-702 battery monitor, however, shows that 80 W is being drawn from the battery. This suggests the inverter is using around 22 W, which seems like a lot for such a small load. What’s interesting is that with my previous inverter, the difference was only around 5 W for the same pump and monitoring setup. I’m aware that inverter efficiency drops at light loads, but I didn’t expect it to be this noticeable. The Victron is supposed to be a high-quality unit, so I’m wondering: Is this normal behaviour for the 12/250 model? Is there any way to improve efficiency at light load? Has anyone had similar results or found better options for running small continuous loads? Any insights or comparisons would be appreciated. Thanks!

Yes, the issue was related to a Tuya smart breaker I added.

Yes I did, I configured it like this:

My battery cells seem to be in balance

I spoke to early, it happened again today except the soc went to 98% I have to switch off the inverter for the voltage to go back to normal.

You correct, my bad. Please see here for more details about my system:

@Coulomb, my apologies I only saw your post now. I have my Solar Assist connect to the inverter and PylonTech battery so I assume the soc is from the battery. Again thank for your input.

Thanks for everyones input, I really appreciate it! I think I fixed the issue! I added this Tuya smart breaker to my electrical panel last week, but it turns out that was the culprit. Once I disconnected it, everything went back to normal.

My Axpert is misbehaving again. It only charges to 89% and then after a while the battery voltage go very high and the solar panels disconnect. Here is my current settings: I did upgrade my battery firmware recently:

Today my soc stayed on 89%

I upgraded the firmware, hoping it would solve an issue I had with the battery. In the afternoons, once my battery is fully charged, I use my excess solar power to heat my koi pond. I automatically switch off the heater once the battery's state of charge (SOC) is 94%. But what happened was that the SOC stayed at 94%, although the battery still had a good charge. Unfortunately, the upgrade didn't help. Today the SOC actually went down from 94% to 93%.