Follow

Who the fuck thought it was a good idea to AC-couple solar panels to a home's electrical network?

Why must this asshat make solar telemetry so fucking hard that one cannot discern solar production from grid consumption?

FML.

@tomas Any MODBUS-level experience with Darfon inverters by any chance?

@vertigo Afraid not. I'm curious about how your system is connected, because not being able to measure production at the meter is very common. You'd need entirely separate circuits and meters for that. My system at least works/will work like that, billing/reimbursement being based on net current at the meter. Actual production would be read off via ZigBee.

@tomas That kind of gets at the root of my issue. I don't know how it's configured internally, but it's clear that you need two inverters (a grid-tie inverter and a battery inverter) instead of just one. It's a more complicated design, with more (metaphorical) moving parts that can go wrong.

For my immediate needs, however, the relationship between battery power, load power, grid import/export power, and ohmic losses isn't clear, and telemetry is producing measurements which make sense.

@vertigo @tomas

@Edent in Oxford, SE England generates solar power at his house, perhaps he also has info available on how the measurements are made (there is also a bot which toots to the timelines showing how much power comes from this installation) @OxfordSolarLive

@vfrmedia @tomas @Edent @OxfordSolarLive Thanks for the offer; however, my circumstances are pretty specific. It only applies to AC-coupled Darfon installations.

@vfrmedia @tomas @Edent @OxfordSolarLive I'm aware that Kirchoff's laws that has to apply (ideally, the sum of all power current, and through Ohm's law, power readings must add to zero; if it doesn't, we're seeing either ohmic losses in the system (which should be relatively constant) or ADC noise (which should be insignificant)). But, what I'm seeing is wildly varying and quite significant, a clear violation of Kirchoff's current law.

@vfrmedia @tomas @Edent @OxfordSolarLive Apologies for my poor typing skills. I seem to be having difficulty seeing and reading this morning. :(

@vertigo @tomas @Edent @OxfordSolarLive

In recent times I seem to routinely miss out entire words in toots or substitute wrong ones, in spite of being able to see perfectly well enough for the driving licence eyesight assessment a couple of months ago.

Perhaps I should start tooting more in other languages than English, as I seem to pay more attention when doing that 😆

@vertigo "it's clear that you need two inverters (a grid-tie inverter and a battery inverter) instead of just one"
Have you considered using a hybrid inverter?

@tomas I'm writing software for a product we're developing. The configuration of the product is already set in stone.

@vertigo You can't sneak a current transformer in somewhere? If not then yeah, sounds like you have quite the pickle

@tomas No; I'm a software engineer, and these systems are already designed, certified, etc. My task is to repair a bug in the telemetry code which works fine for our DC-coupled inverter systems, but not for the AC-coupled. Making HW changes now would require us to spend the time and $ to re-certify, and it wouldn't do much for our existing deployments.

@tomas My measurements show that CTs will only slightly help, but won't cure the bug. Solar production would *still* reported on the essential load reading. :(

True, I can at least separate out load from solar with the existence of CTs; but I still need to maintain two separate patches of code (one for DC and one for AC) to produce the proper telemetry.

This is a bug which is best fixed via a firmware update on Darfon's side.

Sign in to participate in the conversation
Mastodon

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!