Halcyon

Very impressive. Can one in a setup like yours create a custom field to monitor inverter efficiency for example and plot on your chart over time? Inverter efficiency = dc power (solar power +- battery power) / inverter load power * 100?

it will be helpful to paste screenshots of your battery settings and all time of use "system mode' settings page to troubleshoot.

Hi Anton, make contact with https://www.alumo.co.za/, they are Pretoria based, supply, install and provide great after sales support for Sunsynk machines (simmilar to Deye). Regards

Hi Duane, I had a similar thing happen on my side. I did not notice the Installer used a battery fuse with a lower rating than what we programmed as the maximum charge and discharge current for the battery. Naturally the thing burnt out. Something to check on the possible reason you lost the fuse in the first place.

To make things really confusing, especially considering they are basically the same machine, the Deye manual states the ATS port makes 230V when grid is on and Ov when grid is off. The Sunsynk manual and video states the exact opposite. So when using a relay to bond earth and neutral when going into UPS mode, be careful to use the correct NO/or NC wiring for the relay depending on which machine you are using. Has anyone confirmed this difference between the machines? To make things even more interesting the Deye manual recommends the following wiring setup for markets like Oz and RSA https://www.deyeinverter.com/deyeinverter/2021/02/06/502012021【b】储能单相说明书sun-3.6-5k-sg03lp1-eu-ver2.0-deye.pdf Refer section 3.9. Would this wiring make the grade to be SANS compliant? The older Sunsynk machines(mine included) have the inverter's internal load CT on the load output neutral wire. Any neutrals shared between essentials and non-essentials in the DB caused havoc on the inverter's load reading in some cases. Apparently this has been changed on the later versions where they moved the CT to the live wire to eliminate any issues. I assume the same goes for the Deye machines. Anyone have further insight into this?

Yes I have the latest Sunsynk logger. (I think there is a wifi/lan version as well). If your issues persist consider swopping the wifi only version to the one with a LAN cable port. You should have rock solid signal then.

Hi, Did you select the closest and strongest access point when you scanned for networks using the logger? You may have chosen a further away one with less strength and the logger sticks to that one.

For those that are interested I took this issue further to analyse the efficiency at a daily level, before and after rebuild. This image will help you better understand how the inverter works and also what I plotted on the graph using the data from the inverter's logger (before and after the rebuild). Label 1 = PV power Label 2 = battery power (positive discharge, negative charge) Label 3 = Inverted AC power from DC to be used for loads and export. Label 4 = power used by backup circuit (load output) Label 5 = power used by non-essential circuit(geysers, pool, stove etc) Label 6 = power measured at inverter's CT coil, placed at main feed from utility. It measures any power ”purchased and controls backed if no export power allowed by utility. I agree that my logged ratio of 3/(1+2) may be a little crude but it does show a trend. Blue line - sunny day before rebuild, orange line, sunny day after rebuild. I did a couple of these comparisons on different days with varying solar irradiance. The trend was the same, flat line before rebuild and a parabolic line after rebuild. Average of daily total of 3/(1+2) before = +- 95%, average daily total after of 3/(1+2) after = +- 85%. According to me a trend like this indicates something is not right? Any ideas what this may be?

Putting a multi meter on the main grid feed is not a reliable way to check if zero power on inverter corresponds with power on grid feed. Even if the CT on inverter screen show 0W a simple clamp meter may measure much more current on the main feed. It’s a long story but a current clamp meter can’t differentiate between reactive and active power. Your street meter only measures active “real” power and same as the CT clamp. see link for more info https://electrical-engineering-portal.com/beer-mug-and-power-factor

I agree with PowerUser. There may be something wrong with the reading at the CT. To test, If you switch on non-essential loads (geyser etc) does a value appear at the house icon? The power used for loads not on backup will appear there. The total power used in the house is your house icon plus ups icon. The total power purchased, measured at the CT, is what should appear on your Eskom bill.

Thank you PowerUser, helpful information.

This should give a good idea of how the efficiency tracks over time. Notice the result after repair. Will monitor for a while and if it stays much lower like this will definitely get in touch with Sunsynk to find out why.

Hi All I am running a Sunsynk 5,5kw with Revov 10,2 KWH battery setup. We have all seen the efficiency claims of inverter's however has anyone actually tested this? For interest I have pasted extract of the Sunsynk spec sheet on efficiency. Max. Efficiency 97.60% MPPT Efficiency 96.50% Euro Efficiency 99.90% I have recently installed the new Sunsynk logger and decided to analyse the results. I have 2 months data so far. (I have two years of data stored on the solarman cloud which I have not yet analysed) Jan 2022 Theoretical load = Import + Discharge + (PV - (export+charge)) 52,7 + 29,9 + (52,4 - (0,1 + 28,1)) = 106,8 Measured load = 103,7 Efficiency 103,7/106,8= 97,09%. (Pretty good and in spec) Subsequently my inverter went in for repairs and they changed all the control boards to the latest versions and the results are as follows? The results below are from my semi new machine. Feb 2022 Theoretical load = Import + Discharge + (PV - (export+charge)) 77,6 +46,7 + 114,1 - (2 + 51,2)) = 185,2 Measured load = 169 Efficiency 169/185 = 91,25% (Quite a bit lower than spec?????) Anyone else willing to share their graphs to check efficiency? Curious how the new machine seems to be less efficient? Any suggestions on a better way to calculate the overall efficiency? Looking forward to hearing from all the Sunsynk/Deye inverter fans Regards

Hi All I have been following this thread for a while now with keen interest. I am a total novice in the DIY home automation and logging arena and have a lot to learn. Using the setup described I did pick up that many of you are using conditional logic based on the inverter data collected to switch on and off connected devices like sonoff switches, relays, lights and send various notifications. Would the same setup be able to use conditional logic to change the inverter settings? specifically time of use and other settings I have in mind. For example : Update time of use schedule to start discharging battery 8 hours before sunrise each day Use inverter time of use schedule but if battery at 17:00 is not higher than 70% SOC and load shedding is active, grid charge battery to 100% SOC, and then return schedule to normal. If weather app forecasts a cloudy day, change priority of solar power to charge battery and not load You get the idea. Thanks your feedback would be appreciated Regards

Hi Bluobul A fellow Sunsynk owner managed to log into the Sunsynk supplied WIFI logger. I pasted his feedback posted on the Sunsynk forum below. Would this assist in anyway for your custom monitoring project? Does it indicate that one may be able to do away with having to use a Raspberry Pi to manage the data/transfer? "Hi folks, somewhat related, I noticed the data logger listens on port 80/http, and has a web UI, and allows doing config for the data logger. First, it has default password of admin:admin. Hehe, um, this is exactly why I have proper wifi setup with my IoT stuff on a seperate VLAN and wifi SSID. IoT things aren't that secure. Anyhow, noticed you can at least change the admin console HTTP password for it to something a little less guessable than admin:admin. There does seem to be a feature to cater for 2 servers, so if we get to know what protocol to use, we could possibly setup our own data logging receiver server.

Hi Johann, Yes the one I have is the same as Justin's as per the link he posted. Just marked as 12V somewhere in error on the page but must be the 230V 40A version.

Yes I have the same question on why the app (SolarMAN Pro) shows battery discharged. I did not even use the battery that day. I am running the Solarman Smart app for comparison as well. Consumption power also Indicated the 29,4 kWh which is definitely wrong. I agree Sunsynk should develop their own monitoring and control portal. Apparently it is in development as we speak. There is definitely some anomaly with the way the inverter reports information to SolarMAN. All the information on the local inverter screen makes sense and is accurate. BTW I found a setting on the SolarMAN app “Calculation of Electrical Energy”. Default settings for Production, Consumption, Grid Feed in power etc , is on “based on device data”. I changed the consumption field to “based on calculation”. Two days running like this and my consumption data reported seems to be in line with expectations and my historic usage readings. Perhaps that is the fix for now.

Hi All, have any of you noticed that the consumption power reported by the inverter on the Solarman app does not seem to be correct? I have re-pasted my post from the Sunsynk forum below. Please check my logic. Solar produced 13,8kwh + Battery Discharged 0,00kwh - Solar sold 9,1 kwh + Grid purchased 6,5kwh - battery charged 0,00kwh = 11,2 kwh (Theoretical kwh used by load circuit) Why does Solarman report consumption as 29,4kwh? (refer screenshot below). Based on past experience my load circuit only use between 11 - 15 kwh in a 24 hour period. How is this discrepancy possible? (past experience is from an old Voltronic inverter which used to run same backup circuit. The 11 - 15kwh used in 24 hour period were corroborated by an effergy energy clamp on same load wire with 2 years data to back it up) After engaging with Solarman on the question and a lot of testing later their feedback was that there is nothing wrong with their algorithm and this has to do with how the inverter reports the data. The suggestion was to take it up with Sunsynk. So here we are.

Thanks Johann. No need to upgrade at point unless there are new features on latest version? They upgraded me the other day to a newer version in any case. I am not using a smart load with reset issues that needs resolving so at this point all is well.

BTW how were you able to get hold of the files? See screenshot below of place on Solarman pro where I presume you save file, select and push to device yourself.

Not to my knowledge, the GPRS logger was literally plug and play with no setup required with 2 years free data. The downfall is that one cannot connect to it locally for firmware upgrade. However I remember now that in the Solarman pro account desktop version has a section where one can select and upload software. @Achmat in theory you could do it yourself with the files Johann provided? Keith once mentioned on the Sunsynk website that they are working on using one of the LAN ports to be able to plug into home network directly so in theory no need for loggers etc, however I have seen no progress on that development.

Awesome, thank you. Unfortunately in my case I have the GPRS logger so there is no way to log into the IP address and do a self upload. I will have to rely on the bunch in China to push to inverter. Unless there is another way to do a self install?

These software updates are super confusing. I requested latest update and my comm version is now VE 406 and control software is V0.1.3.6. Support assured me it’s the latest version. On a 5,5kw machine.

Hi Justin, for interest does the current functionality of the smart load feature not have enough programmability "almost" to achieve what you want to do? To my knowledge the smart load can be programmed to switch on and off based on the battery SOC and the power produced on panels. e.g, you can program if the battery above 30% and solar power in excess of 4kw then smart load will switch on and could power your geyser.