I have a Sunsynk 5 kW inverter installed, and running 2 MPPTs (5x550W facing East) + (7x 550W facing West) [No north facing roof unfortunately].
Storage is 2 x Narada 7.2 kWh batteries.

The system is performing well in general, but I have a few questions, and Sunsynk Support has not answered in 10 days, hence posting here.

Firstly, the internal consumption on the system is constantly between 175 and 195W.  From other posts and other forums, the view is that internal consumption should be on 45 - 50W. 
(Screenshots attached)

Secondly, the Sunsynk (running latest firmware), seems to be ignoring my values set for Grid Trickle Feed.  I see posts advising setting it to 20W, but even when setting it as such, the Grid is always between 97 and 120W, except during good old Eskom Loadshedding or when I am actively drawing power from the Grid.

I am not feeding back to the grid.

5 minutes ago, 007 said:

Firstly, the internal consumption on the system is constantly between 175 and 195W.  From other posts and other forums, the view is that internal consumption should be on 45 - 50W. 
(Screenshots attached)

Secondly, the Sunsynk (running latest firmware), seems to be ignoring my values set for Grid Trickle Feed.  I see posts advising setting it to 20W, but even when setting it as such, the Grid is always between 97 and 120W, except during good old Eskom Loadshedding or when I am actively drawing power from the Grid.

I have a Deye 5kw inverter. I see the exact same behaviour. Around 180W self consumption and it always pulls at least a 100W from eskom. Sad but true. I confirmed the values using an external meter just to make sure. Just a top tip, do not believe the numbers that you see on the inverter, those are not very accurate unfortunately.

Edited August 1, 20233 yr by HendrikBigChief

22 minutes ago, HendrikBigChief said:

I have a Deye 5kw inverter. I see the exact same behaviour. Around 180W self consumption and it always pulls at least a 100W from eskom. Sad but true. I confirmed the values using an external meter just to make sure. Just a top tip, do not believe the numbers that you see on the inverter, those are not very accurate unfortunately.

Thanks HendrikBigChief.

Makes me wonder where the people quoting 50W get their data?

Edited August 1, 20233 yr by 007

50 minutes ago, 007 said:

Thanks HendrikBigChief.

Makes me wonder where the people quoting 50W get their data?

I think they are just looking at what the screen shows, they did not measure the actual usage with a third party meter.

@HendrikBigChiefDo you think that the CT coil might be suffering from interference or is faulty?

6 hours ago, DashPole said:

@HendrikBigChiefDo you think that the CT coil might be suffering from interference or is faulty?

From the user's of a Sunsynk/Deye. Is there no change of setting from the standard 2000/1 I think? Some makes do allow you to change the ratio up or down. If this can be done then may be users can fine tune it to give a more accurate reading of power being used in and from the grid. 

My experience with CTs are that they normally fail and not short only a few turns as per a voltage transformer. 

On 2023/08/01 at 2:08 PM, 007 said:

Firstly, the internal consumption on the system is constantly between 175 and 195W.  From other posts and other forums, the view is that internal consumption should be on 45 - 50W. 

I wonder if there can be a single figure for self-consumption. These inverters are actually several devices in one. My Goodwe's internal consumption depends on what it is doing. I know from recent experience that at night, when it's just staying on, not supplying any loads, with the only power source the battery, then it draws about 50W. But during the day when there is PV and it is taking that and converting it to AC then self consumption is higher. There would be other combinations PV AND charging the batteries, PV and not charging the batteries. Point is, self-consumption changes according to what the inverter is doing.

Working from diagrams like those you post (neither of which show export to the grid) might also not be an accurate way to work. You need to confirm this, but it seems to me that if the property requires 494W, then the draw from PV and/or battery must be higher than 494 because of the power lost during conversion from DC to AC, or in stepping up the AC voltage. None of those processes are 100% efficient.

On 2023/08/01 at 2:08 PM, 007 said:

should be on 45 - 50W

I'd say realistically for a SunkenSynk, you're looking at 120W to 150W, that's what I am seeing on mine, whoever quoted the 45 to 50W figure should please let us know the model and serial number Sunsynk he's referring to an give us more data, since I don't think I've ever seen below even 100W self consumption....

This is mine(5kw sunsynk), 3273-2813-406=54W , i have seen as low as 30w . Not sure how you guys are getting more than 100w draw. The most i have seen was about 80w. 

• Total DC Input Power：3.273 kW
• Total Consumption Power：2.813 kW
• Battery Power：-406 W

First, the trickle feed

Mine behaves in line with the setting. I have it on 10w, and it will stay between 0-15 quite reliably. But this is as reported by the inverter - I have not used an independent CT to verify this. I have, however, noted that if you enable "solar export" then the grid import will hover around 50w while there is PV production. I don't understand why this happens. Anyway, I've stopped experimenting with export.

Second, the internal consumption (or losses)

As per some previous comments, the internal "losses" definitely vary depending on what's going on at the time.  And those losses usually manifest as heat, which climbs notably when you're harvesting PV and/or using heavy load.

Anyway, here's my graph of net losses from yesterday:

 

This is from Home Assistant. I created my own sensor to calculate the net loss. It is basically the sum of inputs (PV, Grid import, Battery discharge) less the sum of outputs (Load, Battery charge, Grid export). I evidently have some timing issues which cause silly spikes in the graph (both positive and negative!) - so try to ignore those.  But it is clear to see that in the "UPS mode" (running on batteries, low load) the internal losses are very low, less than 50w most of the time. But when the load increases, and when I'm generating PV power, the losses hover in the 150w-350w range. And, as I said, the increasing temperatures of the AC and DC circuits are evidence of these losses.

Here's my DC temp graph over the same period - it gets worryingly hot, despite the fact that I'm running a fan pulling air over the radiator fins constantly. I do unfortunately have it installed in my garage, which I have now realized gets rather warm on summer days - ambient temp in the garage is in the 30C-35C range.  

I find it surprising that the unit does not even have any cooling fans of its own.

Does anyone have any wisdom to share on improving the cooling on these 5.5KW SunSynk units?

On 2024/03/18 at 8:43 AM, Kalahari Meerkat said:

I'd say realistically for a SunkenSynk, you're looking at 120W to 150W, that's what I am seeing on mine, whoever quoted the 45 to 50W figure should please let us know the model and serial number Sunsynk he's referring to an give us more data, since I don't think I've ever seen below even 100W self consumption....

This is mine 8kw Sunsynk Inveter - 52Watts but mostly 51watts

I wonder whether more 8kW Deye/Sunsynk owners could chime in on their consumption...

Here with a 5kW, just like @007 the consumption is rarely below 110W or so... example below of now, a cloudy day...

so to do the calculation, 2478 - ( 895 - 1499  ) = 2478 - 2394 = 84W wow, a little while later...

3535 - ( 883 + 2479) = 3535 - 3362 = 173W, hmmm.... and yet more some while later...

2880 - ( 885 + 1883 ) = 2880 - 2768 = 112W....

Since I am looking at an upgrade, if enough real evidence comes in that the 8kW Deye/Sunsynk is more frugal in the own consumption, then it may be an option for me... here's hoping that all those 8kW owners will give me some feedback here...

I have an 8kW. Using your formula it averages around 48W. It's evening now so no PV and the battery is discharging, load around 550W. FYI I'm not on the latest firmware

Here is another view. You can see how inverter power increases as the load increases. Load scale on the right.

Edited March 21, 20242 yr by slipx

I have an 8kW Deye

On low load it also stays on about 51/53W. I will check on the higher loads later today.

I found that the Sunsynk dongle reports a way higher (incorrect) self consumption than what it should vs Solarman dongle ... here is my friends 8kw sunsynk, his 'self consumption' is always around 200-300w which cant be right.

1226+144+19=1389-1175= 214W

@slipx, @GVC and @Nexuss, thanks for your reports, ok, I should have added that I'd want to see what the self consumption is, when there is no grid, during load shed, or like in my case, when there is no grid, period, I need to create a new entity, so I don't have to do the calculation, right now, pretty much clear sky, I get...

so, 3402 - ( 2189 + 1076 ) = 3402 - 3265 = 137W self consumption on the 5kW Sunsynk...

Let me see, pv_total (in my case) + SS Battery power - SS Inverter power, should give me any time of day or night the self consumption, I think... it would be good to be able to draw a graph and see what is going on...

For others that want to try and do this, you'd need SS PV1 power + SS PV2 power + SS Battery power + SS Grid power - SS Inverter power, to come up with a reasonable self consumption figure, I think, but I am really only interested for now where the SS Grid power = 0, since there won't be a grid here again... ( and on the 8kW there may be PV3 etc. as well, not sure...)

Also any opinions in Deye vs Sunsynk? the Deye is selling for less here and there, so would be the preferred device, since I am counting the cents here 🙂 I would assume, other than the firmware and thus user interface, they are otherwise identical, is my thinking correct here? and apologies to @007 for hi-jacking his thread somewhat...

137W seems high.  When its blending from the grid i have seen as low as 9-30W which is pretty crazy. Right now with no grid i get 46W.

• MAIN：3384-086B
• HMI：0000-E42A

• Total DC Input Power：4.604 kW
• Total Consumption Power：2.15 kW
• External CT Power L1L2：0 W
• Battery Power：-2.408 kW

26 minutes ago, Kalahari Meerkat said:

Also any opinions in Deye vs Sunsynk? the Deye is selling for less here and there, so would be the preferred device, since I am counting the cents here 🙂 I would assume, other than the firmware and thus user interface, they are otherwise identical, is my thinking correct here?

I would go with the Deye if its cheaper , The sunsynk dongle has not impressed me much and its servers are down pretty frequently lately compared to solarman which basicly never goes down. I am very happy using solarman as it can do everything the sunsynk dongle can. As far as the hardware goes its the same.

@Kalahari Meerkat My template entity is based on the following. I don't use grid, added to the graph below. Seems like its all over the place

{{ states('sensor.sunsynk_totalsolar') |int - (states('sensor.sunsynk_inverter_power') |int - (states('sensor.sunsynk_battery_power') |int)) }}

@slipx you would have to add battery power, I'd imagine, in my case when the battery charges, the power is negative, thus add the negative and you have inverter power + inverter self consumption remaining...  graph for a few minute here now...

 

the self use is generally around 120 to 150W, although spikes down to -1108W (not possible that it generates power) and up to 1601W are in there, but also can't imagine that the inverter would self consume quite that amount either...

the below may or may not be correct I ain't a horses-ass^Whomeassistant boffin by any means... but it produced the above graph...

      ss_self_use:
        device_class: power
        unit_of_measurement: W
        value_template: >-
          {{ states('sensor.pv_total') | float + states('sensor.ss_battery_power') | float - states('sensor.ss_inverter_power') | float }}

also note the clouds messing about me trying to get the battery full after 3 days of only partial charge... purple graph should not have the dips it is showing ...

Edited March 22, 20242 yr by Kalahari Meerkat
also...

@Kalahari Meerkat using your template above it's currently averaging 142W

 

With inverter power

Edited March 22, 20242 yr by slipx

2 hours ago, slipx said:

@Kalahari Meerkat using your template above it's currently averaging 142W

Dang, and that's a 8kWer?

Yes but is varies depending on what the inverter is doing e.g. averaging 37W now 

I recently discovered a very interesting situation in our solar system after the installation of 2 additional batteries.  The system now consists of a Sunsynk 8kW single phase inverter, 4 x 5.3 kWh batteries, 16 x 545W panels.  The inverter was set to accept 0W of trickle feed from the grid.

After the installation of the additional batteries the Sunsynk app showed that the household used 45 kWh for the month of February, but the utility bill showed that we utilised about 150 kWh of power from the grid.  I then started monitoring my electricity meter and app on a daily basis and found that the electricity meter still showed around 5kWh of grid consumption per day on days where the Sunsynk app showed zero grid consumption.

Some research on the Internet pointed me to a possible problem with the CT coil, which was connected with 30m of flex cable running adjacent to the 230V cables to the inverter (similar thickness to the original black and white cable attached to the CT coil supplied with the Sunsynk inverter).

I installed a new cable from the CT coil to the inverter.  It was a CAT 6a foil shielded cable where the 4x internal white cables were joined together to form one conductor, and the 4x internal coloured cables were joined to form the other conductor.  It would have been ideal to ground the shield of the CAT cable on the earth connection of the house, but my house has a neutral to earth bond which caused problems, and the shield was left unearthed.  The CAT cable was also installed so that it did not run directly next to the 230V cables to the inverter.

The results were excellent:  in a 24-hour period where the Sunsynk App showed zero grid consumption the actual consumption via the electricity meter dropped to around 300W for the same period.  This result was also confirmed over the next 7 days where the App showed zero consumption from the grid and the actual consumption remained around 300W per day.

On 2024/03/19 at 1:29 PM, Patrick OReilly said:

I find it surprising that the unit does not even have any cooling fans of its own.

Does anyone have any wisdom to share on improving the cooling on these 5.5KW SunSynk units?

@Patrick OReilly I in stalled 2x 120mm fans at the bottom on a timer to control the temperature and it works well