Is This Normal Behavior For EG4 6500 and Other EG4 Inverters.

[WindWizard]

I know this guy must be a member on the forum, but I am wondering is this kind of constant Grid Power consumption normal when in this mode of operation.

The thing I noticed that I do not like is that no matter what was going on whether he was using the Grid or charging the batteries, there seemed to be around 300 to 400 watts that was continually used by the system which seemed to be just power that the System required just to run the Inverters and etc. I stopped the video a number of times and got out my calculator and took the amount of solar power coming in and then subtracted what was going into the batteries and the loads and there were at least 300 watts of unexplained power consumption. If you need 300 watts continuously that is 216k watts a month of power that is wasted. My power costs me 22 cents per kw which adds up to $47.52 a month.

 

[timselectric]

The thing I noticed that I do not like is that no matter what was going on whether he was using the Grid or charging the batteries, there seemed to be around 300 to 400 watts that was continually used by the system which seemed to be just power that the System required just to run the Inverters and etc. I stopped the video a number of times and got out my calculator and took the amount of solar power coming in and then subtracted what was going into the batteries and the loads and there were at least 300 watts of unexplained power consumption. If you need 300 watts continuously that is 216k watts a month of power that is wasted. My power costs me 22 cents per kw which adds up to $47.52 a month.

Those discrepancies aren't inverter losses.
They are inaccurate reporting from the inverters.
Each inverter is using around 65w.
The wattage pulled from the grid is applied to the loads.

 

[PreppenWolf]

Using a clamp meter on my system right now "standby" usage of the grid fir each 2P1 and 2P2 is max 0.64A and min 0.00A. The majority of the time if I leave it on live reporting is 0.55A for each unit.

 

[WindWizard]

So that explains 130 watts of the 300 watts. It could be that the inverter is not reporting correctly. It would be interesting to measure the Solar power coming off the panels with a clamp meter, what the house in consuming with a clamp meter and what is going into the batteries with a clamp meter to see how much power is being lost due to inefficiency.

 

[PreppenWolf]

The inverter is likely reporting raw incoming power, not power converted and then sent to a load?

Remember if you have everything in your house off, the system still requires power to operate and be on. There is no escape from this.

6500s are not the pinnacle of efficiency. They are a compromise between cost and functionality.

 

[Zwy]

The thing I noticed that I do not like is that no matter what was going on whether he was using the Grid or charging the batteries, there seemed to be around 300 to 400 watts that was continually used by the system which seemed to be just power that the System required just to run the Inverters and etc. I stopped the video a number of times and got out my calculator and took the amount of solar power coming in and then subtracted what was going into the batteries and the loads and there were at least 300 watts of unexplained power consumption. If you need 300 watts continuously that is 216k watts a month of power that is wasted. My power costs me 22 cents per kw which adds up to $47.52 a month.

Is it raw PV power being reported or the actual PV power converted by the SCC? Most MPPT's convert at around 93-95% efficiency. Some claim higher but it could depend on voltage vs amps coming in. More amps, more heat loss. Higher voltage, less amps for same watts = less heat.

 

[timselectric]

The watts reported by these voltronics inverters are just not very accurate. Close, but not exact.
However a shunt is very accurate. Which is what is reporting the battery wattage in that video. So you can't really calculate the losses, exactly. Because the readings are coming from two different measuring devices.

 

[Adam De Lay]

I’ve found a few things out since making that video.

• After having some discussions/complaints registered with SS, I’ve found out that any time your inverter is in “utility” mode (whether that’s using SBU or SUB), all the solar power your system is producing doesn’t always go to the loads first (as the output priority states). Even though the modes and the manual dictate Solar, then Utility, then Battery as your output priority, apparently it doesn’t work that way.
  
  According to SS: “In SUB mode, the inverter views Solar and Utility as its main sources of power, so it finds the most consistent, reliable source of power to power loads. In this case, it's utility power. as the voltage doesn't fluctuate as often, and it is available in the evenings as well. We recommend setting your inverter to SBU so that the inverter will use solar as its source to power your loads.“
  
  Supposedly they’re supposed to be working on a firmware version to correct how solar is used when you’re in SUB mode so it properly handles the output priority specified.
  
  So that might be one reason why there’s a discrepancy in load vs grid vs solar.
  
  
• Something else I discovered was regardless of how many inverters you have set up in Solar Assistant, you’re only gonna see cumulative information on the dashboard. I have 2 inverters in my environment and each has 2 separate pv arrays. At any given time, you might have more solar coming in on inverter 1, but more of your loads are coming from inverter 2. In that instance, you’ll see power still being pulled from the grid while there’s power being put into the batteries (Because the battery power is being put in from inverter 1, but the grid is pulling power for inverter 2). Hope that makes sense.
  
  
• The last thing I noticed was the different values in Solar Assistant don’t seem to all update at the same time. PV might update, then a second later the loads update. You’ll really see this when you’re on your Inverter Cluster page in Solar Assistant. So it’s kinda hard to be able to say for certain if the values you’re seeing are exactly the values at the moment when you’re trying to do any manual calculations. Whether that’s Solar Assistant updating slowly, or the inverter’s passing the data to SA slowly, I’m not sure.

 

[Zwy]

Update on this regarding the mystery 60W draw.

 

[timselectric]

The last thing I noticed was the different values in Solar Assistant don’t seem to all update at the same time. PV might update, then a second later the loads update. You’ll really see this when you’re on your Inverter Cluster page in Solar Assistant. So it’s kinda hard to be able to say for certain if the values you’re seeing are exactly the values at the moment when you’re trying to do any manual calculations. Whether that’s Solar Assistant updating slowly, or the inverter’s passing the data to SA slowly, I’m not sure.

It's the inverters.
Solar Assistant reports what the inverters provide, as they provide it.
My Victron Smart Shunt is the fastest reporting device in my system. It's very close to real time.
The inverters are slower and out of sync with each other.

 

[Adam De Lay]

Update on this regarding the mystery 60W draw.

Unfortunately it still doesn't explain the draw though. Since the Victron Shunt is what's used to track power usage in Solar Assistant, anything that's wired before the shunt won't be tracked. With the test charge controller and raspberry pi wired before the shunt, those couldn't have been the draws. The only thing after the shunt was the inverters.

So I know I've seen people stating that the inverters (when the grid input is hooked up) will have a constant grid draw in order to allow for quick switching over to the grid. I've never seen this draw reflected in Solar Assistant when I was in battery mode, but could that be the reason for the draw? Not sure. Really hope not.

 

[timselectric]

Unfortunately it still doesn't explain the draw though. Since the Victron Shunt is what's used to track power usage in Solar Assistant, anything that's wired before the shunt won't be tracked. With the test charge controller and raspberry pi wired before the shunt, those couldn't have been the draws. The only thing after the shunt was the inverters.

So I know I've seen people stating that the inverters (when the grid input is hooked up) will have a constant grid draw in order to allow for quick switching over to the grid. I've never seen this draw reflected in Solar Assistant when I was in battery mode, but could that be the reason for the draw? Not sure. Really hope not.

Solar Assistant only reflects what the inverters report. And we know that they don't report everything. (Especially self consumption)
You can check with an amp meter.
I don't remember ever checking this. And I'm no longer connected to the grid.

 

[Adam De Lay]

Yeah, kinda what I figured.

Solar Assistant only reflects what the inverters report. And we know that they don't report everything. (Especially self consumption)

Yeah, kinda what I figured. Especially since you have to program in the self consumption into SA.

 

[Zwy]

Unfortunately it still doesn't explain the draw though. Since the Victron Shunt is what's used to track power usage in Solar Assistant, anything that's wired before the shunt won't be tracked. With the test charge controller and raspberry pi wired before the shunt, those couldn't have been the draws. The only thing after the shunt was the inverters.

SA won't integrate with the Victron shunt it appears. One of the shortcomings of SA.

When I watched your video, I saw the shunt right away and said to myself, "Why aren't all the loads on a busbar after the shunt?"

I've been there myself sometimes, can't see the forest for the trees.

So I know I've seen people stating that the inverters (when the grid input is hooked up) will have a constant grid draw in order to allow for quick switching over to the grid. I've never seen this draw reflected in Solar Assistant when I was in battery mode, but could that be the reason for the draw? Not sure. Really hope not.

As Tim posted, in order to prevent backfeed in SUB mode, there has to be draw. Inverter voltage is kept lower than grid so when grid is adding power all current would be to the loads and nothing backfeeds the grid.

 

[PreppenWolf]

As Tim posted, in order to prevent backfeed in SUB mode, there has to be draw. Inverter voltage is kept lower than grid so when grid is adding power all current would be to the loads and nothing backfeeds the grid.

This is a completely random thought. I wonder if the flickering lights some users see on EG4 AIOs is related to this function? It's possible those users have their units set to 110V or 120V but their house power is significantly lower than 120V. I've seen houses below 110V, like 107V. Is it possible the grid voltage is so low (or fluctuating) the AIO is trying to stay below to avoid backfeed and then "bouncing"?

 

[Adam De Lay]

SA won't integrate with the Victron shunt it appears. One of the shortcomings of SA.

SA can read the information from the Victron Shunt. It just can't read anything that's wired up on the battery side of the shunt, but SA only shows the same information that displays in the Victron App for the shunt. Nothing wrong with SA, just the dummy that wired it all up wrong... :D

I've been there myself sometimes, can't see the forest for the trees.

I'm sure we all have. Just gotta slow down and take our time...

 

[timselectric]

This is a completely random thought. I wonder if the flickering lights some users see on EG4 AIOs is related to this function? It's possible those users have their units set to 110V or 120V but their house power is significantly lower than 120V. I've seen houses below 110V, like 107V. Is it possible the grid voltage is so low (or fluctuating) the AIO is trying to stay below to avoid backfeed and then "bouncing"?

Anything is possible. But I believe that the flickering complaints are with led lights. Most led drivers for 120v. Have a voltage range that starts at 70 or 90 v.
I'm not sure about other countries but US grids are generally within a 10% tolerance of nominal voltage.
So plus or minus 12v on 120v. (108 to 132v) which is well above the led drivers minimum.

 

[Adam De Lay]

This is a completely random thought. I wonder if the flickering lights some users see on EG4 AIOs is related to this function? It's possible those users have their units set to 110V or 120V but their house power is significantly lower than 120V. I've seen houses below 110V, like 107V. Is it possible the grid voltage is so low (or fluctuating) the AIO is trying to stay below to avoid backfeed and then "bouncing"?

I guess with these inverters, anything is possible. The fact that SA doesn't show information down to the second in its graphs wouldn't be any help in seeing that.

I guess I haven't payed much attention to it, but do most of the people that experience flashing/blinking lights have the grid connected?

I know I've been told by SS that if you have an extremely low draw on a phase and you turn on an LED light, it could flicker till you have more load on that phase. I actually experienced that very thing on Sunday morning. Turned on the light over the vanity in the bathroom (SBU mode, on battery power) and the lights were flickering. Turned them back off and on again and still flickering. Turned on an additional set of lights and the flickering stopped. Tested several times and had the same results (just wish I would have had my camera to document).

 

[SignatureSolarJames]

Hello.

The wattage measured by a non- reactive power compensating meter is higher than what is actually used.

The amps measured are reactive power.

If you put an in-line kwh meter on your AC input side the idle consumption will be the listed idle on the spec sheet.

this is nothing to do with the inverter and everything to do with the meter being used; the grid load shown is a product of the 3rd party software doing the A x V math without realizing that PF is wrong.

@timselectric @robby @WindWizard

 

[timselectric]

Hello.

The wattage measured by a non- reactive power compensating meter is higher than what is actually used.

The amps measured are reactive power.

If you put an in-line kwh meter on your AC input side the idle consumption will be the listed idle on the spec sheet.

this is nothing to do with the inverter and everything to do with the meter being used; the grid load shown is a product of the 3rd party software doing the A x V math without realizing that PF is wrong.

@timselectric @robby @WindWizard

Is there a draw from AC input, while in SBU mode?
Idle consumption should come from battery, right?