diy solar

diy solar

Growatt constant draw from the grid

slychocobo

New Member
Joined
Mar 8, 2023
Messages
4
Location
UK
Hi, New to the whole Solar PV life, I tried searching but couldn't find anything that quite matched my question so I hope this is not a duplicate.

Recently had a Growatt SPH3000 inverter installed at my home the in roofspace along with 10x panels for about 4.2KwH generation, EPS for backup/UPS, 6KwH battery for storage and an 'Eddi' water tank heater for excess generation as our energy provider has a pretty lousy rate for feedback.

Generally the system works quite well, everything seems to be running as expected but we are seeing two odd behaviors we cant seem to pin down and honestly we don't know if its just oddities of how these AIO systems work or if something is wrong and could use an experienced hand to give advice.

1. We are often drawing power from the grid, even when there seems to be no good reason for it.
To give an example of what I mean. We had a situation where we were generating 1KwH from the panels, the battery has a solid charge of about 90%
The Inverter itself have a good charge in it and our house load according to the inverter will be around 400w, The Growatt will report (from the inverter display) that we are exporting -2/4w's, the rest is going to the battery.
But if I check the smart-meter itself, it will constantly show we are drawing 60w+ from the grid pretty consistently.
Even under low loads, around 300w with no solar, the inverter will say the house load is 300w, the battery is supplying 300w, but we will still be drawing 40-80w from the grid.

We've also noticed that the Inverter and the Eddi (Which both get their readings from CT clamps attached to the same cable (which is outside the consumer unit) will give quite different readings. While the Growatt will think there is no load at all (or that it's exporting slightly) the Eddi CT clamp will be reporting the 'correct' value that we are drawing 60w from the grid so we are working under the assumption that this mis-reading by the inverters CT clamp is the cause of the problem?

We arnt sure if the CT clamp might be at fault, or perhaps the make/cable difference is a factor (Both are wired up using Cat5e twisted pair cabling, but the CT for the Inverter has to travel up from the ground floor to the roofspace, while the Eddi is a few feet away) or if this is a wild goose chase we are on.

2. The other issue is also an odd one
We've noticed that on days of low solar production and the batteries run down they will reach about 10% then stop supplying the property with power and we'll go full back to the grid.
My understanding is that this is normal behavior to help maintain the life of the battery (although 10% seems high?). The problem however, is that battery will (quite slowly) over hours continue to discharge to about 8% battery life. The inverter then seems to switch over and will start charging the battery from the grid until the charge level reaches 12%. It then switches back to normal operation and discharges the battery to supply the property till it hits 10% again then stops.
This cycle then repeats till morning when the panels start generating power again.

We are unsure why this is happening, we are looking to switch to a variable rate tariff so do not want to risk the battery suddenly charging from the grid at 'peak' hours and costing us a small fortune.

But anyone with more experience with these systems or ideas, would be appreciated :)

Thank you
 
There is another thread regarding growatts allowing the grid to supply more power than the screen is saying. I think the conclusion was to get a separate limiter box and CTs that the inverter can use. It’s more accurate. Most grid tie inverters allow the grid to supply some power so that backfeed into the grid isn’t occurring. I think the inverters are configured so that the it thinks it’s covering all the load when it’s not so that backfeed into the grid doesn’t happen. kinda like a safety threshold to ensure no backfeed. It’s better than back occurring when the inverter says it’s not backfeeding.
your other issue can most likely be solved by adjusting settings.
 
Hm... Thanks for replying. Interesting about the Inverter purposefully drawing from the grid, that would go to explain the effect we we are seeing to a degree, but we are a little surprised by how much. I think on average its about 1.5/2KwH a day.

I'm not quite clear on what you mean by a 'seperate limiter box and CT' though, I've seen some other posts around this that mention a "Eastron SDM230" which seems to be a digital version of the CT clamp (although it seems to be less a CT clamp as its wired in-line with the Grid-in feed)
Or do you mean something else?
 
I have just fitted the Eastron SDM230 and it has massively cut the continual feed in of 50-100W down to 0 to -10W. In the process of fitting it I discovered the CT clamp was actually broken across the ferrite 'C' core and making an audible buzz like a transformer (which it is). Carefully and accurately glueing it back together reduced the feed in down to around 50W ( from 100-120W before repair).

At the same time the measurements displayed by the Shine App for power in/out for the inverter were different compared to the readings from both the DNO meter (via the in home display) and the Myenergi/Zappi app - both of which agreed with each other. Changing to the Eastron meter now shows all readings to be identical - so a really good move.

Not sure about the charge/discharge thing as mine charge at the moment via cheap rate at night (2130-0130) to 100% then discharge overnight until sunup then charge as required via solar or discharge if no sun. In the depths of winter they did deplete to 10% a couple of times but never saw them charge from the grid outside of the 2130-0130 slot.

Maybe there is an incorrect setting somewhere?
 
Hm... Thanks for replying. Interesting about the Inverter purposefully drawing from the grid, that would go to explain the effect we we are seeing to a degree, but we are a little surprised by how much. I think on average its about 1.5/2KwH a day.

I'm not quite clear on what you mean by a 'seperate limiter box and CT' though, I've seen some other posts around this that mention a "Eastron SDM230" which seems to be a digital version of the CT clamp (although it seems to be less a CT clamp as its wired in-line with the Grid-in feed)
Or do you mean something else?
the eastron is what I mean
 
OK, Sounds like this Eastron is the way to go.. Will see if that helps the situation and hopefully report back :)
 
Just a little follow up on this really, The Eastron has made things notably better, the vampire draw we were seeing before is gone. As for the strange draw behaviour, apparently (according to Growatt) is 'expected' if the battery levels drop too low.

Oh well!
 
I am about to fit an SDM230 as having the same issue with parasitic draw and potentially a long ct cable to the inverter being a culprit. Did you just cut the CT clamp wires and wire to the +/- on the SDM230? Or did you but a new RJ45 cable? I assume that the CT RJ45 cable to the inverter has the same wiring configuration at the terminals at the RJ45 pins, so i can just snip the wires to the CT and wire direct to the SDM230?

My inverter settings are also set for a discharge limit stopping at 10%. I guess this relates to the usable capacity of the batts being 6kw instead of 6.5kw... or have i lost 500w to begin with and then another 10%? ?
 
I have just fitted the Eastron SDM230 and it has massively cut the continual feed in of 50-100W down to 0 to -10W. In the process of fitting it I discovered the CT clamp was actually broken across the ferrite 'C' core and making an audible buzz like a transformer (which it is). Carefully and accurately glueing it back together reduced the feed in down to around 50W ( from 100-120W before repair).

At the same time the measurements displayed by the Shine App for power in/out for the inverter were different compared to the readings from both the DNO meter (via the in home display) and the Myenergi/Zappi app - both of which agreed with each other. Changing to the Eastron meter now shows all readings to be identical - so a really good move.

Not sure about the charge/discharge thing as mine charge at the moment via cheap rate at night (2130-0130) to 100% then discharge overnight until sunup then charge as required via solar or discharge if no sun. In the depths of winter they did deplete to 10% a couple of times but never saw them charge from the grid outside of the 2130-0130 slot.

Maybe there is an incorrect setting somewhere?
Ever since my Growatt Inverter was installed I was seeing the same thing as you, although about 100-150w of excess draw. So glad I found this forum as I wasn't sure what was going on.

On an evening, with the battery at >80% and the baseload from the house at ~200w we would see a draw of >100w. I would wake up to find we had imported 1kWh overnight.

Measurements on my smart meter IHD (assumed to be correct) vs. those reported by ShinePhone app or http://server.growatt.com.

I've taken note of your very useful comments and had an electrician install an Eastron SDM230 V2 Modbus metering device in a separate enclosure. Wiring is Main Fuse > Smart Meter > Isolator Switch > Eastron SDM230 > Consumer Unit > Growatt Inverter.

What I'm seeing now from watching the display on the Eastron SDM, it is hovering around 0, which is correct behaviour when the house is idling. However, the smart meter is STILL showing >100w of draw. Do I have a faulty smart meter???
 
Last edited:
Back
Top