EG4 18kPV stopped "inverting"

[455dan]

My neighbor and I are the last two houses at the end of the line up the mountain. We’re both off the same transformer and both read 253v-255v consistently. No issues though with power, bulbs/appliances or inverter.

Were they able to provide 200 amp service to both locations? Is the transformer larger than a standard single home unit?

Just interested on a standard transformers ability to provide two locations.

US grid (ANSI C84.1) allows 2 ranges for household AC V.
Range A can be designed for 240V +- 5% (252-228) from POCO while the electrical makers (utilization) can design to +5% -10% (252-220). Manufacturers can choose to go +10, but not narrower.

Range B allows a couple V higher for both poco and manufacturer, but down to 208V for utilization, but doesn't apply to lighting loads.


My question for @EG4TechSolutionsTeam is why can't firmware be coded to disconnect grid relay if say grid V > 255V for x amount of time or some other reasonable figure?

Grid passthrough above 252V is non optimal since 254V is allowed, but then still passing through above 255V is a no no to put it politely.

Is there any room left on chip to add code that if grid V>255V, grid relay should open?


If this was me, I'd be calling poco just to have them come out and have a look at transformer just to be safe since it shouldn't be above 254V for extended periods of time.

Fnnwizard would that not require a full 200 amp relay controlled internal disconnect?

I get that the grid and load circuits can be broken by opening the manually actuated 18K pv Breaker/ disconnects, but can an 18K PV actually open the "full" 200 amp load and Grid circuits by itself?

I am showing my ignorance now, but thought these units only had control of their inverters output amperage not the full pass through rated amperage other than the manual disconnects?


ps. Have not seen a reference yet, but is anyone aware of a schematic or any tear down look into what's under the cover of an 18K PV.
After a full year no one has opened one up but understand, because that would nullify the warranty.

 

[ChrisG]

Were they able to provide 200 amp service to both locations? Is the transformer larger than a standard single home unit?

Just interested on a standard transformers ability to provide two locations.

Would that not require a full 200 amp service breaker? I thought the 18K PV and the other Deye 12K units just use a buss and the disconnect is only for the power they can supply from the inverter itself. So throwing the grid breaker just disconnects the inverter from the pass through buss.

No idea on size but here it is. We are the last two houses up the mountain. I did call before and PoCo said this voltage expected. We each have 200a service.

 

[fnnwizard]

Fnnwizard would that not require a full 200 amp relay controlled internal disconnect?

I get that the grid and load circuits can be broken by opening the manually actuated 18K pv Breaker/ disconnects, but can an 18K PV actually open the "full" 200 amp load and Grid circuits by itself?

I am showing my ignorance now, but thought these units only had control of their inverters output amperage not the full pass through rated amperage other than the manual disconnects?

ps. Have not seen a reference yet, but is anyone aware of a schematic or any tear down look into what's under the cover of an 18K PV.
After a full year no one has opened one up but understand, because that would nullify the warranty.

The grid connect relay is rated 200A x 830V.

Chances are small that it would have to break that connection at 200A for some, but no chance for me (max would be 15kVA, ~60A.) Most times I'm well under this power for this test system.

But even if that relay is a 1x use on breaking at max rating, I'd be fine since its not expensive or too hard to replace vs some expensive and delicate electronics like higher end audio.

I did open one up .

 

[455dan]

The grid connect relay is rated 200A x 830V.

Chances are small that it would have to break that connection at 200A for some, but no chance for me (max would be 15kVA, ~60A.) Most times I'm well under this power for this test system.

But even if that relay is a 1x use on breaking at max rating, I'd be fine since its not expensive or too hard to replace vs some expensive and delicate electronics like higher end audio.

I did open one up .

Wow THANK YOU fnnwizard

I have been looking but did not see your post. with pics

 

[455dan]

The grid connect relay is rated 200A x 830V.

Chances are small that it would have to break that connection at 200A for some, but no chance for me (max would be 15kVA, ~60A.) Most times I'm well under this power for this test system.

But even if that relay is a 1x use on breaking at max rating, I'd be fine since its not expensive or too hard to replace vs some expensive and delicate electronics like higher end audio.

I did open one up .

double post delete

 

[TM48]

If this was me, I'd be calling poco

This. I'd call the Poco and see if they will re-tap the xformer.

 

[sunsurfer]

US grid (ANSI C84.1) allows 2 ranges for household AC V.
Range A can be designed for 240V +- 5% (252-228) from POCO while the electrical makers (utilization) can design

My question for @EG4TechSolutionsTeam is why can't firmware be coded to disconnect grid relay if say grid V > 255V for x amount of time or some other reasonable figure?

Starting the grid passthrough above 252V is non optimal since 254V is allowed, but then still passing through above 255V should be a big no no to put it politely.

Is there any room left on chip to add code that if grid V>255V, grid relay should open?

I can't answer for EG4 but I would guess it has something to do with a safe operating window for having the higher local voltage (from EG4 drivers) for when there is less grid assist needed. I suppose they could make it throttle down more but there should be a decent window to work with. What happens if it hits 260v (for less grid assist) to over come grid voltage that already max? What gets fried in the house?

 

[fnnwizard]

I can't answer for EG4 but I would guess it has something to do with a safe operating window for having the higher local voltage (from EG4 drivers) for when there is less grid assist needed. I suppose they could make it throttle down more but there should be a decent window to work with. What happens if it hits 260v (for less grid assist) to over come grid voltage that already max? What gets fried in the house?

I would have to ask LuxP to find out more. Honestly, I don't think anything would fry in most homes if V is 130V. But I don't know for sure for all devices.

If I have an 18k running off ESS and it switches to grid because grid V is >152... that just doesn't make logical sense. What then is the upper limit?

I have a power conditioner though for audio stuff that will disconnect from the grid within .25 seconds if grid reaches 140V.

 

[Oldphile]

FWIW default disconnect setting for high voltage on my Sol-Ark 15K is 264 volts.

 

[crossy]

We have an under/over voltage relay on our incoming supply, power goes outside the set range it disconnects.

The village supply from PEA is a "nominal" 220V, it was always about 215V with occasional dips to <200V.

A couple of weeks back PEA came and installed nice fat 25kV cables and a shiny new (physically bigger) transformer. So, after a day of no village power the supply hasn't dipped below 238V (still within the specified +-10%)

If you're worried Amazon has various options, although most have single-pole relays so you'd need a contactor to open both L1 and L2.

 

[Quattrohead]

All the settings you need are right here -

 

[Dave911]

All the settings you need are right here - View attachment 252454

Can you change the Grid Connect High voltage above 252?
I believe that most of Indiana runs at the high end of +10% which is 264V. Thats the norm.
And that's common at my place. The grid is extremely stable here. We are 1/2 mile from a substation that feeds the entire northern portion of our county. Under voltage only happens during power outages which are very rare.

 

[ChrisG]

I have not checked my primary home in Charlotte area for voltage in quite some time but just did. It’s running about 255-256v also. So my cabin and primary home in different states and different PoCo’s are consistent.

 

[Dave911]

I have not checked my primary home in Charlotte area for voltage in quite some time but just did. It’s running about 255-256v also. So my cabin and primary home in different states and different PoCo’s are consistent.

250 to 260 is also common in the Detroit, Mi area. I take care of a house there.
I have always thought that the entire midwest typically runs runs near +10%. Brown outs are bad.
I have seen plant voltages run way over +10% at the 480 volt levels, usually at night when most of production is down.

 

[Quattrohead]

Can you change the Grid Connect High voltage above 252?

I don't know and I don't want to try it, I'm off grid anyway.

I believe that most of Indiana runs at the high end of +10% which is 264V. Thats the norm.

That is scary high.

 

[ChrisG]

Just called a buddy in NJ, they are running about 256-257v also. Seems there may be a new normal voltage in the U.S.?

 

[Oldphile]

In NH my grid is currently 245 volts with PV at 50% of rated. Don't forget when you back feed the grid, voltage rises.

 

[fnnwizard]

FWIW default disconnect setting for high voltage on my Sol-Ark 15K is 264 volts.

This is what I would expect and assumed for 18kpv too based on the GVL1H in the interface screenshot, but am not completely sure since there are other settings and question how those may interact with applicable 252V, 264V.

However, based on the way Eg4 answered it, makes me think they themselves don't know and is passing on incorrect info!
If 252V is in fact the cutoff (its not, it should be range of limit 1's), 18 should not make grid connection at 253V as shown in OP's 1st pic. So the issue with OP's 18 is not due to grid V but perhaps something else.

There are hysteresis bands at both ends if inverter disconnects due to grid V<> than the Limit 1s ( 211 & 264). (252V-264V and 211V-220V)

We should all press them for a more indepth explanation that should correlate with the LuxP manual. And if not, where/what is/are the divergence(s).

 

[Recovering EE]

The grid connect relay is rated 200A x 830V.

Chances are small that it would have to break that connection at 200A for some, but no chance for me (max would be 15kVA, ~60A.) Most times I'm well under this power for this test system.

But even if that relay is a 1x use on breaking at max rating, I'd be fine since its not expensive or too hard to replace vs some expensive and delicate electronics like higher end audio.

I did open one up .

I'm not seeing a lot of ventilation inside that box...

 

[EG4TechSolutionsTeam]

US grid (ANSI C84.1) allows 2 ranges for household AC V.
Range A can be designed for 240V +- 5% (252-228) from POCO while the electrical makers (utilization) can design to +5% -10% (252-220). Manufacturers can choose to go +10, but not narrower.

Range B allows a couple V higher for both poco and manufacturer, but down to 208V for utilization, but doesn't apply to lighting loads.


My question for @EG4TechSolutionsTeam is why can't firmware be coded to disconnect grid relay if say grid V > 255V for x amount of time or some other reasonable figure?

Starting the grid passthrough above 252V is non optimal since 254V is allowed, but then still passing through above 255V should be a big no no to put it politely.

Is there any room left on chip to add code that if grid V>255V, grid relay should open?


If this was me, I'd be calling poco just to have them come out and have a look at transformer just to be safe since it shouldn't be above 254V for extended periods of time.

Definitely a good suggestion that I will forward to our engineering team.

 

[fnnwizard]

I'm not seeing a lot of ventilation inside that box...

It is actually quite well laid-out/designed. All the heat producing components are on the backside in separate "chamber" that can't be seen, attached to proper heatsinks and forced airflowed.

I was negatively biased during the teardown but honestly didn't have much bad to say about it.

 

[ULR]

What's the distinction between the 1, 2, and 3 sets of grid voltage limits, i.e. what does each do?

 

[fnnwizard]

What's the distinction between the 1, 2, and 3 sets of grid voltage limits, i.e. what does each do?

1 (i abbrv as GVL1H 4 posts above) should be what LuxP defines as normal range but not according to EG4.

2 and 3 I have no idea, and, if setting any of these from default, will it affect anything else.

 

[Oldphile]

What's the distinction between the 1, 2, and 3 sets of grid voltage limits, i.e. what does each do?

Time that inverter should stay connected to the grid. The utilities call this ride through requirements.

 

[ULR]

Time that inverter should stay connected to the grid. The utilities call this ride through requirements.

i.e. Requirements that grid feeding generators must continue to feed for a bit if there's a short voltage droop. to avoid catistrophic grid collapse if a long-haul cable has a momentary short or the voltage sagged because of some temporary shortage of generation vs. load - don't everybody stop feeding and turn a glitch into a blackout. (I see that's mandatory for 1 MW + wind farms and big solar, but hadn't heard a version applied to little residential sites. Makes sense, though: A LOT of little guys dropping out when the juice dips is just as much of a problem as a few big guys. The grid gets nibbled to death by ducks.)

Can we get that in more detail, i.e.:
- Grid limit voltage high/low 1: If grid is outside this limit for xxx seconds the inverter (opens the bypass relay and contnues to power the local load?)
- Grid limit voltage high/low 2: If grid is outside this limit for yyy seconds the inverter (ditto or whatever - 180 db alarm? emits green smoke? blinks the house lights three times?)
etc.

Also: Are these settings mandatory, or partially mandatory (e.g. can't be narrower but may be wider. timeouts can't be shorter but may be longer)? Do they vary by jurisdiction (e.g. US vs. other countries due to different versions of the spec, etc.)