EG4 new AIO rated 12K output and 18kPV aka "EG4-18Kpv-12LV"

[cs1234]

Personally, for a stable system that isn't being rebuilt every week, I can't imagine preferring open loop. Just having an accurate SOC on the inverter would be worth it.

Sure, depending on what features you have enabled in an AIO , accurate SOC can be very useful. Say it's controlling microinverters and doing zero or limited export and you want to push it really close to the edge of battery capacity. Some manufacturers manage to do it without requiring closed loop, but it is kind of kludgy compared to trusting the BMS to report an accurate SOC and capacity.

 

[Zwy]

So now it's the charger making the decisions for all the BMS's? Or the BMS's collectively working as a team? Everything is more complicated with more BMS's.

In a perfect world, we shouldn't be using multiple BMS's. There should be only one, connected to all the cells we want to make up total pack volume. It would have powerful active balancing and be stronger than steel, able to stop a speeding bullet! A SuperBMS if you will.

Batrium does it. But if it trips the breaker, all power is cut. Balancing is less than 1A but I don't see why you need more if the SCC is set properly.

Which can be a good thing or a bad thing depending on your perspective.

All controlled thru the Batrium.

 

[Zwy]

Saying you're going to be on a DIY forum and expect a manufacturer to honor a warranty is like being on a car tuner forum and bitching that Toyota won't warranty a blown motor you tinkered with.

If an item is defective right out of the box, one should not expect a warranty?

Some defects are not known until an item has been in use.

 

[timselectric]

My AIO's don't need to know what the SOC is.
Charging is based on voltage. That's all that it needs to see. I also prefer the settings based on voltage.
Strictly off grid, here. No export. No AC coupling.

 

[robby]

With BMS communications. The BMS tells the charge controllers when and how much to charge.
Which in my case, it put the charger in crawl mode above 80%.
And some days never got fully charged.
No thank you.

Agree that it tells it how much to charge and I disagree that this is a bad thing.
What I am pointing out is that all of that is based on the pack level Voltage or SOC. The inverter has no way of knowing if one cell is cooking off.

 

[timselectric]

Agree that it tells it how much to charge and I disagree that this is a bad thing.
What I am pointing out is that all of that is based on the pack level Voltage or SOC. The inverter has no way of knowing if one cell is cooking off.

Correct, it doesn't provide individual cell information.
It just controls the charging accordingly.
Correctly?
Sometimes
Maybe

 

[kennyb]

The battery test is the 9450A. To my knowledge they use an external heatsource to simulate a thermal event and track the progress of flame or lack thereof spread.

UL9450, my understanding is the test for the whole ESS. Ie batteries and inverter. But they have to have been passed as pairs.

This only matters in jurisdictions like Canada where it's now required for installations.

UL-9540 (A)
Seems to require the battery to pass the fire test without closed loop communication.

UL-9540 (Ed.2) talks about ESS having an Inverter that is tested together with it and certified together but I see nothing about closed loop comms being required. Did someone else find a section that requires that? These documents are extremely boring to read.

I am still looking at the other certifications in regards to the Inverter itself.

I have never seen an Inverter that can read battery voltages at the individual Cell level.
The only reporting from the BMS is the total Voltage, Current, SOC and Temp. All individual cell management I have observed is done by the BMS itself.

If one cell is at 3.75V and the other fifteen are at 3.30V the INVERTER treats the pack as though it is at 53.25V and charges the cells the same way as if all 16 cells were at 3.328V. It is the BMS itself that will try to discharge the one high cell into the lower cells and balance back the pack.
I am open to seeing or hearing something different but I have not personally seen or read of an Inverter doing a single thing to fix or stop an issue with a bad cell. If someone has seen something different I would love to hear about it.

 

[rhino]

What I am pointing out is that all of that is based on the pack level Voltage or SOC. The inverter has no way of knowing if one cell is cooking off.

I don't understand the point of this statement since a BMS in an open loop system would have disconnected the battery before then. It would have disconnected in a closed loop system as well so I don't see how one is more advantageous then the other in this case.

 

[Zwy]

UL-9540 (A)
Seems to require the battery to pass the fire test without closed loop communication.

UL-9540 (Ed.2) talks about ESS having an Inverter that is tested together with it and certified together but I see nothing about closed loop comms being required. Did someone else find a section that requires that? These documents are extremely boring to read.

One thing I've looked for an can not find is the UL listings for EG4 batteries.

UL 1973? UL 1741?

Second, the UL-9540 SE limits battery bank size to 20Kwh non commercial and 50Kwh on a commercial system. I don't know about you but this is unrealistic, I'm already over the commercial size. So what good does a UL-9540 SE cert do if the system size has to be so small?

I started watching this video but not all.

I am still looking at the other certifications in regards to the Inverter itself.

I have never seen an Inverter that can read battery voltages at the individual Cell level.
The only reporting from the BMS is the total Voltage, Current, SOC and Temp. All individual cell management I have observed is done by the BMS itself.

If one cell is at 3.75V and the other fifteen are at 3.30V the INVERTER treats the pack as though it is at 53.25V and charges the cells the same way as if all 16 cells were at 3.328V. It is the BMS itself that will try to discharge the one high cell into the lower cells and balance back the pack.
I am open to seeing or hearing something different but I have not personally seen or read of an Inverter doing a single thing to fix or stop an issue with a bad cell. If someone has seen something different I would love to hear about it.

 

[houseofancients]

With BMS communications. The BMS tells the charge controllers when and how much to charge.
Which in my case, it put the charger in crawl mode above 80%.
And some days never got fully charged.
No thank you.

so you didnt have it setup correctly tim , is that what you are saying ?

 

[cs1234]

All controlled thru the Batrium.

View attachment 148538

How is the batrium controlling the ones with no wires? What kind of magic is this?

 

[cs1234]

Agree that it tells it how much to charge and I disagree that this is a bad thing.
What I am pointing out is that all of that is based on the pack level Voltage or SOC. The inverter has no way of knowing if one cell is cooking off.

If it gets cell voltage readouts on the closed comms, then it does know one is cooking off it knows cell chemistry. It just can't do anything about it other than cut charge to the BMS. The BMS still ultimately has to do what it is supposed to do. All of this could be very bad though, if using other chargers, other batteries, etc.. on the circuit. In that situation, you don't want any type of dumb decisions being made on anything other than overall circuit voltage by each individual device.

 

[robby]

I don't understand the point of this statement since a BMS in an open loop system would have disconnected the battery before then. It would have disconnected in a closed loop system as well so I don't see how one is more advantageous then the other in this case.

Exactly, which is why the inverters I have seen don’t even seem to care about cell voltage. So i don’t see how closed loop adds any value to fire protection from the point of a single bad cell cooking off.

 

[timselectric]

so you didnt have it setup correctly tim , is that what you are saying ?

Correct, I was using communications. lol
Once I pulled that crap , everything was great.

 

[timselectric]

I don't understand the point of this statement since a BMS in an open loop system would have disconnected the battery before then. It would have disconnected in a closed loop system as well so I don't see how one is more advantageous then the other in this case.

If it worked correctly.
Closed loop should slow or stop charging. Before a cell goes off script. Because the BMS is in control of the charging.
If it actually worked correctly.

 

[Quattrohead]

I have never seen an Inverter that can read battery voltages at the individual Cell level.

Come on Robby, don't make me do this to you.... Solark. ?

 

[Zwy]

How is the batrium controlling the ones with no wires? What kind of magic is this?

That was during the build.

Finished here.



Complete system although the EG4's aren't there now.

 

[Zwy]

More on UL 9540 SE https://collateral-library-producti...4/CT26157086_UL9540A-whitepaper_vDIGITAL1.pdf

How many of you will intend to stay at a 20Kwh max size for ESS?

 

[SparkyJJO]

Heck I'm at 22kWh and I think it way small (but all I can reasonably afford right now).

They want to limit it so you're not too self sufficient.

 

[Zwy]

Heck I'm at 22kWh and I think it way small (but all I can reasonably afford right now).

They want to limit it so you're not too self sufficient.

You're catching on........

So what good would it do to purchase a system that meets UL 9540 SE if you had to stay under the ESS size limit?

 

[timselectric]

You're catching on........

So what good would it do to purchase a system that meets UL 9540 SE if you had to stay under the ESS size limit?

That wouldn't align with the desire for 100% EV adoption.

 

[Lt.Dan]

I personally plan to get to the 50-65kWh range. No way I'm staying with 20kWh.

 

[robby]

You're catching on........

So what good would it do to purchase a system that meets UL 9540 SE if you had to stay under the ESS size limit?

Get Inspected and certified and then add more batteries.
It really is a stupidly low number, but I now realize why Tesla was pushing two Powerwalls in their packages.

 

[Rusty959]

Another big problem I have with the open loop idea and "the bms will shut off if needed" is it ignores real world situations such as inductance.

For example, why is a charging resistor so important when connecting a battery to an inverter? I mean, it's just a bit of a spark, why not just close your eyes and go for it?
In reality it's not just a spark, the huge inrush plus inductance means you will have massive overshoot, easily severely overvolting the battery input.

Similarly, if a battery is charging at a high rate and all of the sudden the BMS shuts off the input, weird stuff can happen. Not even the same weird stuff each time. Additional surges, overvolts, spark gaps crossed, and potentially blown equipment. I can write out some examples but I started getting too wordy.

 

[Rusty959]

One thing I've looked for an can not find is the UL listings for EG4 batteries.

UL 1973? UL 1741?

Second, the UL-9540 SE limits battery bank size to 20Kwh non commercial and 50Kwh on a commercial system. I don't know about you but this is unrealistic, I'm already over the commercial size. So what good does a UL-9540 SE cert do if the system size has to be so small?

I started watching this video but not all.

What do you mean you "can not find", do you mean you haven't looked?