Benefits of Closed Loop Comms?

[chrisski]

I have “partial loop” communication with my fortress eFlex batteries.

This did pass the fire inspection and AHJ inspection.

This “partial loop” is all 8 batteries are in closed loop communication with each other not the inverter. I made this term up myself, but haven’t seen this battery only networking discussed elsewhere.

Overall, I don’t know what to think of “partial loop” but would give that a thumbs down. I have no way to “check” what the battery is doing and the algorithm is not discussed elsewhere.

I can monitor individual battery SOC by 4 LEDs on each battery, which oddly seems the most reliable method.

I can also monitor SOC through an inverter shunt, but oddly, this is totally unreliable.

On an app, I can also monitor SOC and voltage to the individual battery, but not cell level. This is more reliable than the shunt but there is always one or two batteries that are up to 50% off.

 

[Osmethne]

I have “partial loop” communication with my fortress eFlex batteries.

This did pass the fire inspection and AHJ inspection.

This “partial loop” is all 8 batteries are in closed loop communication with each other not the inverter. I made this term up myself, but haven’t seen this battery only networking discussed elsewhere.

Overall, I don’t know what to think of “partial loop” but would give that a thumbs down. I have no way to “check” what the battery is doing and the algorithm is not discussed elsewhere.

I can monitor individual battery SOC by 4 LEDs on each battery, which oddly seems the most reliable method.

I can also monitor SOC through an inverter shunt, but oddly, this is totally unreliable.

On an app, I can also monitor SOC and voltage to the individual battery, but not cell level. This is more reliable than the shunt but there is always one or two batteries that are up to 50% off.

That doesn't sound like partial loop. It sounds like no loop. Not unless the batteries talk to each other for whatever reason. Maybe to trade rumors.

I have a victron shunt and I've had good luck with it so far. Just as long as I charge to 100% once a week.

 

[Cmiller]

Closed loop gives SOC reading to inverter. This is needed for TOU settings to function well (voltage based doesn't work very well for partial discharge scenario). However, if the inverters these days could/would track SOC properly, then the closed loop wouldn't be needed for this, and also, most BMSs are sketchy at best on accurately tracking SOC if your batteries don't get full every day or 2....

Closed loop doesn't usually utilize float voltage, which is actually a bummer when someone has batteries that are purely backup for a grid outage, and therefore always full. You end up holding the batteries at a higher voltage than you would need to...

Closed loop can tend to cause inverters to nuisance shut down due to BMS reporting low SOC, when the BMS's SOC actually ran off...

Overall, as long as no TOU/partial discharge is happening, or if the inverter can accurately track SOC, then closed loop is more annoying than beneficial IMHO.

(Side note, Victron Smartshunt with Cerbo GX is an easy way to remote monitor SOC, and when set correctly is as accurate as any internal battery BMS. Another option would be to closed loop to a Solar Assistant in order to have remote capability of batteries without closed looping to inverter.)

 

[CommandoJoe]

They still make those things?

Indeed, and some are still made in the USA. Some of us actively seek them out... My last acquisition was 40kwh for $1000.

 

[Cmiller]

That doesn't sound like partial loop. It sounds like no loop. Not unless the batteries talk to each other for whatever reason. Maybe to trade rumors.

Not sure about other brands, but in this "partial loop/battery to battery comms" Rubix brand batteries will share SOC to master, and then all show average SOC on their screens. As an installer, the benefit there is that you don't get the question "why is are my batteries at 55%, 57%, 60%, and 56%"? Another "benefit" would be that if 1 unit has an error, all the batteries might shut down... causing power outages, but also notifying the end user that there is a problem...

 

[Cmiller]

Indeed, and some are still made in the USA. Some of us actively seek them out...

Personally, I think my savings in work pants alone would offset the extra cost of a few lithium battery banks. Maybe that is stretched a bit, but as an installer, I would never even remotely consider going back to lead for myself! Dealing with the acid part, the extra weight to handle, the losses in efficiency, the additional maintenance...... the list goes on and on.

 

[timselectric]

Closed loop gives SOC reading to inverter.

I guess that would depend on what inverter.
Both, my Deye and Growatts give SOC % in open loop.

 

[SeaGal]

most AIO inverters will let you charge from the grid as well, without using SOC, but voltage instead.

Yes, obviously. But without SOC information you can't, for example, charge to 85%.

Or, say, use the batteries to power the house until 35% is reached overnight.

 

[Cmiller]

I guess that would depend on what inverter.
Both, my Deye and Growatts give SOC % in open loop.

Are they accurate though? The Sol-Arks do as well, but they track horribly! As soon as voltage jumps a volt or 2 due to cloud effect, or drops a volt or 2 due to cloud cover or heavy load spike, they will jump to 100% or 0%. As you can imagine, that causes some problems when you are trying to do something like a "staged" TOU, with different SOC targets for different parts of the day!

 

[marionw]

I guess that would depend on what inverter.
Both, my Deye and Growatts give SOC % in open loop.

How can the inverter accurately calculate SOC based on voltage alone. Does the Inverter give you the option to reset SOC like the Victron shunt and does the inverter give you the other option the Victron shut provided that allow for a more accurate calculation of SOC

 

[ncsolarelectric]

Here's a reason. Certain manufacturers have made their products so they shut-down if comms are lost. This is true for some LFP batteries, Hybrid inverters, and Battery chargers. So the reason is, you want to use these products and it's easier and less frustrating to go with the flow than to fight the tide.
Seriously though, do you know how many things are in the NEC that fall into this category of "why bother"? We could nitpick stuff like this for decades, and they'll still add more every 3 years.

 

[empowered]

Yes, obviously. But without SOC information you can't, for example, charge to 85%.

Or, say, use the batteries to power the house until 35% is reached overnight.

sure you can, just extrapolate SOC to voltage...

 

[Solar Guppy]

From an operational view in normal operating conditions ( specifications adhered too ) the way its been done for over 100 years is fine so no surprise it is functionally ok to just monitor voltage and current.

The fireman code, better known as the national electric code ( NEC ) watched as planes crashed, ships went up in flames and sank, cars, busses and buildings go up in flames that can't be stopped thanks to the NMC chemistry. So they decided to use their advantage and added successively more regulations in the name of protection for structures and the ones the occupy them..

This is were closed loop communications come from, a means to prevent end-users from making unsafe conditions and manufactures to go beyond stating safe operational limits but actually enforcing them ( charge voltages, current, temperatures etc. ).

There is also double standard as orders of magnitude more destruction, thanks to NMC powered mobile devices, ebikes and the like but since the NEC can't regulate the contents of your home.

So if you can run open loop, makes no difference to the battery, never has, just remember when you run this mode you are the engineer making sure the system you built and installed not to exceed manufactures operation limits.

 

[Cmiller]

Yes, obviously. But without SOC information you can't, for example, charge to 85%.

Or, say, use the batteries to power the house until 35% is reached overnight.

Inverter brands like Victron can track SOC accurately enough (via Smartshunt) to be able to do this without using closed loop comms.

But then the likes of Sol-Ark are a disaster to try to rely on the inverter for SOC tracking...

So really the need for closed loop for SOC depends heavily on the brand of inverter being used.

 

[Chuck2]

?

 

[timselectric]

How can the inverter accurately calculate SOC based on voltage alone. Does the Inverter give you the option to reset SOC like the Victron shunt and does the inverter give you the other option the Victron shut provided that allow for a more accurate calculation of SOC

I have no idea how they do it. But they are usually pretty close.
They do have internal shunts, on all inputs and outputs.
Don't get me wrong, I only use the victron shunt for SOC. (It's my one truth)

 

[SunDave]

Sort of an idiot proofing measure.

Then I took off the training wheels.

I'm the idiot with training wheels. Been called worse

As with everything it is situational. The comms give me good data in Solar-assistant. The Pytes battery SOC count seems really accurate compared to most I have seen. As others stated above, the SOC based TOU settings are nice to have and intuitive.

With comms you can see individual cells in the pack, so instead of things going wrong, and relying on some kind of safety device to save the day, you can instead see it coming.

Plenty of advantages to communications for those that have a use for them. If you don't use them, great. I like being able to stick on percentage on something. Simplifies the thought process.

 

[CommandoJoe]

Personally, I think my savings in work pants alone would offset the extra cost of a few lithium battery banks. Maybe that is stretched a bit, but as an installer, I would never even remotely consider going back to lead for myself! Dealing with the acid part, the extra weight to handle, the losses in efficiency, the additional maintenance...... the list goes on and on.

Sounds like Big Lithium propaganda...

 

[JAS]

Just as long as I charge to 100% once a week

This! If I don't get a full charge, all my soc monitoring starts to deviate from one another

 

[Osmethne]

My question is..... why can't the BMS be smart enough to keep track of the time that it spends at a current value BELOW a known good low current threshold value and for that amount of time, utilize an equation that factors in live battery VOLTAGE to help autocorrect away from the error created by these low current usage time periods? The same way that a human would? If a man was sitting next to your battery bank all day, every day, with constant access to the current and voltage info, he could give you a VERY good SOC estimate. Why does the BMS have to be so much dumber than this man?

Someone decided that it was all or nothing logic. EITHER...... it tries to track any and all current flows and add or subtract those from the capacity of the battery at the point of last calibration.... OR.... it just stares at voltage, which on Lifepo4 is flat and high current situations give less than stellar results. Why can't the BMS use BOTH things in the calibrations for SOC? Use the shunt for the times when current is high (avoiding the time when voltage is the least accurate) and use the voltage for the times when the current is low (avoiding the time when the shunt is the least accurate)????

Why can't the SOC value be a multivaried value?

This has been a question of mine for as long as I have been playing with these closed loop systems like this that float WILDLY out of reality if not cycled deep and fully charged regularly.

The explanation I've heard is that most BMS systems can't detect current accurately below a certain threshold, I think it has something to do with saving money on good CTs. So all those times when your system is barely using any power sort of throw things out of whack.

 

[TM48]

I do think closed loop has a place in systems for people with no inclination to understand how it works

I see this said all the time but it's a totally lame argument. It takes all of 30 minutes (and that's if you're a slow reader) to learn enough to set up via voltage. It's not a deep subject. For me it gets down to why give up directed shutdown..... IF... it works correctly closed loop. Most people run voltage because they have to with either self built batteries with unsupported / under supported BMS's or for economics have a mixture of different batteries.

 

[ksmithaz1]

Depends, On balance I think closed loop should be superior to rolling your own and winging it. The BMS in a battery should not be "training wheels" they should create a symbiosis between your inverter and your batteries. It isn't about "cutoff" is about management in general. I want my battery management controller to control it's batteries in concert with the inverter optimally. That being said, ... The problem is not closed loop in general, it's multi-fold. There is a severe lack of standardization in the various BMS systems, all the manufacturers really need a well defined specification to follow, like an ODBII. Thus if you want to use closed loop, you really need to try and stick with a single vendor for your batteries/BMS.

The biggest advantage to a BMS is tracking SOC. This is the #1 reason i want closed loop to my inverter. I want the inverter to be able to make sane decisions based on the state of charge of my batteries. This is well nigh impossible to do with just voltage, though one can make educated guesses, the BMS should do a better job. Further the batteries should be able to tell the inverter/mppt systems to ramp up or tone it down in the most optimal fashion if properly implemented.

The problems I've seen with closed loop are all related to (firmware?) issues with the BMS. For example the BMS on the EG4-Lifepowers will sometimes get sticky requesting very low charge rates indefinitely, fixable only by resetting the batteries. . . Annoying. Further the BMS can still just shutdown your batteries, in the case of no comm this could happen without warning. If the BMS is crap, then by all means you are better off just running on voltage and experience and educated guessing.

This stuff needs to be turn-key. Buy an inverter, snap in some solar panels, snap in some batteries, connect the power and go. The last thing i want to do is fiddle around with some manufacturers battery tool to 'balance' my batteries. The fact that 'closed loop' has generally sucked really bad is not an indictment of the idea, it's simply an indictment of the quality of the communiction around it. My inverter and the BMS on them should be managing my batteries not me, or some 3rd party tool. The BMS built into your battery should be able to make the best decision for the life and health of that battery period. If it does not, then the problem is not the closed loop communication.

<rant>
Lastly, and I have mentioned this multiple times. . . The inverter cannot just shut the f*ck off if it loses communication with the batteries in closed loop. @EG4TechSolutionsTeam can you hear me? There is a difference between the batteries sending a SCRAM and not being able to talk to them. On the former you shut down, on the latter you pop a f*cking warning lamp, ring a bell, whatever. If I unplug either comm cable from either of my two racks, my EG4 inverters will drop to standby in a few seconds. *THIS* is the only reason I would consider just running on voltage settings. As long as the voltage is within a specified parameter the inverter should continue to run with or without comm. Comm should be an enhancement to running on voltage, not a replacement. This is the stupidest most annoying thing that still bugs the crap out of me. The inverter should run as it would with an open loop if it loses comm in closed loop. Ridiculous crap.
</rant>

 

[timselectric]

I'm the idiot with training wheels. Been called worse

As with everything it is situational. The comms give me good data in Solar-assistant. The Pytes battery SOC count seems really accurate compared to most I have seen. As others stated above, the SOC based TOU settings are nice to have and intuitive.

With comms you can see individual cells in the pack, so instead of things going wrong, and relying on some kind of safety device to save the day, you can instead see it coming.

Plenty of advantages to communications for those that have a use for them. If you don't use them, great. I like being able to stick on percentage on something. Simplifies the thought process.

I use Solar Assistant. (Wouldn't want to be without it)
That has nothing to do with closed loop communication between the inverter and batteries. (Would never do that again)

 

[LydMekk]

BMS comms are for people who does not understand the settings.
The only thing it does is fill in the voltage fields in the inverters settings automatically.
This forum would be 90% smaller if all ran voltage control and not BMS cables.
Most of the posts are problems with BMS comms.
Everything from bad settings, bad BMS firmwares, incorrect cable pinnings...

There are SO much wrong with each manufacturers inverter and battery BMS setups due to no established proper standards.

I am running voltage and have been for 3 years.
My inverters are showing SOC close to 3-5% of reality which I am very satisfied with.

I prefer to be in control myself.

 

[empowered]

I'm the idiot with training wheels. Been called worse

As with everything it is situational. The comms give me good data in Solar-assistant. The Pytes battery SOC count seems really accurate compared to most I have seen. As others stated above, the SOC based TOU settings are nice to have and intuitive.

With comms you can see individual cells in the pack, so instead of things going wrong, and relying on some kind of safety device to save the day, you can instead see it coming.

Plenty of advantages to communications for those that have a use for them. If you don't use them, great. I like being able to stick on percentage on something. Simplifies the thought process.

DATA, i wouldnt go as far as calling it GOOD data. my SOC readings when i was in closed loop were ALL OVER the place, had nothing in common with what my bms readings were. garbage.