[RE-wrenches] AC Coupling

Mon Oct 30 08:37:54 PDT 2023

Jason:

Thanks for contributing to this thread.  I did not realize it would create
so much interest.  My main point is to establish a definition for
AC-coupling, and I believe we do have consensus on that.  The next question
to ask is AC-coupling a good thing?

You make the point that BMS systems protect lithium batteries from damage.
They do so by shutting down, not correcting the problem.  But what good is
a system that shuts down?  When that happens someone has to roll a truck to
reset the system.  We should not design outages into our systems.

I spoke at length this weekend with a very knowledgeable member of this
forum (thank you, sir) about closed loop.  Their experience is that there
were packet losses in the communications and when that occurred the system
shut down.  I am going to set up my first closed loop today and I hope I do
not have the same experience.

Traditional charge controllers may seem “crude” but they operate on real,
measured values of voltage and current, values that are at the core of
battery care.  Modern charging systems seem to rely at least partially on
sate of charge.  State of charge is a calculated value that is notorious
for being inaccurate and drifting more so over time.  Every SOC scheme I
know needs to be recalibrated frequently.  The recalibration requires
hitting certain charge parameters and if you can’t hit those you don’t
recalibrate.

I assumed that a BMS system provided within a lithium battery assembly by
the manufacturer would be accurate.  I spoke with a tech support person
from a prominent LFP manufacturer at length recently and was told their BMS
is not foolproof on calculating SOC.  I am not confident that any of them
are.

I am new to lithium battery systems.  It may be my inexperience talking,
but it seems they incorporate a lot of complexity that may make them
unreliable unless charged and discharged very carefully.  I do not believe
AC-coupled charging is a good way to charge lithium batteries, or any
battery.

William Miller

Miller Solar

17395 Oak Road, Atascadero, CA 93422

805-438-5600

www.millersolar.com

CA Lic. 773985

*From:* RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org] *On
Behalf Of *Jason Szumlanski via RE-wrenches
*Sent:* Monday, October 30, 2023 7:33 AM
*To:* RE-wrenches
*Cc:* Jason Szumlanski
*Subject:* Re: [RE-wrenches] AC Coupling

I read through this thread, and I don't think one important aspect was
covered with respect to "overcharging" batteries. There may be risk with
lead acid batteries connected to a hybrid inverter that is AC coupled. But
LiPO batteries have a BMS that is there, in part, to protect the battery
from an overcharge scenario. The BMS is going to disconnect the battery
before catastrophe strikes. In my opinion, AC Coupling is better suited to
modern equipment that talks to each other in a closed loop. Crude
algorithms for throttling PV don't seem particularly confidence-inspiring.
This is one argument for sticking with a single manufacturer system
architecture (i.e. Enphase). While there are obvious downsides and
limitations to that, the system should work safely and flawlessly in an AC
Coupled scenario.

With respect to mixed systems with DC and AC Coupling, while I see the
advantages, it really complicates the design and installation. It also
greatly confuses the monitoring aspect for PV production. I think it
depends on the scenario, but going one way or the other makes the most
sense to me in most situations. With that said, there are exceptions. For
example, I have a Sol-Ark 15K client right now where all of the MPPT inputs
are taken, and reconfiguring the DC coupled strings is not easily achieved.
So, to add more PV, the obvious choice is to AC couple. But this system has
a generator, so the AC Coupled PV needs to be on the AC load output,
meaning it will not be monitored by the Sol-Ark. That's not ideal, but it's
nice to have that flexibility.

Jason Szumlanski

Principal Solar Designer | Florida Solar Design Group
NABCEP Certified Solar Professional (PVIP)
Florida State Certified Solar Contractor CVC56956

On Sun, Oct 29, 2023 at 4:19 PM Kienan Maxfield via RE-wrenches <
re-wrenches at lists.re-wrenches.org> wrote:

William,

My whole email is in regards to the point you made when you said

"This is not how battery inverters were originally designed to operate.
These systems require careful consideration to avoid battery overcharging."

I both agree and disagree with these statements... depending on what you
mean. I don't think it's super helpful to say that careful considerations
are required without mentioning what those considerations are, and I'll get
to those below.

When you say that the battery inverters "weren't *originally* designed for
this" I'd agree to an extent.. The "original" battery based inverters 25
years ago were certainly not designed for AC coupling. They weren't
"originally" designed for it. But that was then and this is now.

Now William is correct that the Gen input becomes an AC output, and you
can't wire the gen into that "input" if you are using it as a PV input...
that being said, Jay didn't think otherwise, he was just critiquing the
wiring diagram in the original post to say that it wires up differently, so
I'd say that you are both right on this one. That said, I don't think
anyone says you have to wire it into the gen port, I think that even with
the hybrid inverters, the way in William's original post is still a fully
acceptable way to do it.

I am not a huge fan of AC coupling, however, I have done it a fair amount
and I have a fair amount of experience with it off-grid with Lithium and
with Lead Acid. Most of my experience is with Outback and Victron, but I
have some experience with Schneider as well. These systems have been
working really well for a good while, but that being said, Schneider had
some real problems at first (major headache).

Now when it comes to hybrid inverters, I'm still not a huge fan... but I'm
starting to try them out. I have not tested AC coupling with any of them
yet. I was at a conference less than a year ago and one of the major brands
was saying that there is no hybrid inverter that supports both AC coupled
PV and a generator at the same time for a grid tied system. They said "you
can't have it all." There may be an exception to that, I've never looked
into it, but with the Victron and Outback (and probably Scheider... IDK)
it's easy to have it all. I've done it and it works well in every mode. The
frequency changes super fast and assuming you have a properly programmed
rule 21 compliant Grid tied inverter, it responds very quickly, and in the
systems I've monitored, the frequency never had to go above 62 Hz, which is
good enough for most sensitive loads. The inverters are totally designed
for this these days, but if you have highly sensitive loads, then it may
not work well.

*Considerations...*

First of all, I will be assuming that you are using quality rule 21
grid-direct inverters and a good battery inverter that works as well as the
Outback Radian or the Victron. The early Schneiders were too slow in
changing their frequency, and special considerations had to be accounted
for because of that, but I heard that was fixed a while ago. You had to
reduce the charging voltages to account for the delayed throttling, but I
can attest that even in the early days, Outback never had this problem, and
Victron works smoothly as well.

The primary problem for AC coupling is in off-grid scenarios or prolonged
power outages, and that's the black start issue (or dark start). Simply
put, if your battery gets low and the battery inverter turns off, then your
PV can't charge the battery. This problem is lessened when you connect the
PV Grid Direct inverters to a dedicated output on the inverter so that the
loads shut down while the battery is still a few percent above the inverter
shutdown level. That is a nice advantage of the Victron inverters. I don't
know if the Hybrid inverters do this or not. In any case, even with this
little safety net, the inverter can still discharge the battery to the
point where the inverter turns off and you have a problem. My solution is
that you should have at least a little bit of the PV DC coupled so that
when the sun comes out, the voltage will rise and the inverter will turn
on. This is also recommended in Victron's AC Coupling manual.

The second consideration is the minimum inverter size... Your battery based
inverter needs to be capable of handling and controlling the full PV power.
Victron recommends that the maximum PV power (DC or AC, whichever is lower)
does not exceed 100% of the rating of the inverter (no more than 10kW of
power from the PV on a 10kW inverter). Outback recommends no more than 6 kW
of Grid-Tied inverter per 8 kW of Radian (so that's 75% of the battery
inverter's power). In some circumstances, this could demand upsizing the
battery inverter, which is another great reason to DC couple half of the PV
and AC couple the other half... because then you could effectively have
more PV with less battery based inverter, depending on your loads.

As was mentioned by William, monitoring is generally not as good with an AC
coupled system, but here again, that's not always true. Victron has been
really leading the way with integrating with other PV grid-tied inverters
so that you can monitor them through Victron's own monitoring. This depends
on the brand you are connecting to, but most of these brands have a way to
get this information via TCP/IP, and if so, Victron will read that and
report it on their own website, in their app, and on their touchscreen.
With some brands, you can even see the error codes etc. remotely through
Victron's monitoring. It almost makes it seamless as though it was all one
brand.

So in case it's not clear, I'm never a fan of 100% AC coupled PV, but these
days there are very few legitimate concerns or issues... Most if not all of
which disappear with a 10% - 60% DC coupled system.

Thanks,

Kienan

*Green-Go Solar Wholesale Distribution*

*maxfieldsolar at hotmail.com <maxfieldsolar at hotmail.com>*

*(801) 631-5584 (Cell)*

*www.distribution.solar*
------------------------------

*From:* RE-wrenches <re-wrenches-bounces at lists.re-wrenches.org> on behalf
of Jerry Shafer via RE-wrenches <re-wrenches at lists.re-wrenches.org>
*Sent:* Saturday, October 28, 2023 11:14 PM
*To:* William Miller <william at millersolar.com>; RE-wrenches <
re-wrenches at lists.re-wrenches.org>
*Cc:* Jerry Shafer <jerrysgarage01 at gmail.com>
*Subject:* Re: [RE-wrenches] AC Coupling

AC coupled systems used to work like this but with solark and sort of with
outback they use the gen set input to the inverter, do the freq shift and
charge via a preset voltage. This seams to work well but if you connect a
genny it must be on the grid input via an ATS, alot more complicated the
any DC side connection.

Funtimes

On Sat, Oct 28, 2023, 8:51 PM William Miller via RE-wrenches <
re-wrenches at lists.re-wrenches.org> wrote:

Jay:

I am not sure how it would be possible for a grid-tied inverter to operate
with any battery inverter if it was not connected to the battery inverter
AC output.  The grid-tied inverter needs to see correct voltage and
frequency before it can start up.  How else could this work?

I looked up all of the inverters you cited.  In each case it may look to
you like you are connecting to an input, it is really just reprogramming
the inverter to turn the generator input into an extra output.  All of the
manufacturers provide warnings about doing this.  Here is more information
from each of the manufacturers:

Sol-arc:  From the manual for the Sol-arc 15K-2P-N: You convert the Gen
input to an output and make that your connection point.  “*A full AC
coupled solar system is not recommended as power control and monitoring is
limited.”*

EG4:  I could not find an “ev4” inverter anywhere online.  There is an EG4
inverter.  I read the manual and it looks like the EG4 works just like the
Sol-arc:  You program the generator input to actually be an output.  The
EG4 abruptly disconnects that output when battery SOC reaches a programmed
level-- There is apparently no frequency shift function.  The manual is
pretty inadequate on this point but they do provide a video
<https://www.youtube.com/watch?v=vRAq9g34VcI&ab_channel=EG4Electronics>.

Fortress:  The same as above is true about using the generator input for
AC-coupling on the Fortress FP-Envy-12k.  From that manual*:  It is
forbidden to connect the Generator in Gen Port when AC Coupled.  *That is
because when AC-coupling, the generator input is no longer an input.

So while it appears one connects an AC-Coupled inverter to battery inverter
inputs, it is just a reprogramming of the generator input to be an output.
In all cases that I can see there is still no voltage control of AC-coupled
battery charging.

I am aware that to try and provide some modicum of control over the battery
charging from AC-coupled sources, a scheme was invented to shift the output
frequency and taper or shut down AC-coupled power.  This is a work-around,
not a design.  In each and every one of these systems it is pretty clear
there is little control over battery charging voltage.  These systems do
not support three stage battery charging like a DC charge controller can.

When AC-coupled charging is connected to batteries with BMS in certain
conditions battery voltage spikes can occur.  See section 2.2 of this
article. <https://www.victronenergy.com/live/ac_coupling:start>

There are limited advantages to AC coupling:

·        AC coupling allows reduced wire gauge for long-haul feeders.  This
is no longer as necessary with new 600 volt input charge controllers.

·        AC Coupling allows an easier upgrade path for a customer that has
grid-tied inverters and wants to add batteries.  I am not sure most
batteries can survive for long with the crude charging algorithm provided,
so I would hesitate to recommend this.

In my opinion AC-coupled battery charging with even the newest inverters is
a crude affair.  I am not a fan.

William

Miller Solar

17395 Oak Road, Atascadero, CA 93422

805-438-5600

www.millersolar.com

CA Lic. 773985

*From:* Jay [mailto:jay.peltz at gmail.com]
*Sent:* Saturday, October 28, 2023 1:58 PM
*To:* william at millersolar.com; RE-wrenches
*Subject:* Re: [RE-wrenches] AC Coupling

I’m going to both agree and disagree

Agree The older inverters ( Schneider, Outback, etc) your drawing is
correct.

Disagree For many new inverters ( sol ark, ev4, fortress) they bring ac
coupling into the inverter directly.

Usually on the generator input and not in the backup loads panel.

And I will also disagree about the statement that these inverters are not
designed for ac coupling.

Agreed that the earlier non rule 21 compliant versions didn’t have any
internal control and required external relay control to prevent overcharge.

Newer versions of Schneider, OB, have  frequently shift which works with
newer gt inverters to address the ac coupling overcharge concerns. While
creating potential new issues for loads with up to 64 hz.

Jay

On Oct 28, 2023, at 2:15 PM, William Miller via RE-wrenches <
re-wrenches at lists.re-wrenches.org> wrote:

Friends:

I am starting to get the impression that there is a diversion on the
understanding of the definition of AC coupling within the industry.  I
think this definition needs to be clarified.

Below is a diagram of what I understand is a DC coupled system (I hope
these diagrams come through):

<image003.png>

A battery inverter that receives AC power into an AC input port is *not* AC
coupled.

Here is an AC-coupled system:

<image005.png>

What makes AC-Coupled systems unique are the problems presented by sending
power *backwards* through a battery inverter.  This is not how battery
inverters were originally designed to operate.  These systems require
careful consideration to avoid battery overcharging.

Do we have consensus on this definition?

William Miller

Miller Solar

17395 Oak Road, Atascadero, CA 93422

805-438-5600

www.millersolar.com

CA Lic. 773985

_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the
other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org

_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the
other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org

_______________________________________________
List sponsored by Redwood Alliance

Pay optional member dues here: http://re-wrenches.org

List Address: RE-wrenches at lists.re-wrenches.org

Change listserver email address & settings:
http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org

There are two list archives for searching. When one doesn't work, try the
other:
https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org

List rules & etiquette:
http://www.re-wrenches.org/etiquette.htm

Check out or update participant bios:
http://www.members.re-wrenches.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20231030/f5ef78ac/attachment-0001.htm>