Build 55 Victron Feed-in questions

[Daniel8181]

Already testing the Release 2.17.55 intensively.

Thanks Batrium-Team @James to push this forward for Victron Users, much Appreciated!

Open for Conversation.

Issues I found out yet.

The Documentation does not tell anything about the activation of ESS Grid Feed in (AC or DC coupled). Assumed it’s not necessary, but would not make sense to me, watched what happens at Batt Full, no feed in happened, checked MPPT`s - they went into State 1 ( limited power ) once batt was fully charged.
Batrium Integration showed in Victron GX Full Power A (not LIMP or else), so Amps might not limiting factor here…

Activated AC Feed in in ESS had no effect (to me)
Activated DC had effect, that Batterylimits from Batrium were ignored by DVCC, Battery went into Charge - Blockmons started burning amps (Batrium was in no recharge allowed), quickly switched AC-Feed in off again…

In Victron Documentation i read, that DVCC adds “invisible” 0,4 Volts to the BUS in order to enable Grid feed in, when you activate it in ESS)
See Chptr. 11.6 in 11. DVCC - Distributed Voltage and Current Control

What makes it even worse is when you limit the grid feed in (because your local Provider told you so…) the excess Power of these 0,4V the MPPT`s push stays “in” your DC net, trying to go into the Battery.

Confused at the Moment, how the whole integration is intended to be configured and work, do I need to decrease the actual values in Batrium about these “invisible” 0,4 Volts, which DVCC&ESS-DC-Feed-In adds, ar does this have negative other side effects I can’t yet imagine?

[James]

Hi Daniel,

Hope you don’t mind that I’ve split your topic into a new thread.

My understanding is that feed-in settings shouldn’t affect us, as he have a few Victron users now reporting that grid feed works well.

Our offset voltages are actually based on Victron’s hidden 0.4V, we found that document too. The difference is that we change that offset to 0 when we are done charging with Dynamic Volt Targets. We changed this to 0.3 when we found it was a bit too agressive for small systems.

Privately, I’ve asked for you to send through your WatchMon serial number and a read-only VRM share link. With this in hand we can make a more informed call, and update this thread.

Cheers,
James

[Daniel8181]

Hi James, nice to meet you.

I`ll send you an separate eMail across with some details.

Hopefully I got your post right, but maybe lets discuss further after you have a better picture.
Sorry non native English speaker - btw. greetings from Germany

Do you know if these other Victron-Users have an ESS-Assistant Installed?
Essentially “Grid Feed-IN” whether it is AC or DC is a switch within this ESS-SUB-Menu.
Did they keep those switches “OFF”?
Because I observed no Feed in while trying Dyn-Volt-Ramped Feature, while Batrium was telling the DVCC give 240A on the BUS (which was my setting).

– ignore #Low SOC, I fixed that after the Screenshot…

does not mean I didn`t do something wrong somewhere, eventually.

At the moment the Sun disappeared in Germany maybe some days and I switched back to my old setup using Ramped Current Targets. I wrote a Node-Red Flow months ago which shifts the Grid-Set-Point from -50W to -5KW dynamically once the BATT finished balancing, by monitoring if all MPPTs are limited or in Full throttle-Mode and if Batt current goes higher then -10A it reduces Gridsetpoint step-wise.

Enjoy your night in Australia going to bed in Germany now.

Looking forward.

Cheers, Daniel

[kleini]

Hi Daniel,

I have a Victron Energy ESS system with AC coupled PV and DC coupled PV. And I am running version 2.17.55 of Batrium. Grid feed in works perfectly from AC and DC coupled PV in my system.

Of course, you need to activate “feed in excess” in the Grid feed-in menu in ESS settings for AC and DC according to your system setup. I read, you manipulated your feed-in previously by configuring a high negative grid setpoint. I think, this is very bad practice, but unfortunately adviced by some very unexperienced tinkerers in German forums. Don’t do it that way. Please ignore my configured feed-in limit. This is necessary in my system as I configured my Fronius Symo for “Zero and Limited Feed-In” and then the Victron system then needs to tell the Fronius through sunspec protocol how much energy to produce at maximum.

Keep in mind, that charge current limits do not apply in the Victron system anymore, if you activate grid feed-in from DC as solar chargers then need to provide their full current. You’re right with the 0.4V charge voltage increase to get the grid feed-in from DC started. In my system these higher voltages only appear shortly, when grid feed-in from DC ramps up. Once this is stable running, charge voltage goes back down to charge voltage limit reported by Batrium.

Kind regards,
Marcus

[Daniel8181]

Hi @kleini Marcus,

thanks for your reply to my Post and taking some time for it, appreciate.

Actually i never felt this Grid-Set-Point-Shifting-Workaround as a ideal solution, I never was satisfied when looking at it, but lets say, it did so far 80% of what i was striving for of course it was somehow a semi-beautiful solution.

• Prio 1 : Charge the Battery and balance it softly by the help of ramped CC targets…
• come into a limited Grid Feed-in (because I may not pump everything into the Grid - as advised by my local provider)
• keep the battery more or less charged (not charging) and avoid getting rebalanced twice a day by having CV13 set to a good value.

I managed to do this in my Node Red Flow, whiles shifting up or down the Grid-Set-Point every 10 Seconds among Shunt being not more then 10A negative and all MPPT`s running in state “2” not limited.

My actual Problem was (and probably still ist) the 0,4V Voltage-add by the DC-Feed-in Feature.
It actually always when I tried looked scary…
Battery at 100% sitting in peace.
Switching on DC Feed-in, boom all Balancers went into “panic” turned it off.
Switched off Feed-in Limiter, was a bit better, but then my PAVE Sensor dropped my Grid-Coupler. (Of course, peak feed-in… forgot), then bms pulled the rope and kicked by SACE-TMAX Breaker in fron of my battery

Questions @James maybe aswell.
To which Voltage Setting is Victron Adding this 0,4V?, which setting is sent by Batrium to victron, where these 0,4v get added by the MPPTs when full? Is it the High Voltage from Batrium (which is sent to GX), or the Low Voltage? when batt is full Or is it the Volts configured in MPPTs? (After Batteries are full? and the Batrium Values are ignored after full? That is what confuses me. When which value is applied.

@kleini In your System:

• If you decrease the Grid Feed-in further, do you see MPPTs getting reduced or Limited in your System? I mean, shouldnt at least the Victron try to reduce the Amps on MPPT`s side in such case eventually?
• Would you be so kind and provide me some screenshots of your values under.
  Charging + Extra Params and Remote, I would be curious about them, to really compare and try to have a 1:1 test how my system behaves when I retry on the weekend?

By the way, which Cellmons do you have?

Cheers, Daniel

[kleini]

Dynamic voltage targets work perfectly to charge your battery softly up to 99%, do the balancing and then reduce the charge voltage target to float voltage. This works absolutely perfect. I don’t remember the exact condition to start a normal charge cycle after float voltage is reached but in my system recharging only happened the next day.

If you need to limit your grid feed-in then configure it this way as seen my screenshots in my previous post. I don’t need anything in Node-RED to fix mixbehavior somewhere in my Victron system. Version 2.17.55 perfectly fulfills all needs of an ESS system.

0.4V are added to the charge voltage limit reported by Batrium to the Victron system. The graphs allow showing something named effective charge voltage, which will show you those additional 0.4V. These additional 0.4V added to either dynamic max voltage or dynamic min voltage should not result in Batrium going into critical state. Then your voltages are not configured correctly. And I already reduced my voltages a lot compared to Batrium’s defaults. Please see my graphs above.

You’re using an ABB SACE-TMAX. This is a bad device as it can only break the connection and it is not able to reconnect your battery. I can only advise to switch to Kilovac and utilize a Pre-charge device or implement pre-charge with Batrium.

What do you mean by decrease grid feed-in? My grid setpoint is configured to 0W. I never tried to limit my grid feed-in. But if I would do so, I am sure, voltages might rise those 0.4V and then solar chargers will not go higher with their voltage and currents will go down accordingly. You need to keep in mind, that Victron systems are controlled by voltages. Solar chargers and MultiPlus and so on always generate some voltage and with the resistance comes the current: I = U/R. Regarding the battery the voltage is the voltage difference between cell voltage and charge voltage resulting in the current through the battery.

Will try to provide you the screenshots of the voltages, but normally I don’t use Windows systems, so running WatchMonToolkit is hard for me. I use CellMate-K9.

Kind regards,
Marcus

[kleini]

Okay, battery is now full on my system for today and I limited my grid feed-in in the ESS submenu in several steps.

For AC coupled PV you need to configure feed-in limiting as I already mentioned above. Otherwise the Victron system is not able to reduce the power produced by AC coupled PV. Frequency remains to the one of the grid and that frequency is within the producing limits of the inverter. Reducing power through frequency shift is with grid connection not possible. For Fronius AC inverters this feed-in limiting is described here: 4. Configuration
I don’t know whether this can be configured with inverters from other manufacturers.

For DC coupled PV, the additional 0.4V are now completely applied as the MultiPlus do not invert the power from solar chargers. Once this voltage is reached by the battery, the solar charger completely stops producing power. My RS450/200 went down to -16W.

So, your battery needs to be able to “consume” these additional 0.4V to the charge voltage and AC coupled PV needs to be configured for limiting grid feed-in.

Kind regards,
Marcus

[Daniel8181]

Dynamic voltage targets work perfectly to charge your battery softly up to 99%, do the balancing and then reduce the charge voltage target to float voltage. This works absolutely perfect. I don’t remember the exact condition to start a normal charge cycle after float voltage is reached but in my system recharging only happened the next day.

→ Thanks this is what I had in Mind now. I will reduce my optimum Charge Voltage, which is 55,5 to 55,1 for my next try in the New Dynamic Volt target feature.

you need to limit your grid feed-in then configure it this way as seen my screenshots in my previous post. I don’t need anything in Node-RED to fix mixbehavior somewhere in my Victron system. Version 2.17.55 perfectly fulfills all needs of an ESS system.

→ That was a misunderstanding. This is where I limited feed in swell (and switched on dc feed in (do not have AC) when trying to get it done without messing with the grid set point. But as said I had some trouble with my batteries, lowering my volts as suggested under quote #1

0.4V are added to the charge voltage limit reported by Batrium to the Victron system. The graphs allow showing something named effective charge voltage, which will show you those additional 0.4V. These additional 0.4V added to either dynamic max voltage or dynamic min voltage should not result in Batrium going into critical state. Then your voltages are not configured correctly. And I already reduced my voltages a lot compared to Batrium’s defaults. Please see my graphs above.

→ Will look for that value, that is a supercool hint, thanks and indeed, as you suggested under Quote #1 will try that.

You’re using an ABB SACE-TMAX. This is a bad device as it can only break the connection and it is not able to reconnect your battery. I can only advise to switch to Kilovac and utilize a Pre-charge device or implement pre-charge with Batrium.

→ Yes, Next Evolution-Step in the System Had this one from the Datacenter as a leftover when designing my system, you never stop learning. Thanks for the hint, maybe I come back to you for some advise how you integrated it.

What do you mean by decrease grid feed-in? My grid setpoint is configured to 0W. I never tried to limit my grid feed-in. But if I would do so, I am sure, voltages might rise those 0.4V and then solar chargers will not go higher with their voltage and currents will go down accordingly. You need to keep in mind, that Victron systems are controlled by voltages. Solar chargers and MultiPlus and so on always generate some voltage and with the resistance comes the current: I = U/R. Regarding the battery the voltage is the voltage difference between cell voltage and charge voltage resulting in the current through the battery.

→ Will observe this one closely when I reconfigure, thanks!
→ But actually you do limit it ( by this Max Feed in 14kW Value)

→ Curious… What do you use instead of WMToolKit? (Sorry for Noobing)

Super grateful for your Help Marcus !

[kleini]

The peak power of my AC and DC coupled PV is 12.4kW. Therefore, 14kW maximum feed-in limit is not a limit. I can never reach it, except on very cold and clear winter days. I need to set this value, otherwise the Victron system does not tell a maximum to my Fronius, which is configured for limited or zero feed-in. Maybe I should remove the dynamic power reduction softlimit. Then I would not need that maximum feed-in limit anymore.

I don’t configure anything on Batrium. It runs now very smoothly for more than a year. Therefore, I don’t need WMToolKit. BMI is currently enough for me. Maybe one day, I will let Batrium send its monitoring values to MQTT or deploy WatchmonUdpListener.

[Daniel8181]

Allright, I think this works now! Tested Today. Thanks @kleini

What I did to make it work in my setup was.

Batrium:

1. I lowered my DVMax to 55,0 from 55,5 (55,5 was my perfect Balancing Voltage to get the cellmons into balancing within the old non DVCC Feed-IN Setup and Ramped A-Targets). Because DVCC adds 0,4V to this Value if you enable Feed-IN). So Still with 55,4V/16=3,4625 i theoretically&practically (tested) reach a balancing which starts at 3,45V CellV.
   → So, you need to Lower this Value at least by 0,4V and tweak a bit…

2. Then I figured out, that even when the Battery is signaling full to Victron I come back into balancing when enough solar power is there - but shouldnt as per Batrium, because we had that Battery Full, so pushing it into balancing all the time wont make sense. CV13 and QC5 were not yet even scratched when I observed the behavior. Did I miss something in here to tweak?
   

I figured out, if i put this Lim Value at the Remote-Page into a Float-Voltage Level (54,5 + 0,4 (Which Victron adds resulting in 54,9V)) would do the Job, and as I can say, it does.
→ Not Sure, if I worked somenting around, which could had ben done easier (or intended), maybe advise?! At Least it seems Batrium isn’t adding the 0,1V from Limited, when Battery is full. So it was plausible for me, that this could be the Float, when the Dynamic Volt offset isn`t added, which wasn’t as I observed.

Victron DVCC:

Also “Maximale Einspeisung” Max Feed-IN works now flawlessly, I was able to observe DVCC limiting the MPPTs accordingly and I dont have overshooting Blockmons when Feed-in gets limited while Sun is pumping.

So at the End, lowering and tweaking the Batrium Setting for DVMax in Normal (55,0V) and Limited (54,5) did the trick, for Charging, Lim-Charging and Floating.
You can see in the Victron Diagram, that after 100% (Full) the battery floats at 54,5V(Lim Value without offset 0,1V)+0,4V DVCC-Offset at 54,9V

Resumee: @James
If my Setup is right, I assume having a bit more explanation of the Values and what effect they have would be super helpful for other Users after me.

Comments, corrections and questions welcome

[kleini]

Something is not correct in your Batrium setup. If Batrium reports 100% SoC then CVL should be sent to Victron from Dynamic Volt Min. According to your VRM graphs it sents Dynamic Volt Max Limited. So, somehow the detection of full battery does not work correctly in your setup. Maybe your charging settings are not correct, especially the bypass energies.

Please see graphs from my system above. Once SoC reaches 100%, CVL softly goes down to Dynamic Volt Min and CCL should go back to normal value.

[Daniel8181]

Hmm, you mean the Value DynVoltMin is the “Float Value” and should be applied after Batt-Full?

Do you see something suspicious ?

[kleini]

Dynamic Volt Min is the float voltage.

I am not familiar with BlockMons. Description for Charging is here: Charging Settings | Batrium Knowledge / Wiki
You need to check now, whether all cells reach either Bypass Amps Hrs or Final Bypass Amps. I think, this is the case. But your cell voltages are not below Cell Hi Resume. They are above. That seems to be the cause for not reaching Charging FULL state. And you Hi Cutout also seems to be very low. Cells need to be able to do some balancing. Some will be above 3.45V, some below. If you stop charging then, the cells below 3.45V can not reach that voltage.

[Daniel8181]

Actually they reach charge Full State, Batrium reports this.

• I thought the Bypass Session (yellow) also reports this, see my Screenshot
• Also the blue Icon in Chart reports this. (Bypass Session)
  
  

So, I really get all cells into balancing and Batrium also puts that blue icon in the below icon-bar, but if i do not put the limited Voltage in a Float “Area” Victron won’t let them float…
In my test, the only way to avoid “recharging” was, to put the Dynamic Volt Max at Limited-Value to a “Float Value”. which was this 54,5(+0,4V)(no offset seemed to be added by Batrium)

Unfortunately when I took the picture it had negative current, but before it had positive (Full-Batt and all cells had balancing in a cycle before), you can see some mons are balancing still.

[Daniel8181]

Maybe I never reached CV14… will observe that tomorrow.
But what makes me wonder is, the fact that Batrium states FULL.

[Daniel8181]

Test done.

What I can say is.

When Batrium Reports Full by following icons…

…Also Checked, bypass is completed…

…also all Cellmons did say “Hello, let’s bypass” …

—> The Target Voltage seems to get set to this value (Dyn.VoltMax-LIMITED), no Batrium offsets or so applied, (sure Victron does, but that is clear) when Battery is full.
This also correlated with what @James said in the first reply to my first thread post…

Our offset voltages are actually based on Victron’s hidden 0.4V, we found that document too. The difference is that we change that offset to 0 when we are done charging with Dynamic Volt Targets. We changed this to 0.3 when we found it was a bit too agressive for small systems.

…
→ So this value needs to be selected high enough to sustain at round about 100% SoC, but low enough, to not stay in balancing all the time.

So still assuming, this Value…

… is somewhat also the “Float Voltage” in parallel to a “limited charge Voltage” during the Cycle in Shot below.
→ Is there a Line in the graph missing (Dynamic Volt Max “in Limited” / Float), see my green Arrow?!
→ there seems to be some text in the Wiki missing, see my red Arrows?! CopyPaste?!

@kleini @James any thoughts?

Cheers, Daniel

[James]

@Daniel8181

→ there seems to be some text in the Wiki missing, see my red Arrows?! CopyPaste?!

Hi Daniel,

You have indeed found an error in our wiki. I’ve corrected it.

I really don’t think you should have to resort to different limited min/max, they actually weren’t really intended to be used like that. I will inspect your system graphs today and see if there is anything else we can tweak or correct.

[kleini]

Documentation for dynamic voltage targets and also the YT video state, that in Charge Off or Charge Full state, the CVL should be reported as the current shunt voltage and not charging should happen. So after some time, shunt voltage should go down to Dynamic Voltage Min. And this does happen in my system. Therefore, I still think, the float voltage is Dynamic Voltage Min and not Dynamic Volt Max Limited.