question

Javier San Juan Costillas avatar image
Javier San Juan Costillas asked

Quattro 15000VA and Pylontech US3000 high voltage alarms


I had this setup running flawlessly without alarms (except due to imbalance at start) for almost 1 year and in the last month I started to get High Voltage alarms. I have been in touch with Pylontech and tell me that is victron causing it due to high charge voltage... But I have all the settings lowered twice in the last week and I still get this issue.

Setup:

  • 2 x Fronius 6Kw
  • 1 x Victron Quattro 15000VA
  • 6 x Pylontech US3000

Looking at battery voltage-current graph, on the left we can see how in the night the battery gets charged (with the grid) and regulates very well and keeps the max voltage on batteries around 51,52V quite stable.

On the right we can see how the battery its charged during the day, with excess solar but firstly it does not charge as smoothly as with grid (even it is charging at less amps) then once charged tthe battery voltage keeps shooting up until sometimes reaches even 54V.

What is causing victron to work differently during the night and day?

1629817725827.png


Here are the limits configured in VEconfigure and even limited on DVCC too. As pylontech support recommended.

1629818054197.png1629818070719.png


Should I even lower more the maximum charge voltage? I even limited the charge current to 2A during a few days and still got overvoltage alarms.

I have checked and all cells are balanced thus no idea what else might be going on.

Looking forward to some experienced Victron member to illustrate me why during the night and day is different.


MultiPlus Quattro Inverter ChargerESSPylontech
1629817725827.png (81.6 KiB)
1629818054197.png (222.8 KiB)
1629818070719.png (107.9 KiB)
5 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·

Here there is a finer look at the moment when the alarm comes. Pylontech support team says is due to Victron putting higher voltage than requested.

1629891813844.png


I am literally going crazy to find this issue....

0 Likes 0 ·
1629891813844.png (262.8 KiB)
Nico Bayer avatar image Nico Bayer Javier San Juan Costillas commented ·

This looks like the system is not following the max charge current set by the battery, but using its own charge algorithm or something.


My system ( 8xUS3000 + 3x Multi2 5000) follows the max charge current set by the battery until its full and then drops to 0.

Is your ESS assistant set up correctly for the Pylontech Battery? https://www.victronenergy.com/live/battery_compatibility:pylontech_phantom

0 Likes 0 ·
Javier San Juan Costillas avatar image Javier San Juan Costillas Nico Bayer commented ·

Do you have feed in active? I thank that might be might problem. Looking at past graphs voltage never shoot up and pylontech support team says is due to high voltage charge from Quattro.

When I had feed in deactivated didn´t got an alarm... I did activate feed-in around 12May or so when I was legally allowed to. When I charge at night from grid, it works flawlesly too, I wonder what the issue is...

1629905464077.png


0 Likes 0 ·
1629905464077.png (204.8 KiB)
Nico Bayer avatar image Nico Bayer Javier San Juan Costillas commented ·

Yes i have feed in active. My battery voltage never goes above 52.5V
I have 15kw AC and 10kw DC coupled PV, all feeding power into the grid.

0 Likes 0 ·
Javier San Juan Costillas avatar image Javier San Juan Costillas Nico Bayer commented ·

Thats odd its properly configured and literally today shows a perfect example of why the Victron its doing its own thing... When DVCC says CCL 0A it has pushed into the battery and goes higher.

Either voltage or current... but something is wrong.

I wonder if @Guy Stewart (Victron Community Manager) or @mvader (Victron Energy) could give me a hand... of support.

1630072820213.png

0 Likes 0 ·
1630072820213.png (254.9 KiB)
10 Answers
Jack Peters avatar image
Jack Peters answered ·

Might be worth giving this a read: https://www.victronenergy.com/live/ac_coupling:start#minimum_battery_capacity

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·
We barely use 10000VA was way oversized by distributor. But having current limits and feed in activated should not be a problem at all, at the end of the day BMS sets its limits via DCVV.
0 Likes 0 ·
Nico Bayer avatar image
Nico Bayer answered ·

Are you feeding excess power back into the grid or is your system set up with zero feed in?

Is the communication between your GX and the battery working correctly? What are the values in the Pylontec parameters tab when the battery is fully charged (CVL CCL DCL)?

Do you have any other charger on your DC system? An MPPT charger?

Without any current into the battery the voltage should not rise. Can you measure the battery current with a current clamp and check if it is the same as shown by the Quattro?

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·

No other chargers only Quattro 15000VA

I have it set to feed in excess, no-limit. Saying this... the issue has been happening since the feed in excess has activated in the last month or so.

Communication seems fine it reduces max current while it starts to get full in the DCVV menu. See screenshot. Perhaps not to 0 CCL because last few days at 51.5V was not reaching full SOC I think.

1629872605844.png


0 Likes 0 ·
1629872605844.png (283.8 KiB)
Rob Fijn avatar image
Rob Fijn answered ·

Hi JSJC, There is no need to set the Charge Voltage Limit in the DVCC menu, as the Pylontech BMS dictate these setting to the Victron system, you have the ESS assistant in the Quattro that takes care of this. This also externally control the MPPT ( you can verify this in the MPPT menu).

Settings would not change just after a year.. Unless the Quattro was updated and the ESS assistant was lost in this action.
Checking communications is a thing to to and verify.

Hope this helps, Regards, Rob


1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·
Thats what I thought but from Pylontech support they told me to limit it in the config and I did as well inside the DVCC too, because still reaching 54V for no reason.


Thats what I thought that after a year should not suddenly change. The only difference is that now we have an ESS with feed-in active, before feed-in was disabled.

0 Likes 0 ·
Alexandra avatar image
Alexandra answered ·

@JSJC

We have experienced something similar on a system. Voltage rise at night while on grid. Not Pylontec though.

For a while we had a work around where we let the battery drop to 95% soc each night. Then found it was sometimes 93% or other random percentages close to 95. So our that was our clue then cell balance or bms was the issue.

We have pulled the battery stack out and sent if for diagnosis. Had found cell imbalances were the cause. So the pack is in rehab.

Is your system doing an absorption phase with feed in active? The other thing wit pylon, sometimes changing which battery is the master can help with the errors.

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·
All pylontech battery logs have been reviewed and was pretty well balanced, so that is not the issue :S

Will try about the master but I doubt is that if has been working for amost 1 year until feed-in was activated without issues, I bet it has to do with the feed in because at night when there is no feed in its all pulling from grid it works flawlessly...

0 Likes 0 ·
Javier San Juan Costillas avatar image
Javier San Juan Costillas answered ·

@Guy Stewart (Victron Community Manager) @mvader (Victron Energy) @victronteam

Is there anybody that can help with this issue???, its still happening and it is definetly a victron issue handling the charge while Fronius its producing and there is feed into the grid happening....

This will degrade batteries overtime, I bought a Victron because was recommended as solid product, well developed, good support... But I am starting to change my mind here.

Its a big issue, not only happening to my system and needs a solution.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Daniël Boekel (Victron Energy Staff) avatar image
Daniël Boekel (Victron Energy Staff) answered ·

Hi @JSJC

It looks like the battery disconnects for some reason (you can see the current dropping to zero suddenly, after that the voltage goes up), and without the battery connected the inverter cannot keep the voltage stable.

Please read out battery logs / cell voltages, there might be an imbalance / defective module.

at the same time you can try to lower the maximum charge voltage in the DVCC menu to about 51 volt to see if this prevents the issues.

2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·

Hi Daniel,

I have reduced the maximum charge voltage to 51V previously and still happens.

It does not disconnect, it simply gets full and requires for 0A.

I have talked with Pylontech Support they got all logs from all batteries and confirm that is the Victron sending more voltage to the batteries than the one set on config. Batteries are fine no defective cells or modules.

Said this, this situation ONLY happens when its day time, batteries full and the two Fronius are producing enough to have excess fed into the grid. At night it charges perfectly and does never go overvoltage.

There is something dodgy going on. I bet if I turn of feed in it will get back to normal (as it was a month ago when was disabled)

Here there is another person having similar issues: https://community.victronenergy.com/questions/95663/pylontech-high-voltage-5396v-when-dvcc-set-to-limi.html

0 Likes 0 ·
Javier San Juan Costillas avatar image Javier San Juan Costillas commented ·
I am guessing that battery disconnections as you mention are due to high voltage and the BMS tries to get disconnected? Physically it has not been touched.
0 Likes 0 ·
Javier San Juan Costillas avatar image
Javier San Juan Costillas answered ·

@Daniël Boekel (Victron Energy Staff) @mvader (Victron Energy) @Guy Stewart (Victron Community Manager)

Hi again guys!

I come back with some fresh data to that shows what I meant and proofs what Pylontech Support Team is saying.

Since I have activated the Feed-in in the system the batteries have been giving overvoltage issues, not due to imbalance or defective modules (it worked great for almost one year). If there was issues on imbalance or defective it would have continued, but I tested one day with feed-in deactivated and then everything was balanced. So here are the screenshots of the data. There are two days that Feed-in was active and not limited. And last day that I deactivated the feed-in and then you can see how voltage during peak sun hours was regular not spiking like when feed-in is active.

This can cause serious harm to the batteries and has nothing to do with the BMS or battery system is about the Victron charger not functioning as expected.

It has nothing to do with charge voltage, I have tested 52V, 51.5V and as low as 51V and still got the High Voltage Alarms, and before (with feed-in disabled) it used to be 53.2V for almost one full year and never (besides initial balancing) had issues with High voltage alarms.

As you can see when charging at night any day the batteries respond perfectly and there are no voltage surges at all. Which is the norm if the charging is done at the correct voltages.


30th-31st October - Feed-in Active and no limited. (High Voltage Alarms)

feed-in-on.png30thoctober.png31stoctober.png

1st of September (No High Voltage Alarms)

feed-in-on.png1stseptember.png


What is the cause of this and most importantly how to get it resolved.


Thanks in advance!



30thoctober.png (320.2 KiB)
feed-in-on.png (90.3 KiB)
31stoctober.png (311.9 KiB)
feed-in-on.png (90.3 KiB)
1stseptember.png (299.0 KiB)
17 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Guy Stewart (Victron Community Manager) avatar image Guy Stewart (Victron Community Manager) ♦♦ commented ·

Hi @JSJC ,

Can you please adjust the settings to be exactly as I have shown in this answer and see if that makes any difference?

https://community.victronenergy.com/questions/95663/pylontech-high-voltage-5396v-when-dvcc-set-to-limi.html

0 Likes 0 ·
Javier San Juan Costillas avatar image Javier San Juan Costillas Guy Stewart (Victron Community Manager) ♦♦ commented ·
Hi @Guy Stewart (Victron Community Manager)


My situation it is the opposite, I want to feed-in to be active and not limited. The other day after reading that post I did try to set a limit on feed-in and keep it high to see if that will maintain things correctly but didnt work.

I don't know what calculations are done behind in the Victron systems, but with feed-in active no limit, should be "simple", no calculations whatever is left goes to the grid and not the battery. There is no need to have a buffer in case of spikes, because they can go to the grid.

So the need is to have a solar setup, with ESS and battery backup, that feeds the grid on the day with whatever is left. But of course not damaging the batteries.....

What else could be tried?

0 Likes 0 ·
Show more comments
Javier San Juan Costillas avatar image
Javier San Juan Costillas answered ·

@mvader @Guy @Daniël, is there a chance I can get somebody to reply either here or email?? Its been over a month since this critical issue has been communicated and there is no active feedback/solutions from Victron.

Really need a solution to the Quattro not being able to stop charging when its set to STOP, and as well not sending more voltage than its configured to. If feed in is activated with no limits it should feed to the grid not to the batteries full batteries.


2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Javier San Juan Costillas avatar image
Javier San Juan Costillas answered ·

Thanks @Daniël for keeping an eye on this over email.

We came to a conclusion after multiple tests, that the issue why the battery was overcharged was due to Input Current Limit being very low, in my case 12A. In this case ESS feed in will have issues feeding into the grid and overcharge battery if there is large charge going on.

I tested now leaving higher grid Input Current Limit when battery is almost 100% charged, and no issues whatsoever with the battery voltage.

Awesome, thanks again!!

PD. I will keep spreading the Victron love around :)

4 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

shaneyake avatar image shaneyake commented ·
Glad it is all sorted now!

Can you explain why you had such a low AC input limit? Being you have a 15kva inverter?
Just curious as I didn't think about that when reading your post.

0 Likes 0 ·
Javier San Juan Costillas avatar image Javier San Juan Costillas shaneyake commented ·
The reason is because during the day the input limit is very low (to pay less to grid) than during the night (that is quite higher), during the day mostly is all feed-in or solar consumption and during the evening-night its all battery/grid consumptions.

I thought was good to keep limit automated in the input limit, and was changing it via MQTT. Took us a few months figure it out because was something that as a user I would think will affect ( it is called Input Limit :) )

So happy all is sorted :)

0 Likes 0 ·
shaneyake avatar image shaneyake Javier San Juan Costillas commented ·
If you are running ESS, it is best to set input current limit at the real current limit. IE, the input breaker limit.

You then control the "ESS grid setpoint". This is what you should be controllering if you want to automate ESS.

0 Likes 0 ·
Paul B avatar image Paul B commented ·

The battery voltage should never go high ie if the batteries are full and the grid has disconnected (ie a Grid power failure) then the ess should raise the Frequency to say 53 hz and then your grid tie inverters should DERATE or turn off so that the battery is NOT overcharged. So make sure you have the frequency settings correct set to suit your grid tie Inverters. The frequency iss et in the ESS assistant Ie mine start at 50.30 then then at 53hz the grid tie solar units are fully shut down

50.3 full output

50.4 : De-rated to 90% max output

50.5 : Derated to 80% max output

50.6 : Derated to 70% max output

50.7 : Derated to 60% max output

etc


0 Likes 0 ·
raymiller avatar image
raymiller answered ·

Thanks for finding out the solution to the High Voltage problem! Maybe changing the wording from "Input current limit" to "Input and Export current limit" would be a simple aid?

The confusion with setting the input current too low for maximum export could be avoided with the Quatro/MultiPlus reporting all logic True conditions, not unlike the MPPT regulators reporting their current operating state or similar, maybe even on the VRM graphic. While the MulitPlus II does report some logic states reporting of opperating/exception conditions would be more than useful.

As complexity is increased in our energy storage systems and more new people are exposed to them adding in smarter tools to support the complexity seems in order.

Victron, keep up the good work.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.