question

it does not say low SOC, but waiting for sun

(rather cloudy now)

It seems the EVCS position is not correctly configured. It is defined as connected on AC In, and actually it is on AC out.

no, its actually connected to our home-grid and the victron ist only connected to it an AC in

Here is a quick illustration of my setup:

Hello,

I have a similare pb, wen i plug my VE. My ESS system have SOC 100% a véhicule start charge, after 30 second ev charging disconnect waiting for sun, after 30 second he restart charge….

My system is an ESS 3 phases With 3 multi 48/5000.

CAN you help me?

Hello

I have exactly the same issue. No matter what setting I setup the charger start charing in auto mode and stops after approx 55 sec. Could you please help me to arrange that. Attaching picture from just now, sunny day, great production but charger is waiting for sun.

Hello folks,

I have the same problem with the automatic mode.

I have a relatively small pv system with 5.6kwp but when the battery is full it all goes into the grid. I have a multiplus 2 48/3000 ESS with a cerbo gx and 3x pylontech us3000c.

everything works fine but the automatic mode on a phase with 6A does not activate automatically in case of excess.

I have already checked all settings from this contribution but without success.

I would be happy if someone had a solution

the problem is that I can't track my solar system via the vrm portal because it goes directly to phase 1+2 via micro-inverters. Should I then add 2 modbus energy meters to the feed points to make the whole thing visible? I thought the grid point is the trigger for the EVCS

the system works perfectly. i defined 2 em24 ethernet via tcp modbus as pv inverter to monitor the complete pv system. one tracks phase 1+2 and the other only phase 3

thanks for the help with my problem <3 great support

A function from version 1.24 is no longer active. If the battery is not yet full but more than the maximum charging voltage is generated (battery can be 30 A) PV supplies 50A then the wall box starts when the 1.3 KW feed is reached and charges the car. From version 1.25 this is no longer possible the start depends on the LOW SOC value and it doesn't matter whether 2kw is then fed into the grid because the battery can no longer absorb charging power. Therefore, I think only version 1.24 makes sense for many people in order to be able to charge the car early in case of a lot of excess despite battery charging max. You give away a few KW, if the output of the PV is very high and the battery cannot absorb that much, which is then missing in the afternoon when the battery is full for charging the car.

Hi, sorry but I don't understand what the issue is, please try to explain better

What do you mean by “battery can be 30 A”, maybe 30Ah? But I still don't understand what is the connection with EVCS

I have a 100A Battery. this Battery can be load with maximum 30A. I have in Summary 10 KW Peak PV-Modules. Up to 10 o´clock I have much more Power as I can load into the battery. Only in version 1.24 the Chager start charging the car in case the Battery load an parallel Charge the case the Soc is not nessesary because Maximum Load to Battery and more the 1.3 KW available. Up to Version 1.24 you configure a min SOC for Battery. And equal if more Power is available the Charger will not start before the SOC is reached. So I put many power direkt into the Grid for the time until the SOC is reached. in the Version 1.24 only a value for discarging able to set. I set this Value to 98 % so the PV System try to load the Battery and the car in Case that I have enough power. And it will stopp direkt if the sun is gone and the Batter is not higer than 98 %. In the newer Version this funktion is not available because SOC must be higher than Discharge Value. This configuration in the software does not take into account that it may be that the min SOC value is not reached but the maximum charge value in amps of the battery is reached. There is further PV excess available for charging the car, which is then not released is because of the minimum SOC value. It is necessary to remove the lock that the min value must always be greater than the discharge value.

Hello @Lucian Popescu (Victron Energy Staff).

Sometimes I am dealing with problem regarding stuck in "waiting for sun" state with newer FWs (starting from 01.26).

I have manual disconnector in electrical box, which can interrupt second (L2) and third (L3) supply phase of EVCS. When there is enough PV power and clear skies, I am charging EV in auto mode using only 1 phase, when I must charge the EV during night or there is no enough PV power for charging I am using manual mode and all 3 phases.

What are the conditions for start charging in auto mode? Is it everytime only feed-in power (sum of all phases) or feed-in in only first phase L1? Or is there some internal parameter which remember previous charging (if it was 1 phase or 3 phase) and then EVCS in auto mode waiting for min. feed-in power based on this parameter (1x 230 V x 6 A or 3x 230 V x 6 A)?

And of course what is the minimum time of feeding into the grid for activation?

I done some investigation and for example if I am reading register for manual mode phase setting (no. 5055) and auto mode phase setting (no. 5056) via Modbus, everytime I get 65535 so no default value and value out of the range according to EVCS Modbus Excel sheet.

I have the same issue, even with sufficient sun (checked with AC loads), the EVCS stays "waiting for sun". Victron is proud of this "shitty non-working" feature and they don't even know how is it working and why not starting (great programmers - I guess India) !!! You sould atleast consider irradiance factor, people are expecting EVCS to start as low sun as possible, not to be stuck on "waiting for sun" with sun in front of his eyes. Is a disgrace for the price you pay, I don't recommend it at all ! Shame on you victron guys.

For instance, in the morning it was available more than 3kW from PV, but evcs didn't start on auto, why is that? On manual mode is working ok.

I’ on manual charging remote right now, the problem is not my system, is the triggering event of auto charging as you could see from almost all users, whete you don’t have a clear answer on that. Maybe you shoud go back and consult with development team, perform some DOE’s and test on the complaints and improve next FW

I have 20% with 80%, in the morning I had 97%, didn’t start even with the clear sun 500W/m2. My pv power istalled is 16kW. Another issue i’m facing with the station is that LCD dispays garbage when it shoud display real values. The connection signal strength is really good -50db. Why is that?IMG_7934.jpeg

Display issue is present in manual mode as well, not in auto mode only and the firmware is up to date.

On the station ip, when i’m connecting to it, the information is correct, exactely the same data as on VRM. When I’m checking on ev charger display, it shows freezed frames, as in my previous picture with incorrect data.

Yes, charger works in auto mode without any issues but not in the morning. I’ve tried in full sun in the afternoon and is working, but if i connect it in the morning and wait to catch the trigger, doesn’t work - with the message waiting the sun and stays like that. And all these with a faulty data LDC.

Or atleast this was my expectation in the morning, may be I’m wrong and I need to wait for more. But is frustrating having the sun and not charging.

ok, thanks, i’ll make more tests in the next morning and keep you posted.Thank for support.

Meanwhile please provide a solution for the faulty LCD display because I cannot operate on the station, will it be corrected in the FW update?

Hello,

I’m coming with some test results and problem solving after spending some time on troubleshooting auto mode and display issue on evcs. Firstly, the LCD display issue with freezed and overlapped frames was caused in my case by some EMC interferences caused by A1(P)-brown and A2 (blue) wires which were overlapping on top of LDC data bus flexible cable. Fixing them on high voltage area instead of close to LCD solved LCD issue(including slow or missing touch response):

Other issue that worried me was LOOSEN SCREWS on both A1and A2 screws but also on green left side connectors(there’s not high power on these wires but nevertheless VERY RISKY AND FIRE HAZARD):

When you as customer buy such evcs, you shoud inspect it internally very carefully as I mentioned above. I assume Victron will take care about these issues in the production for at least next ev charging stations and I hope my investigation helps. Regards

Yes : HQ2301VEUAZ

Update, EVCS working great since my last workaround. No more display issues + auto function working as well.

Hello Lucian, I just installed EVCS, it works well, only I cannot find option "Allow low excess solar power for auto mode" in the menu. I have v1.28.

Picture from Lucian:

My menu v1.28:

Hi, that will be part of 1.29 FW, that will be released next week

Hello team,

I'm sorry, but I can't understand what he is waiting for?
My AC-Coupled PV is already feeding in quite a lot and the car supports 6 Amps loading as well.
Even if there potentially isn't enough power on 1 phase, Battery being at 99% could compensate

Will try that 1.29 FW once it comes out...

I've had this problem, and 1.29 does not appear to solve it. Offgrid system, ~12kW of PV, EVCS wired for single phase, and system is set to use Auto mode to minimize battery discharge and charge the EV PV direct as much as possible.

I've found if the car achieves charge and the EVCS goes silent; if I unplug and drive into town, when I return and plug back in the EVCS remains in Waiting for Sun mode, even on a brilliantly bright day with plenty of solar potential. I can force a manual charge and manually dial it up to full array capacity, but that's not the point.

Let me know what additional information you need to help track down and squash this bug!

Thanks Lucian, it took me a while to find that setting - it's not available in VictronConnect app on Android v6.08, and on iOS the app doesn't work with the EVCS (results come back as null). It took me some poking but I finally found it via the web access portal, which seems less than ideal. Shouldn't this setting be accessible via VictronConnect as well?

I've enabled low sun and will let you know how it behaves. Looking forward to this!

This bug still appears to occur in 1.29, even with Allow low excess solar power for auto mode enabled. I just came home and plugged the car in, and it appears that if the MPPT is in a curtailed mode it doesn't "wake up" and go back full throttle. PV input continues to match load, and EVCS stays in "waiting for sun" mode. I can switch to manual mode and ramp up the charging current, and the MPPT matches it up to the full ~10-12kW presently available from the array, but if I switch back to auto, the EVCS goes back to waiting for sun.

It appears that Allow low excess solar power for auto mode just allows the house battery to discharge to the car - perhaps needed for grid connected applications, but it’s not ideal for offgrid. There needs to be a signal for the MPPT to test the load, to see if it should stop curtailing and ramp up output.

Remember, in off-grid systems the battery is not the backup, it's the foundation. If the EVCS does not control the MPPT (or provide signals to what can control the MPPT), then I don't know whether you call that a bug or a design flaw, but it's a foundational error.

You can see in the attached screenshot video that the house battery is at idle, the array is simply matching load. When I plug in the car (which desparately needs a charge) the EVSC remains stuck on Waiting for Sun, because the battery remains at idle. It's only when I change to manual and initiate charging that forcing the charger to activate "awakes" the MPPT, which immediately matches the new load. Increasing the charge limit shows clearly there's more available from the sun.

Surely it would not make sense in an off-grid system for Auto to kick in only when the battery begins to discharge, as that's typically long after sunset, correct?

In short, the equation is a three-legged stool. Currently the EVCS/Cerbo combo is very good at sending surplus PV above what's needed to charge the battery (or send back to grid, if I had that) to the EV, so those legs work. The leg that appears to be lacking is the ability for the EVCS/Cerbo to probe to see if there's surplus available from the PV, especially when the battery doesn't call for it. I expect this would also be a problem for homes with non-export limits.

Trust me, I'm truly impressed with the overall Victron package, and I've seen behind the scenes of a lot of inverter platforms so I know a bit about how hard it is to get to the level that you're offering; my hat is off. I'm really hoping we can work together to implement fixes to this as your platform is so close to the holy grail of dead nuts reliable, with deep configurability but the ability to be seamlessly mindless from the customer interaction POV. At the end of the day people just want things to work, and cars charging one day but not charging if plugged in mid-day is the opposite of that.

Video upload didn't appear to work so let me try this link share...

https://photos.app.goo.gl/VJqLAGuyUGmUf5iR6

When this is happening, can you tell me the VRM portal ID, so I can connect and take a look?

c0619ab46187, feel free to take a look and let me know what you think

Also while you're in there looking at data, in response to my other post https://community.victronenergy.com/questions/328222/evcs-needs-response-to-inverter-temperature-alarm.html take a look at August 17th as an example

@Lucian Popescu (Victron Energy Staff) I just returned from an errand and plugged in, and the battery is idle and the EVCS is in waiting for sun mode. I'll leave it this way for a bit so you can check but then I'm going to need to override so I can charge my car :-)

I’ll check that next time. As it was, it hung in that mode for an hour until I switched to a scheduled charge for long enough to discharge the battery slightly, then changing back to auto it worked fine.

I made the mistake before of choosing manual, only to forget to change back and I tanked my house battery :-(. At least a scheduled charge has an end time :-)

Good evening, everyone!

I've had my EVCS for a while, but only got around to installing it today (the PHEV it was meant for was also delayed). I also bumped into the problems cited above and eventually got it to work by updating to the latest firmware and setting a schedule that the unit appears to follow without issue.

Good news:

• it integrates nicely with the rest of my Victron system, though the overall UI concept does need a bit of polishing (EVCS only shows up in VRM, not on the dashboard of the CCGX; the EVCS web UI, while probably well intended, appears goofy with those large fonts etc.)
• my car has no trouble negotiating with the EVCS, charging always starts when commanded and my car has no trouble following the advertised current limit, down to 6A

Bad news:

• an electrical safety issue around the connector, that was probably deemed not important enough to attend to during product design (I am an educated user myself and would never willingly abuse any power supply to make it dangerous -- but IEC 62196 was not designed with an educated user in mind, on purpose)
• the sun issue, outlined in the posts above and which I will detail below, noting that I'm willing to volunteer any support required by Victron (including writing and donating code, running beta versions, providing logs etc.) to have this fixed, if there is a will
• the screen doesn't appear to turn off if a car is connected, there is a schedule configured and we're currently outside of it (i.e. according to the defined schedule, the EVCS should do nothing right now).

The sun issue:

• the EVCS has to draw power from somewhere to charge the car. Owing to the overall theme of the Victron brand, it defaults to renewable power for that, but the way this is communicated to the user is unclear at best. This causes the device to violate the principle of least surprise (i.e. you plug it in and it just works), causing upset customers to show up here and complain
• for some unknown reason, despite the Victron EVCS not being the myenergi Zappi, it kinda tries to behave like the latter. The difference is that the latter has a CT with which it can sense, in real-time, the energy posture of the customer's ESS and, as a result, can start/stop charging the EV at a moment's notice. The Victron EVCS, on the other hand, uses the data gathered by the GX device for that, but somehow fails to understand its semantics:
  • if the ESS is configured to export to the grid (and the grid is currently not limiting export), the EVCS will read the PV Output Power data and if it's above the proverbial 1.4kW, it will start charging -- this is OK
  • if the ESS is configured NOT to export to the grid and the (house) battery is already full, the MPPT bank will be kept at idle by the GX device. The EVCS will read a low PV Output Power and decide there's nothing to send to the car -- this is NOT OK
• so, how do you decide if the sun really is up (by at least 1.4kW :D )? Well, the first question should really be: why not allow the user to choose where the power should be coming from, in the first place, instead of hard-coding it? Just display a list and allow the user to sort the elements of that list (sun, battery, grid etc.) in their preferred order. Second, if the user did choose "sun" as their first option, we have the following options ourselves:
  • common sense (is it night in the current location?)
  • educated common sense (are we after sunrise and before sunset in the current location?)
  • aristocratic common sense (are we during daytime and is the sky clear in the current location?)
  • some form of sun light sensor (cheap or expensive -- I'm not sure VenusOS supports anything like that)
  • look at what the MPPT bank is doing vs. what the current policy set by the GX device tells it to do (i.e. 1W solar power + stay down == we expect more PV power to be available vs. 1W solar power + go nuts == it's probably dawn, dusk or overcast)
  • ask for forgiveness instead of permission: start charging with 6A and observe the system. If the MPPT bank follows up, raise the power to the configured maximum in steps. If the MPPT doesn't react, stop charging after a few seconds.
• bonus points (remember, I volunteered to implement these :-) ):
  • support for astronomical schedule, that is the ability to use "from sunrise" and "to sunset" instead of only hard-coded hours
  • priority queue, that is the ability to set a charging policy like so: "Sun if available, then battery maximum 3kW for a maximum of 5kWh, then grid maximum 1.4kW for a maximum of 10kWh only after 22:00 and before 05:00"
  • ducking, that is the ability to tell the EVCS to follow a certain maximum power curve, referenced to an arbitrary device in the system. For example "Charge with 16A, but duck if the battery draw approaches 50A" or "Charge with 10A, but duck if the inverter load approaches 4kW" etc.

Thanks for reading, I've been a Victron user, fan and installer for a while now and I have a vested interest in your products being a masterclass in their respective areas.

- There is no safety hazard, that's just your opinion. It is allowed to disconnect the charging cable under load. If you look at the pins inside the connector, the CP pin is shorter, so when unplugged, it will force to stop charging. I gave you the same answer on the other post.
- the automatic mode is not working as you are describing above, there is no need to have extra 1.4kW to start charging.
The algorithm has to be functional for all locations, for all system types, for non-Victron MPPT or PV inverters. Adding extra light sensors will make the system more complicated, if the sensor is not correctly installed or dirty, the EVCS is no longer charging?
The simplest way to do it for now was to probe the available power from time to time, but only when the battery level is above a certain value (configurable by user). And if during this test, if the battery is not discharging, the power will be increased.
We are planning to move the AUTO algorithm into the GX device as part of DESS, and then it will be way more precise, with dynamic price implementation etc. But that's for later.