question

Dave Rae avatar image
Dave Rae asked

AC/DC Coupled Off-Grid System

Hello,

I am running a fully off-grid system. Fronius Primo 8.2-1 AC coupled to a Multiplus-II 48/10000.

DC coupled side consisting of 2x 150/70 MPPT. Total solar is 15KW. 10KW on the Primo & 5KW on the MPPT's. 8x US3000 Pylontechs are the storage specs and a CCGX brain box.

After recent firmware updates I have noticed that after the batteries are fully charged that the Primo's output is 0-watts at an inverter frequency of 52.5 kHz, and any load is being directly covered by the MPPT's and not the Primo.

Before this it was the other way around, the Primo was covering the loads consumed directly and the MPPT's were brought into the system only when an increased load was applied and slowly reduced as the frequency was reduced to increase the Primo's output. The system seems to be biased towards the MPPT's now and not the Primo. Has Victron changed something here recently?

Cheers.


AC PV Coupling
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.

1 Answer
Adam avatar image
Adam answered ·

Hello

I'm not sure if this is a result of an update in the new firmware, but its defiantly acting the way I would want it to in those conditions, if it is to do with the new update then nice work Victron.

In my opinion it makes more sense when batteries are full to prioritise DC coupled solar. As AC coupled on a full battery can be a bit of an issue, EG: if your battery was full and you turned on a 8 KW load, and the Fronius was covering the full load , the moment you turn that load off you have a momentary 8KW that needs to go somewhere for the split second it takes for the Victron to ramp up frequency to limit the Fronius, so that power is sent straight from the AC Side to the DC side ( to absorb) where it will shoot into already full batteries. Not such Big issue if using lead, but if using lithium or flow might trigger a BMS overvoltage alarm and in the right circumstances could cause some surging on the AC side if there is literally nowhere to put the power. I think this may be able to happen a little on the DC coupled if you had a large array but the point is that its less likely to cause any issues by using coupled over AC coupled when on a full battery.

Well that’s my opinion anyways and what my experience has shown , but I’m happy to be wrong and look forward to other opinions , hopefully this helps answer your question

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.

Dave Rae avatar image Dave Rae commented ·

Hello, yes that is a consideration but prior to this I never had any problems as the Pylontech BMS reports a full charge at 53.2v and Victron has derated this to 52v in their charge parameters, so it leaves headroom for AC load overshoot. The whole point of AC coupled system is to use the direct conversion from the string inverter straight away, it's pointless to have AC coupled operating in this way. As the MPPT's are controlled through the VE Direct connection and allow instantaneous control that is what should be ramping up and down to make up the slight variations. Victron would be better off using the direct control of the MPPT trackers in the Fronius as do SELECTRONIC for instantaneous control instead of the primitive frequency method which has a huge reaction time problem not to mention some timing related issues on certain appliances that require a stable 50Hz.

cheers.

0 Likes 0 ·
Adam avatar image Adam Dave Rae commented ·

Yes you are correct and it shouldn't pose a issue with pylontech , but can pose a issue with other batteries. It's always a issue with Flow batteries, well Redflow to be specific and on those sites I am happy for the system to prioritise DC coupled as the reward is less chance of over shoots. In mind it's much of a muchness, its being throttled anyways so to me it doesn't really matter what source the solar comes from, as long as it ramps back up reasonably swiftly once the load increases.

But yes Interesting change in operation and I suppose there are some pros and cons to both depending on perspective and system set up. probably need someone form victron to chime in to tell us if it's intentional or just a unintended bug

Also were you running DVCC correctly before you updated firmware ? as the DVCC is forced on with the updated firmware , which could ,depending on how your DVCC was set up before change the way the system interacts with its components. Just a though , thats about all I got , ahahaha


I don't suppose you have read the firmware change logs to see of it is mentioned in there ?

0 Likes 0 ·
Dave Rae avatar image Dave Rae Adam commented ·
Yep have read all the change logs no mention of it. Yes DVCC is all configured correctly. No big deal I guess we will wait to see if future updates correct the issue.

Cheers.

0 Likes 0 ·
Adam avatar image Adam Dave Rae commented ·
Interesting, , as you mentioned I suppose we will wait and see if it changes in a future update, please if you do notice a change let me know as I'm interested to see what happens.


one last ditch idea, I don't really think this will work because DVCC should have control over all the Victron equipment anyways. But as a experiment you could try dropping the absorption voltage/ float by a couple points of a volt in the MPPT, so that the limits in the Multi are higher, that might make it take priority over the MPPT but i doubt it as DVCC is the overarching control there.

this doesn't answer the question as to why the change happened but maybe might get it operating the way you wish again until the "issue" is resolved.

0 Likes 0 ·
Dave Rae avatar image Dave Rae Adam commented ·
Yep that was one of first things I tried. Interesting side note just today one of the ve:direct inputs has failed. Done all the standard testing, cable is fine swapped to other port it works fine, so definitely a hardware failure. Unit is 3years old.

Cheers

0 Likes 0 ·