question

Thanks Alexandra. I have checked and double checked the wiring, also disconnected both AC inputs (from grid) and turned on grid power and checked both have 230vac and are line and neutral as expected. All good. It may be grid is out of range, but I checked both the voltage and the frequency of the grid, they measure correct and close expected.

I assume from your reply that grid should be to both Multi's at their AC input via equal length lines, which is how I have connected them.

I have thought to remove the parallel settings, run just one Multi reconfigured for ESS and test again. This would verify if its related to the ESS assistant setting or if related to the parallel Multi's setting.

Another question related to above: how important is the equal length cabling when running more than one Multi? I ask in terms of accuracy and also power. For example, I have relatively short cables from Multi's to batteries and Multi's to AC inputs and outputs. All are typically less than 0.8 meters. To what accuracy should the cables be matched? 5cm? 1cm? 0.1cm?

Also would it be correct to say that any ill effects resulting from cabling of enough difference in length to cause problems, would only be present when significant power is being drawn from the system, in other words, ill effects would not be noticeable when system is in idle and there is no, or minimal load ?

I have read this: https://www.victronenergy.com/live/_media/ve.bus:rva_-_theory_on_wiring_large_systems.pdf

@phil.r

You should start simple.

Try u run one MP at a time and as a plain system without ESS assistant.

Than make it parallel and load ESS assistant only when everything works.

Cable lengths diferences on AC-in and on AC-out in parallel setup would not hinder the system work. They will cause power transfer imbalance between master and slave. To have imbalance it must work first.

Dear Marekp. I have now been able to run some more tests. Reset one of the Multi's as a stand alone single device. Installed ESS assistant, and configured for Pylontech batteries, all works just fine. No problems.

I then reconfigured for two Multi's single phase, both with ESS assistant, but this time, they run for about 15 - 20 seconds, then some relays change over, the slave warns of overload, the fan comes on, and relays clunk about, and it tries again.

I then removed the ESS assistants and ran them just as two parallel Multi's without ESS. but with AC In to both. This time all worked until I applied AC in which then attempted to charge the batteries (which are fully charged so no appreciable current drawn) and similar problem arose, overload warning and fan, and then relays. No notifications anywhere, no error messages.

System is very simple, AC in connected together by equal length cables of same rating, same for AC out and same for battery in, coming from nearby common bus bar.

It seems this is not an ESS issue, but a parallel Multi config issue

I used VE Bus System Config to set up the two Multis in parallel single phase and also to then set up each with ESS Assistant.

Settings for all the above were identical, ie BMS external, all voltage settings the same for all above situations.

Phil

Yes, when running both Multi's I use Victron RJ45 cable to connect data together, as per recommendations, ie, nothing else on end of the string

Yes, run the ESS set up individually, starting with the Master. Upon configuration of ESS Assistant, there is an option to write to both devices, but it it fails to write to both with error message, so I have to write to one, then config ESS second time on the Slave

@phil.r

Just to be sure that both MP-IIs are working, maybe try the single setup with the other MP-II as well.

How you know which one is master and which one is slave?

How are the LEDs when you have them connected as a parallel?

I can try the other in single set up, will do that next site visit.

Master and Slave are easily identified in set up and also by test in VE Bus Config - ie flash the led's

I am not at site now, but from memory, the Master has inverter green LED on continuous and the Slave has same LED flashing, but that is my memory which I know is prone to error :-)

Probably should have done the Multi 10k, but I bought was was available from local supplier

Multiplus2 10k not grid approved?

Hello @phil.r did get to the solution for this issue? i am having almost identical setup with the same problem.

kaspars - see my comment dated 4th May 2022 at 12.29. I found this allowed the system to work, I took very great care to ensure linking cables were same length, exact same length. I reduced cable size on AC side to borderline schedule, I read this helps due to increasing linkage resistance, and I removed the data cable from the trunking carrying the power and DC cables. It worked, and has been working, but I cannot say I am satisfied - the victrons make a lot of varying humming noise. Lack of support from Victron, a very iffy performance and terrible patchwork of data protocols have led me to make the decision to abandon Victron, I now install other brands, Studer being my favorite. A Studer parallel system in ESS type mode is so simple, seamless and solid to set up, I would never go back to Victron