Good day everyone
Noticed something odd on my VRM app today when looking at the power used from my 4 BYD 5kw batteries for this month and the amount of power supplied to charge them for the same period.Power into the battery was 400kwh and power used was 194 kwh for this month.This does not seem correct as BYD battery efficency is around 95% and currently it is less than 50% .
Am I missing something here?
Any feedback would be greatly appreciated.
The battery might be that efficient but the system can be less so.
Remember there is converstion before and after the battery.
Though you are right. I would expect closer to 75% for the entire efficiency.
There is alot you can look at for evaluation purposes, for example when inverting a load see what the load is vs what power is being taken out if the battery.
How warm does everything run when it is charging and working hard? How long do you spend balancing? How is everything connected?
I see feedback to grid amount as well - I assume you do some of that from battery. The one screenshot only includes from battery to inverter loads (not to grid) if i am not mistaken {Consumption tab?}. So probably alot of it is there too.
Thanx for the input…will do a lload check tonight.I have also noticed that battery was fully charged by 10.00 am this morning but since then the battery has received 4.5kwh additional from PV without any power draw on the batteries…this also seems odd.
I welcome input on this.
You can try to download the “Production report” from the download section of the “All installation” page.
You can see there the real efficiency…
Hi Guy @guystewart
Happy New Year! It’s new year on Sydney time. 
There was a topic opened by me about Pylontech efficiency. I can’t seem to find it anymore… among many other topics…
Can you please help?
Thanks!
Yes, do this for e.g. the last 6 months. At my system (3x MP2 48/5000 + 15 kWh 280Ah/16s low voltage) discharge : charge = about 73%
Thanx Alex…just downloaded the info and I am at about 58% for charge /discharge which is not good.Also did a load conversion test now.DC battery power converts to AC at around 93% under load which seems OK.
If fully charged by 100% SOC?
AT that point, batteries may be only 95% - 98% charged, and balancing takes some power too.
I have a question.
Does your time period include your initial charge?
That does skew the initial stats a bit as you start at around 50% soc depending on where it was shipped. (Although still not enough to skew it that bad.)
I agree 20 kw either way will not change the percentage alot.Just cant figure out why its so bad.
How fast are they discharged and charged?
Do you have them individually connected to a bus bar?
Discharge is normally slow with a spike here and their and charging is controlled by the BMS.
Batterie are connected in parallel to each othervia the battery cables
These are the byd flex or flex lites? When you say parrallel connected you mean they are daisy chained? (One battery terminal connected to another battery terminal?)
It is much more efficient to have them each having their own connectiln to a bus bar. That is how they describe it in their install manual.
There are a few reasons for that, one is efficiency.
Hi @Trevor
The problem here is that VRM only uses AC as Consumption. It doesn’t count system losses, DC>AC conversion losses, nor DC direct use (even if you have a shunt allocated to that). So those losses end up getting blamed on the battery.
Doing it that way, my own Pylontechs show ~54% RTE. Useless figure, best ignored.
Hi John …thanx so much for clearing that up it quiet a relief knowing my system is operating normally.
I was thinking wiring losses in the dc side, but not because you have dc direct loads (since I don’t see any). How hot do the cables/connections get?
Hi John
I dont have any dc loads which I know of and the cables are relatively cool when I feel them under load…
Just the standby load on a Multiplus is significant. The spec sheet will show that. And under load, have you ever heard the cooling fans running? Rest assured, the figures you’re using are attributing that heat generation to battery losses.
Then there’s mini-cycles. You may be assuming one battery cycle per day, but if you have loads that exceed solar generation you may be prone to that too. I have a switching inductiom cooker that cycles >1000W on the batteries when solar is fading at dinnertime. Being generous, the batts may well have a 99% RTE, but that’s happening every few seconds. So there’s losses there too, and almost impossible to quantify, although the batts continue to warm a little even though ambient is dropping.
All these little things add up, and the standard VRM figures won’t even come close to showing something useful for this purpose.
I had this concern when I first started using Victron. Got so many suggestions on the reason for the discrepancy like what you have received here. Later found out that the self-consumptions of the Victron devices (like Quattro, et cetera) add up over time and accounts for the discrepancy. When I configured “has DC” in Cerbo, the figures started adding up correctly. Then several months later when I added multiple smartshunts as battery and/or energy meter to account for addition of (different) battery bank, et cetera, VRM / Cerbo messed up the calculation again and the energy figures stopped adding up correctly.
It does not matter if you have BYD, Pylontech or other brands, all theire BMSes suck.
The Self-Consumption is horrible high and is is to 90% the case of all people complaining about the efficiency of theire Li-* Batteries.
Have a nice day
