It was meant as a joke, But I am now going to take a safety measure that the Multi rs will switch off completely if battery power fails.
1 Like
Ah okay, thank you… with some of the comments/questions I see in Community and elsewhere, I genuinely can’t tell when people are joking anymore!
That’s because they are not using the emojis when posting…
And I admit, I am guilty of being one of those who like to push buttons and take the bear’s side… 
Life is too short to be frowning/glum.
Surprising is how many are falling for this… 
But now, really and seriously…
If one would know how many issues and potential problems that are waiting to happen are on some products’ firmware, would take all and any supplementary measures for protection… Just that the conditions are not met… (yet).
Still one question / issue remains:
I have an 48 V battery combined with multiplus 3000/48.
Battery is 280 Ah LFP 51,2 volt with JK BMS.
It happened to me that the BMS triggered a sudden shutdown (discomnect) because of low cell voltage. Could that damage the Multiplus?
The below is my point of view, but this can be far removed from the truth…
When inverting, no. (your case)
When charging, well…, that’s debatable. Depends on the currents.
Still, I don’t think that Victron didn’t take into account all situations and took all precautions, loosing DC connection being one of the most encountered situations: BMS disconnect, loose wire, fuse blow, etc.
Generally, the hardware is well built/designed, but one can’t vouch for the components and workmanship. Therefore surprises can happen.
2 Likes
Yes, agree. That’s why Victron only recommends their own BMS’es
These BMS’es - at least the CAN based - have a pre-warning when the BMS senses a near critical situation
I assume the Multi is then instructed to ramp down the charging current to safe levels.
The JK BMS is not cerified or recommended bij Victron, so i am on myself…
does victron have there own bms they recomend?
Yes Victron has a range of BMS systems from small to large (Victron) batteries, only 12 and 24 volt, not 48 V
Further there is a list of approved battery manufacturers on their website, for which the CAN communication is tested.
For separate BMS to be added to a custom made battery, they approve 4 manufacturers - see Victron website. Maybe more to come.?
JK, JBD, Daly etc. i.e most Chinese manufacturers are not recommended, because of whacky and unpredictable software and untested upgrades.
The batteries I have already have the jk 
I have a multiplus 24v 3kw and I was testing AC connect / disconnect programming by using a small 5 amp power supply. I had also set it to prioritize solar/wind power (sustain mode.)
(The following is a bit of a repeat from another thread I have on here, but there is a bit of additional information at the end of this post.)
In the past, when the BMS units in the WattCycle batteries hit charge disconnect, I noticed the voltage spiked up over 30 volts. I put my Fluke 87 in max/min and re ran the test. 31.5 volts peak.
What is interesting is that both Victron MPPT units I have used, a 1kw Phoenix inverter I was using at one time, AND the Smart Shunt all read within 0.02v of each other.
The Multiplus consistently read several TENTHS of a volt low, no matter it the Multiplus was under load or not. (Yes, I went through all the DC cabling testing, remote sensing wires on or off, ect.)
I further found out that the charging circuit would OVERSHOOT the voltage limit programmed by another 0.5 to 0.75 volts in addition to the BASIC ERROR in the measuring circuitry. So, if I wanted a peak of 27.7 volts on the batteries, I had to OFFSET it LOWER by over a volt and a quarter or so. (Look at my other thread for actual numbers. I’m just noting this down by memory..)
The point is with LITHIUM, several TENTHS of a volt make a difference, and errors of over a VOLT are completely unacceptable.
I squawked at the dealer, and they offered to swap it out.
I made them test their stock unit BEFORE I tore my inverter out, and it also had the issues.
They sent it in for repair, and only when it tested ok, did I trade. It was still off slightly, but acceptable.
Jumping forward to the last few days…
Well, that voltage overshoot issue is still present as I discovered the other day, when I was testing the AC disconnect/reconnect set points in combination with the renewable priority setting, (sustain mode,) and the unit configured in UPS mode.
When I slowly ramped down the voltage to hit the AC reconnect point, the power supply went from providing a few amps at roughly 26.3v (trip point was 26.3v,) to the inverter and the voltage display on the power supply jumped up to over 30 volts, and slowly reducing back down to 27v or so. (No batteries connected at the time, just the power supply, and yes, I understand the lack of a load can cause overshoot/instability.)
When testing WITH batteries connected and then running the batteries (100ah @ ‘24’ volts,) down to the lower reconnect point, the multiplus went into the Sustain mode correctly, supplying the load with ‘shore power’ and trickle charging the batteries at 5 or 6 amps or so. I had set the Sustain voltage to about 26.5 volts. I let it sit for several hours and rechecked. The current had dropped to fluctuating to an amp or two negative, then positive and back to zero, but it had obviously stabilized. The measured voltage was over 0.3 volts HIGHER than the programmed set point, AND the Multiplus was CORRECTLY displaying the voltage measured on the batteries. I repeated the test several times. Same results.
So, the Multiplus KNOWS the correct voltage, and is displaying it correctly, but the charge control is STILL overshooting and HOLDING that voltage 0.3v higher than the programmed set point.
My question / observation is simple.
/rant mode on
Why is it that Victron can produce equipment such as the MPPT controllers, the Smart shunt, etc. that have DEAD NUTS ON measurements compared to each other AND my 87 Fluke, (which has been calibrated to a NIST traceable source,) but produce a line of product like the Muliplus / Multiplus2 that are such an obvious TRAIN WRECK (lots of comments all over the place about MP and MP2 units,) when it comes down to measuring simple things like battery voltage, AND not regulating to a PROGRAMMED value? The MPPT controllers I tested kept the output voltages within several hundredths of a volt of their set points without any difficulty, why can’t the Multiplus units do the same?
/rant mode off
Justin, since you said you used to manage a Victron depot, I would really like to hear your thoughts on these topics.
JK isn’t a bad balancer.
I’m starting to see more “DIY” (even though marketed as proper brand) batteries that are using JK as balancer and Seplos for BMS.
This is wildly off-topic so should probably be moved to a separate question if you want to pursue it, but in short: yes, I ran a Victron Repair Centre in the US at a major distributor for a couple-few years, not sure if that’s what you meant by “depot”, but my personal thought is, the MultiPlus is a big inverter/charger that’s pretty reliable, almost insultingly cheap, and not at all a precision measuring instrument.
Off by a few tenths on a 24vNom system, so that’s around 96% accuracy, which -again, my personal view that no one else needs to like or agree with- I don’t think is worth yelling about when it costs a third or less of a comparably-capable comparably-reliable system.
2 Likes
So if my batteries have the jk bms i shouldnt have an issue using a multiplus?
I never said that and if I’m reading this entire thread I think it’s pretty clear that running a Multiplus without batteries is A Very Bad Idea, EOD.
Next to that there is a common misunderstanding about a BMS and a Battery Balancer.
With Lithium batteries, you want both but they do not necessarily need to be from the same brand or in the same package.
A Battery Balancer will make sure your cells are properly balanced (= evenly charged) so the complete pack remains in top shape.
A BMS (Battery Management System) communicates with the inverter/charger management system so (dis)charging voltage and current of the battery pack happen in an optimized way, again to keep the battery pack in top shape.
To repeat: the JK boards make good battery balancers, but rather lousy bms’es.
So no, I wouldn’t rely on them too heavily to think that the entire system will be flawless and incident-proof.
What you could do is add a proper (ideally an officially supported) BMS board to your batteries and use the JKs purely as balancers.
Those will add protection to you batteries but if you’re looking to protect you Multis then the best advice I can give is to repeat what Marc and Justin have already stated: don’t run them without batteries and don’t try funky business.
My system is just going to he a small rv set up with a max of 4 batteries. Maybe some watching tv using lights etc nothing heavy duty. do I really need a balancer and separate bms?
Just trying to wrap my head around all this.
4% error will get you a speeding ticket, too.
It’s advertised as a lithium compatable inverter / charger.
Tenths of a volt matter with lithium.
Lead acid, not so much.
It really sounds like an engineering screwup or factory calibration /procedure error somewhere.
Not really, but I saw that Victron programmers love to use, for smoothing out the measured values, single-pole recursive low pass filters.
Depending on those filters’ parameters, they tend to exhibit the above behavior, usually a lagging one…
Uh… I mean I guess that probably depends on what country you’re in, but in general no, it certainly will not.
And it is - no false advertisement there.
Ah yes, the “lithium is a delicate flower” argument. Look, everyone can treat their batteries the way they decide. The neat part about Victron’s chargers is that, if you don’t like how it’s acting, you can adjust it to suit.
Yeah I mean, that may well be the case, and I’m honestly not here to defend the company - I have my own issues with some of their choices over the years. But again, it’s cheap, it’s pretty reliable, it’s highly configurable, it’s reasonably accurate. Could it be better? Of course, no question… and it could be a lot more expensive too. These days I’d count my blessings, personally. It’s a little odd -and again, this is just me personally- to pay for a Toyota, be delivered a Lexus, and complain that it’s not a Range Rover.
If you’d like to raise the topic for Community discussion, please do so in a separate thread, as per the Community Guidelines. This thread, with the actual original question having been asked and answered, is now closed.
2 Likes