I installed Victron BMV702 a few days ago to my batteries (16 x Vision gell AGM 4 strings making 48V), I already installed 4 x battery HA-02 balancers a couple of weeks ago as I was getting hugely varying voltages on the various batteries.

The batteries in my opinion are getting tired, only get about 6kVah till 50% charge measured in volts to 48V

This morning when voltage had reached 48V, the Victron BMV still said 77.4% SOC?

Is this perhaps because it needs a few cycles to accurately measure the batteries state? Or is there an issue?

Hi Colin

The key is calibration with the BMV. Have you checked the calibrated settings...bank ah, etc?  Also make sure the unit is zero'ed under no load... ie with inverter and all other loads off (leave for some minutes so the bank can stabilise.  I made that mistake myself at one stage.

Then see how she goes...

Regards

Mark

 

According to Manual, it should auto calibrate when fully charged.Bank iis set to 400AH

Please can you elaborate on how to Zero under no load?

Please ignore last question Just read manual and did the Zeroing

Busy charging fully with the general and ill see if any difference  tomorrow 

Nice.  Keep us updated...

yes the BMV will autocalibrate but only from a known point therefore the manual zero start point.

You set the Amp-hour capacity when you configure the BMV, and that is where the issue lies. Let me get into that in some more depth.

Over time the battery degrades. In fact, it is probably good for the full 100% only once or twice, then its good for 99% for another ten times or so, and so it goes on. After a 1000 cycles or so, it is only good for half its original capacity. If you started with a 100Ah battery, by that time you technically only have a 50Ah battery left. If you only need 20Ah... then it's still serviceable. If you need 51Ah, you throw the battery away.

So... you fitted a BMV to a tired battery, and probably configured it to the original NEW Ah rating. Of couse it will show an optimistic number... :-)

thanks for opening this thread @Colin

my battery bank is disconnected, and there is no load although the ha02 is attached, and the bmv702 shows -120w and -2.29a, -0.9ah. the bmv702 back-light is on and it must use some power, i presume, but after switching the back-light on/off there was no difference.

the batteries are full and the main voltage is 52.54

is that the normal drain on batteries not being charged and standing on their own?

this might be a stupid question, if so, flame me, i'm coated with teflon

1 hour ago, gabriel said:

is that the normal drain on batteries not being charged and standing on their own?

If you remove the negative cable from the inverter side of the shunt, then there can be no current running through the shunt. It must read zero. Even if the HA02 is still attached, it doesn't matter, that current (if any) doesn't run THROUGH the shunt and won't show. The BMV should measure zero. If it does not, then zero it as the manual explains.

The other end of the shunt must remain connected to the battery, as should the thin red cable to the positive side. The BMV gets its power from there.

thanks @plonkster, i double checked and yes, i thought i had it zeroed [bmv702 program 09] but i did not hear the beep, have done so correctly just now with ears open. i also re-synced to 100% soc as the bank is still unconnected and has been so since 16:00 yesterday, showing main voltage as 52.46.

waiting for a computer to flash the axpert before reconnecting everything.

enjoy ascension day!

God bless

g

Ok, zeroing voltage did make a difference, thus morning 48.8V and 81%

However when wife starts hair drier (1200W) voltage dips to 47.2V

I then run the geny for only 5 minutes then off and she was able to run microwave (1200w) for 3 minute and to toaster (1800w) for one slice without dipping below 48V

Should I ignore voltage dip and trust BMV? 

13 minutes ago, Colin said:

Ok, zeroing voltage did make a difference, thus morning 48.8V and 81%

However when wife starts hair drier (1200W) voltage dips to 47.2V

Should I ignore voltage dip and trust BMV? 

Does the voltage recover after the hair dryer, etc without running the gennie?... I would trust the BMV and see how things go... 48.8V and 81% maybe the best you will do with old batteries (per @plonkster previous post).

30 minutes ago, Mark said:

Does the voltage recover after the hair dryer, etc without running the gennie?... I would trust the BMV and see how things go... 48.8V and 81% maybe the best you will do with old batteries (per @plonkster previous post).

Yes the voltage does recover, never tried the hair drier a second time, maybe I should try that too, a load a few times and monitor the voltage

5 minutes ago, Colin said:

Yes the voltage does recover,

Any battery has an internal resistance which will cause a volt drop in the battery proportional to the current flowing through it (V=IR), so when you attach a high load the current will increase and the battery voltage will drop while that load is attached and then rebound when the load is removed. This is the main reason why state of charge cant be estimated using battery voltage when the battery is under load.

11 minutes ago, pilotfish said:

Any battery has an internal resistance which will cause a volt drop in the battery proportional to the current flowing through it (V=IR), so when you attach a high load the current will increase and the battery voltage will drop while that load is attached and then rebound when the load is removed. This is the main reason why state of charge cant be estimated using battery voltage when the battery is under load.

Thanks, issue is at what stage (how much load and how much volt drop) does it become too far a drop in that we need to believe the voltage more than the BMV?

Or do should I totally ignore the volt drop as long as it recovers?

I would ignore the load related volt drop (within reason) as long as the batts are doing a job for you. When they get too tired a medium load will drop them to your alarm voltage, then time to cough up for new batteries.

If you suspect battery problems then maybe you should check individual batt voltages again now that your HA's have been in for a while. If voltages are reasonably level now with HA's in place then disconnect the HA's in early evening (while high SOC) and check batt voltages in early morning (while low SOC) to identify bad batteries. Do morning tests while the batteries are IDLE (no charge or load).

Maybe you have some bad batteries spread around your bank, if so you could pull those out and reduce to less strings, for example if you identify 3 bad batts then pull them out and reduce to 3 strings which would be better than running 4 strings with 3 bad batteries.

2 minutes ago, pilotfish said:

I would ignore the load related volt drop (within reason) as long as the batts are doing a job for you. When they get too tired a medium load will drop them to your alarm voltage, then time to cough up for new batteries.

If you suspect battery problems then maybe you should check individual batt voltages again now that your HA's have been in for a while. If voltages are reasonably level now with HA's in place then disconnect the HA's in early evening (while high SOC) and check batt voltages in early morning (while low SOC) to identify bad batteries. Do morning tests while the batteries are IDLE (no charge or load).

Maybe you have some bad batteries spread around your bank, if so you could pull those out and reduce to less strings, for example if you identify 3 bad batts then pull them out and reduce to 3 strings which would be better than running 4 strings with 3 bad batteries.

Sounds like a good plan.

With HA's in voltages are pretty stable now within 0.3V (this with cheap Chinese voltmeter on each battery so not 100% accurate), need to check when removed

I think I know those meters you are referring to and they are junk! (because I tried a similar thing using that rubbish conected to my 24x 2V batteries and the results were unusable).

Get a proper multimeter and measure the batteries, O.3V is quite a significant difference in battery SOC so lets hope that they are more balanced than that.

2 hours ago, pilotfish said:

I would ignore the load related volt drop (within reason)

Agreed. When you start a car, the battery voltage can drop as low as 9V for a few seconds. I'd happily accept drops to 46V for even a few minutes. Of course, during these events the batteries heat up a bit (the internal resistance generates heat at such high currents), so it absolutely must be within reason. Efficiency also suffers, because the inverter is doing the work at a lower voltage, so it has to increase the current to get the same wattage. The battery however doesn't gain more amp-hours out of this transaction, so basically the lower the voltage, the faster it discharges as well.

On another note, I want to use the BMV to disconnect load when SOC low, I have ready a few threads on doing this, however the manual says max 60V.

Has everyone adhered to this? Means I need a 12, 24 or 48V contactor / relay instead of simpler, more readily available 230V

28 minutes ago, Colin said:

Has everyone adhered to this? Means I need a 12, 24 or 48V contactor / relay instead of simpler, more readily available 230V

@The Terrible Triplett has put 230V on his and it seems fine. I think the 60V is a DC rating, and since we know that DC arcs over a lot easier, this is probably fine.

The reason why I am not fond of the idea is quite simply that I don't like 230V cabling in the vicinity of my battery cabling. I'd probably use a Solid State Relay to switch the contactor (or switch it directly possibly, up to a few hundred watts), that also gives you galvanic isolation as the SSR typically uses opto-coupling.

20 hours ago, plonkster said:

@The Terrible Triplett has put 230V on his and it seems fine.

The reason why I am not fond of the idea is quite simply that I don't like 230V cabling in the vicinity of my battery cabling.

I have and I do and I am still doing it and it is nowhere near the batteries nor any DC things. 

Victron, the kind company they are, includes (for me) a "free" 15m cable to connecting said BMV to it's shunt thereby ensuring my personal safety. They thought of me see?

So, in my case, said BMV relay switches 220v loads between Eskom (220v AC) and Inverter (also 220v AC apparently) which means I was forced to use said 15m cable, having all on the 'other side' of things solar, where, allegedly, 220v comes out, neatly inside a custom built Interlock where the only current inside is ...

... wait for it ...

... wait ...

YES!!!
YOU.
ARE.
RIGHT!!!

It is all 220v AC! No DC nor batteries even close by!

Give THAT man a Bells! 

(... tears are running down me cheeks from laughter.) 

11 minutes ago, The Terrible Triplett said:

It is all 220v AC! No DC nor batteries even close by!

Except of course the battery power arriving across that 15m cable. But granted, that's fused and properly contained.

My case is a bit different. My BMV sits in a box next to the CCGX, with the vebus, vecan, and rs485 cabling to the modbus energy meter all terminating in that box on the back of the ccgx. An an ethernet cable, and a USB cable going to more equipment. The idea of adding a 230V to the middle of that low voltage mix would horrify me.

I vaguely recall a plane that crashed even though all the cabling in the fuel tanks were low voltage... for similar reasons. Perhaps not the best analogy... but there it is ;-)

19 minutes ago, plonkster said:

My BMV sits in a box next to the CCGX ...

My VenusGX is also close to BMV, USB cables are short, but still, the 220v is isolated and far from the planes fuel tanks, or in my case, hydrogen buildup from the batteries in the wooden box in the braai room. 

You are not winning this one! :-) 

Your setup is quite unique, being a developer and all that, with a lot of stuff to connect and test.

Me and @Colin , we can do things easier and simpler.

1 hour ago, The Terrible Triplett said:

Your setup is quite unique, being a developer and all that, with a lot of stuff to connect and test.

It's actually not that unique. It is pretty common for an ESS setup to have two vedirect cables (a solar charger and a BMV), a vebus cable to the inverter, and an RS485 cable to the power meter. The CCGX is the center of a star network. So as I said, it's not really the BMV that's the "problem" in my case, it's that the CCGX is right next to it. I would never use 230V on the relay inside the CCGX for example!

The one thing I would still prefer to do, is look inside a BMV at the creepage distances around the relay. That would really be the last item before I change my answer from "probably safe" to "definitely safe".