Battery balancers HA-02

[___]

31 minutes ago, SilverNodashi said:

One could easily argue that since the Zhiel device is on the list, the rest doesn't  

But if that is true, then can't I put a Ziehl between an Infini-V (or any other inverter) and declare it good enough? I think one has to understand that inverters aren't the only co-generating devices found in cities, there are even companies running synchronous Diesel generators that also need anti-islanding, or farmers running hydro plants. Perhaps passive protection is good enough and our local propeller-heads are just insisting on that second physical disconnection device? Clarity would indeed be nice :-)

Maybe I need to read NRS097-2-1 again. Not that I understood it the first time I tried...

[Chris Hobson]

I have one HA 01 (2 x 12V) for R300 + postage

[Chris Hobson]

20 minutes ago, plonkster said:

Oh that is nothing. It goes over 3% for a short period. Hey, it's Ex-UPS batteries that I've had for 18 months now. When they were new they were only rated maybe 700 cycles to 50%.... :-)

I have one HA 01 (2 x 12V) for R300 + postage.

[___]

On the topic of anti-islanding, here is my take. Reading the doc itself, starting around page 8, there is a list of things your device should not do, like:

• Disconnect outside the prescribed voltage ranges, your job isn't to try and regulate the voltage
• Don't cause flicker
• Don't inject DC
• Harmonic distortion below 5%
• Don't mess with the power factor. Definitely don't make it worse. If you make it better, stick to 0.9.
• Synchronisation limits of 5% on voltage, 0.3Hz on frequency, phase angle of 20 degrees.
• Disconnect if any of the above is out of bounds

So up to this point, my amateur understanding is that a passive device such as the Ziehl could potentially do the job, though don't quote me on it.

Then comes the next section, 4.2.2.4.2, which says you need at least one active anti-islanding component. In theory, this might then allow any inverter that has at least one such measure, combined with the Ziehl device?

The rest of the document describes other parameters, such as what to do when the grid recovers, how long you have to wait before you may reconnect. This could also rule out some inverters (so if your inverter reconnects in 30 seconds... then on that technicality they could throw it out).

I think the Multiplus passes (and should be on the list) simply because it has an active detection component (it is legal for grid connection in the UK for example).

Anyway... that is my take on it. Would love to know why the Infini-V failed.

[___]

There you go, it's changed sides!

[Mark]

16 minutes ago, plonkster said:

There you go, it's changed sides!

@plonkster What software are you running please ...

[SilverNodashi]

15 minutes ago, plonkster said:

There you go, it's changed sides!

Do you think, that since the BMV only has one current sensor, and can only monitor 2 voltage points, that it's calculations could be wrong? On a 24-battery bank, it wouldn't be able to monitor batteries 2,3,4,5,6,7,8,9,10,11,13,14,15,16,17,18,19,20,21,22 and 23 - thus relying on the values of batteries 1,12 and 24. 

Same applies to a 4-battery bank, 8-battery bank, etc. 

Nor can it monitor each cell individually, so this begs the question: how accurate are these values, really?

[Mark]

19 minutes ago, plonkster said:

There you go, it's changed sides!

Another Q:  When it flipped sides... was the voltage and deviation exactly the same...  maybe the software is flipping and not the actual bank?

[SilverNodashi]

1 hour ago, plonkster said:

Oh that is nothing. It goes over 3% for a short period. Hey, it's Ex-UPS batteries that I've had for 18 months now. When they were new they were only rated maybe 700 cycles to 50%.... :-)

how many cycles have you done so far?

[___]

2 minutes ago, SilverNodashi said:

that it's calculations could be wrong?

Could be, but unlikely. It has a second input that can be used to monitor either 1) the starter battery on a boat/RV, 2) monitor the battery temperature, or 3) measure the midpoint voltage. So it is a generic voltage sense input. The top and bottom values are not calculated on the CCGX, they are in fact calculated by the BMV itself. So if there is a bug here, it would be in the firmware of the BMV.

I'm pretty sure there is no bug though, because I've induced an imbalance before by connecting a 3A automotive lamp to the high side to pull it down. The BMV correctly showed the top side being pulled down, and when I disconnected the lamp it showed it creeping up again. So no... pretty sure it's working as expected.

5 minutes ago, SilverNodashi said:

how accurate are these values, really?

The multimeter agrees. Or did last time I checked. Of course I don't know what's going on on an inter-cell level, but I think just about everyone on this forum has that problem too!

5 minutes ago, Mark said:

When it flipped sides... was the voltage and deviation exactly the same

I wasn't watching it that closely. I don't know. Previous tests says there is nothing weird going on... it actually flipped sides.

Clearly the charging impedance vs SoC curve isn't 1) linear nor 2) the same for two batteries in this case.

[___]

19 minutes ago, SilverNodashi said:

how many cycles have you done so far?

Sadly, I do not know. They have been cycled very shallowly, deepest was probably around 35%. According to the spec sheet I might get 1500 cycles out of them. Life time is maybe 4 years at the temperatures they generally live at, and I'm now approaching 3 years total age, so I suspect the cycles doesn't matter so much as the general age at this point. If they can last until December... maybe we can put in a Lithium bank this time!

[SilverNodashi]

23 minutes ago, plonkster said:

Could be, but unlikely. It has a second input that can be used to monitor either 1) the starter battery on a boat/RV, 2) monitor the battery temperature, or 3) measure the midpoint voltage. So it is a generic voltage sense input. The top and bottom values are not calculated on the CCGX, they are in fact calculated by the BMV itself. So if there is a bug here, it would be in the firmware of the BMV.

 

How many batteries do you have?

23 minutes ago, plonkster said:

 

I'm pretty sure there is no bug though, because I've induced an imbalance before by connecting a 3A automotive lamp to the high side to pull it down. The BMV correctly showed the top side being pulled down, and when I disconnected the lamp it showed it creeping up again. So no... pretty sure it's working as expected.

The multimeter agrees. Or did last time I checked. Of course I don't know what's going on on an inter-cell level, but I think just about everyone on this forum has that problem too!

 

I'm thinking more along the lines of: there aren't sufficient monitoring points in the system as a whole. I would personally have wanted to measure every battery's volts and current flow. With more measurement points, one would probably see different data from each battery - but that would mean a whole new design on the BMV side

And, yes, then everyone would be sitting with the same problem. 

 

 

 

[___]

1 hour ago, SilverNodashi said:

How many batteries do you have?

Two 12V batteries. So I have one measuring point... the rest are covered under the plastic :-)

 

1 hour ago, SilverNodashi said:

there aren't sufficient monitoring points in the system as a whole

True.

I'm now at 0.2V difference. I'll see if there is any improvement but it seems this is it.

[___]

2 hours ago, Mark said:

What software are you running please

That's Venus, the official CCGX firmware. It's from an actual CCGX. It has a really cool VNC/Web access feature, so I simply screen-grab the remote view to make these pictures.

There is a version for the Raspberry Pi and it is free. It is not crippled and it has all the official software on it that you would get on a real CCGX. With the exception of the canbus devices, this thing can do everything that the CCGX can (just with no LCD screen). And there is no official support, only informal support on the venus mailing list. If you want official support you should buy the Venus-GX (the hardware was delivered last week).

I spent 6 months getting it where it is. Well, not solidly, but many evenings over the course of 6 months... please try it! :-)

[Deonsr]

What does it do and where do I find it. Currently I am running Watchpower on a Nextbook on Windows 10.

 

[Guest]

Do you have a axpert?

Sent from my SM-G920F using Tapatalk

[Guest]

Here is the website

http://iccsoftware.co.za/

Sent from my SM-G920F using Tapatalk

[Guest]

After you download ICC for windows you can email me the machine id in the settings page . I will then send you a trail key

Sent from my SM-G920F using Tapatalk

[Deonsr]

Yes, 2 x Axpert 4,000w 5,000VA in Paralel with 16 x 170 A/h lead crystal batteries and 16 x 250 w PV's

[Deonsr]

Where do I email it to or do you mean I should post in on the forum?

 

[Guest]

Email it to me.

Sent from my SM-G920F using Tapatalk

[Deonsr]

I do not have your email adress.

[Chris Hobson]

9 minutes ago, Deonsr said:

I do not have your email adress.

[email protected] is the email address associated with ICC.

[pilotfish]

On 2017/03/27 at 6:35 AM, Chris-R said:

Hi Pilotfish

I am sending you a PM with all details required for the HA-02 order

Welcome to the forum, I am sure you will enjoy it !

Cheers

Chris-R

Hi Chris,

My 2x HA02's arrived this morning in good order, thanks for the prompt and efficient service from both you and Zane'

[___]

So I absorb-charged this bank yesterday until the charge current dropped to 0.5% of the capacity (and the Multi finally decided that's good enough to go to float)... and look at the beauty I just witnessed. Yeah I know... doesn't mean everything is in sync internally, but this certainly looks much better.