Setting up 8 Pylontech US5000 for first time...one with highest voltage shows lowest SOC

Question

question

alaskannoob asked Sep 20, '23

Setting up 8 Pylontech US5000 for first time...one with highest voltage shows lowest SOC

I got my Quattro mounted but not powered, same with my RS MPPT, and my bus bar is set up with an EG4 chargeverter. All 8 of my US5000 batteries were initially showing 49.4V with the voltmeter.

I used the EG4 to charge them all up to 51.8V. The first one I charged by itself. The SOC lights were 4 solid, 1 blinking, and the last one never lit up at all.

I then used the EG4 to charge the other batteries up in a group. They all showed 5 solid SOC lights, and the final one blinking.

I thought this was odd.

The next day I put the voltmeter on them and the one with the lowest SOC indication actually was showing the highest voltage.

Battery 4 (the one with lowest SOC light indications) was 51.3V

Battery 3 was 50.6V

Battery 2 was 50.9V

Battery 1 was 51.1V

Battery 5 was 51.2V

Battery was 51.0V

Battery was 50.9V

I then charged them each up again with the EG4 setting 51.8V and 10A of current.

Then the next day I measured:

Battery 1 = 51.6V

Battery 2 = 51.6V

Battery 3 = 51.4V

Battery 4 (the one with lowest SOC light indication) = 51.6V

Battery 5 = 51.6V

Battery 6 = 51.5V

Battery 7 = 51.6V

Battery 8 = 51.6V

Anybody else seen something like this with one battery showing less SOC on the lights than the others at same or higher voltage?

I know the batteries have to balance and I haven't put a load on them yet and haven't yet powered up the MPPT or Quattro. Although I don't know if balancing is limited to each battery balancing its own cells, or if the group balances batteries (sharing charge or discharge amongst themselves). I have my batteries each wired individually into the bus bar and am not using any other positive or negative short cables to connect batteries to each other. So if the master battery is somehow providing extra charge/discharge to various batteries I'm not sure how it would be doing that.

Perhaps a battery that isn't full shows as a load on the bus bar, and another battery that is more full would then supply power to the bus bar and so the batteries would balance out that way? I have no idea.

I should have the MPPT online with about 1.5KW of panels in the next couple of days so maybe after setting up the Cerbo GX to DVCC and getting solar to charge, things will work out?

Pylontech

1

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

alaskannoob commented · Sep 20, 2023 at 03:26 PM

My dealer suggested I discharge Battery 4 completely, and charge it back up. I don't have a 48V DC load that can do this at the moment though, but I did buy a 48V DC heater and should get it shortly.

0 ·

Modifications Space

Node Red Space

Answer 1 · 2023-09-20T15:30:26Z

alaskannoob answered · Sep 20, 2023 at 03:30 PM

I should have the MPPT and 1.5KW of panels ready to go later today, so before I discharge Battery 4, I think I'll just see what happens with the MPPT charging in DVCC and the batteries (perhaps) balancing themselves and such. Maybe this will sort itself out.

I haven't put any loads on the batteries, so the only load will be the MPPT itself when I finally flip it on.

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 2 · 2023-09-20T15:53:30Z

Alexandra answered · Sep 20, 2023 at 03:53 PM

@AlaskanNoob

TLDR leave them charged connected to each other so they can balance as individuals and as a pack.

The Pylontecs sort themselves out. You can have a whole stack at different state of charge.

They do need a good charge when new (this can be done altogether) and be left at 99/100% for a few days to get things balanced on the initial charge particularly.

You can get a higher pack voltage than soc if some cells are higher than others (so not fully charged but refusing charge for balancing purposes)

Maybe charge them up a bit slower as well.

6

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Craig Chamberlain commented · Sep 20, 2023 at 04:15 PM

This is my experience as well. As long as you start with all batteries powered down and then connect them all up including the BMS connection and then perform the normal start procedure from the master, the batteries will soft-start and balance out over the next few days.

I've only got three US5000s but when I added the third one I just waited until the voltage on the existing pair was in the vicinity of the new one I was adding (49.4V as I recall) and then I powered everything down, connected the new module as the new master and powered it all back up. I then ensured I ran for extended periods at 100% SOC over the next few days to assist the cell balancing process and I did get one cell imbalance warning about 3 days in, but this appeared to just be a transient issue and I've not seen any warnings since and cell balance is very tight.

HTH

1 ·

alaskannoob commented · Sep 20, 2023 at 07:58 PM

Okay, thank you and Craig for the information. I have not previously powered them all on and left them on, I have always turned them off after charging.

I just measured them again:

Battery 1 = 51.6

Battery 2 = 51.6

Battery 3 = 51.4

Battery 4 = 51.6

Battery 5 = 51.5

Battery 6 = 51.6

Battery 7 = 51.6

Battery 8 = 51.6

And I powered them all on, connected to the bus with no loads on it. For 30 seconds to a minute, Battery 1 and Battery 8 were discharging, and Battery 3 and 5 were charging.

So it seems two of the 51.6 batteries were charging the two lowest batteries. But then after 30 seconds to a minute, all the batteries went to idle which is where they remain now. I'm not sure if I should leave them all on in idle right now even though nothing is charging or discharging.

0 ·

rack.jpg (441.6 KiB)

Craig Chamberlain alaskannoob commented · Sep 20, 2023 at 08:56 PM

It doesn't seem unexpected that the lower voltage modules might be charged by the higher voltage ones so I'd say that was normal module balancing. If you have a clamp meter, it might be interesting to measure the currents flowing in the various wires as there may still be some currents flowing to further balance the modules. Perhaps the BMS software is smart enough to accept module voltages once they get within a certain threshold of each other and then go idle until a charge or load appears. I'm not sure exactly how the BMW operates.

But it's really once you get the batteries up to near 100% that the "cell" balancing happens. IIRC there are 15 cells in each module and the sum of all the cell voltages gives you the module voltage. Right now my modules are at 49.14V and the min and max cell voltage across the 3 modules is 3.28V. The min and max isn't always identical with 0.01 - 0.03V difference being a typical range once balanced in my experience at least. Right now you'll probably have a wider range of cell voltages but they will coalesce over time.

If I was you, I would leave them all connected together on the bus bar and get the Cerbo connected up so you can easily monitor the cell imbalance and any alerts from the batteries. Just take care to get everything set up correctly, especially the max charge voltage and other key params from the Victron document here: https://www.victronenergy.com/live/battery_compatibility:pylontech_phantom

(Side note: Are you aware of the need for an external capacitor on some Cerbo GX devices on 48V systems? If your Cerbo is a revised model then you'll be fine but best to double check if unsure. More details on serial number ranges to check on the forum.)

Once the Cerbo is up and running then yeah I'd probably try to get the MPPT online and set with a relatively low charge current to help the cells balance. Again, take care with the setup of the MPPT to ensure you set appropriate params and get help from your dealer or on the forum here if need be. I got my own MPPTs set up nearly spot on but my dealer noticed one param that wasn't quite right. It's always good to get a 2nd pair of eyes to look things over.

HTH

ps. I'm really jealous of your 8 modules! :)

1 ·

alaskannoob Craig Chamberlain commented · Sep 20, 2023 at 10:02 PM

Thanks Craig. I think I'll put the clamp on some wires and see if there is anything. The batteries are still just blinking in idle but who knows. At some point I'll measure their voltages and maybe I'll see that a low battery has increased in voltage.

I just wired my PV into my DC isolator and checked that it was getting voltage, so I'm now going over wiring the PV wires into my MPPT RS 450/200 which is the part that makes me the most nervous since the connections are those plastic things rather than nice metal lugs. Fingers crossed I'll get the PV in there in the next hour or so.

Then I'll turn off and disconnect all my Pylontech. Then use a secondary smaller 48V battery to charge the capacitor of the MPPT, then disconnect that smaller battery and turn the Pylontech all on.

Then I'll power up the Cerbo GX.

Then I'll get to programming. I think I'm just setting DVCC really and Victron will override the BMS requested charge voltage on its own to 52.4V or something like that.

I think there are a couple of other steps in that Victron Pylontech compatibility guide though so I'll go through it again.

Then I'll turn the PV isolator to ON and let the sun power flow and see what happens. No loads other than the Cerbo and the MPPT itself, and just let it run like that for a week or so.

The 8 US5000s are nice and I'm very happy that I haven't seen any alarms yet. But we're so far off grid in Alaska that it was no fun getting them up here. So heavy!

0 ·

alaskannoob Craig Chamberlain commented · Sep 20, 2023 at 10:10 PM

I so far have not grounded anything. Since I'm off grid, and I'm not using AC power yet, I think I can get away with a "floating" system where nothing is grounded. But I don't actually know. I don't have the batteries or the MPPT or the DC isolator or the solar panels grounded. So I'm not sure if I need to do that even during my temporary testing.

The electrician will come out when we're ready for the AC output from the inverter and will make sure everything is grounded for us, but in the meantime...???

0 ·

alaskannoob Craig Chamberlain commented · Sep 20, 2023 at 10:23 PM

Oh, we did check the Cerbo GX for the capacitor issue and it was not one of the effected units. We turned it on successfully to program several devices about a half dozen times. Then it would no longer power up. We got another one and I tried the new power cable to the old one, but it wouldn't power up.

The new one also has a good serial number and so far is working but it's only been powered up once...

0 ·

Answer 3 · 2023-09-21T00:05:23Z

alaskannoob answered · Sep 21, 2023 at 12:05 AM

Powered up the Cerbo and got this info.

bat-1.jpg (24.2 KiB)

bat-2.jpg (32.4 KiB)

bat-3.jpg (37.4 KiB)

bat-4.jpg (78.9 KiB)

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 4 · 2023-09-21T01:39:44Z

alaskannoob answered · Sep 21, 2023 at 01:39 AM

And after cracking one of my trackers in my MPPT RS using the 2.4Nm torque setting in the printed OLD manual (which has been removed in the NEW manual), I wired up our 3 solar panels into a different tracker.

And powered it all up for a first attempt in the rain and fog...

Amazingly enough, no sparks, no smoke, no errors and even with only 3 panels temporarily set up, we are getting a little something and batteries seem to be charging. Most we've seen in the hour or so of charging is 450W or so, but not bad for late in the evening on a foggy rainy day. That's more power than we've ever made with our little Renogy Panels and Goal Zero on even the sunniest blue sky day.

DVCC is on and everything looks fine to me perusing the settings.

cracked.jpg (245.3 KiB)

fog.jpg (163.7 KiB)

latest-battery-status.jpg (41.6 KiB)

settings-1.jpg (29.1 KiB)

settings-2.jpg (32.2 KiB)

settings-3.jpg (27.8 KiB)

settings-5.jpg (32.5 KiB)

settings-6.jpg (30.5 KiB)

settings-7.jpg (30.2 KiB)

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 5 · 2023-09-21T15:06:12Z

Craig Chamberlain answered · Sep 21, 2023 at 03:06 PM

Good to see your progress above although I am gutted for you at the cracked tracker connection. 2.4NM is a pretty common torque settings here in the UK for the larger connections inside a typical consumer unit along with 2NM on neutral and earth bus bars and 1.2NM on RCBO terminals. I also have a torque screwdriver which I use for everything these days and would be really annoyed if something broke while using it. I think the AC connections inside the Multiplus-II are something like 1.6NM off the top of my head but I always refer back to the manual any time I torque them up. I really should print little labels to put next to the terminals as should have been done at the factory IMHO.

As for your cell balance, it already looks pretty good already.

I can't speak authoritatively about grounding as it varies by regional wiring codes and of course you are off grid which I'm not. In my case I have a MET (main earth terminal) as part of my AC feed from my power company and this has been supplemented as required by the UK wiring code for use in island mode (essentially off grid when grid power has failed). So all my AC earths come back to this earth point along with the earth terminals from my Pylontech batteries. I have not earthed my MPPTs or my PV array as there are different opinions on what to do in that area. I am in a very low risk area for lightning so don't feel the need for DC surge protection or array earthing. If I did earth my array then I would need a completely separate earth rod for this purpose and that's actually very hard to do on a metal structure and metal skinned garage where the panels are mounted. I do have AC surge protection because my AC feed is from overhead 11kV lines with a dedicated transformer on a pole outside our house. This is at greater risk of being hit directly or indirectly by lightning and so AC surge protection is important. No point in frying my inverter just as the grid fails...

We've actually had 3 or 4 grid failures in the last year and have barely noticed. We only actually notice when we realise that our range cooker isn't working or our EV charger isn't working because those loads are not maintained in island mode. But apart from that life goes on as normal until the batteries die, and since the batteries are DC coupled to the solar array, in daytime we can keep going from solar power + battery. It's really a great system.

Good luck with your next steps.

2

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

alaskannoob commented · Sep 21, 2023 at 03:30 PM

Grounding and AC power are both magic to me. Might as well try to learn quantum physics. I know others get it, but I'm dumber than most.

Grid failures seem like something we'll see more and more. Good on you for preparing for them, invaluable investment in your own security. Thanks for the help, cheers!

0 ·

Craig Chamberlain alaskannoob commented · Sep 21, 2023 at 04:10 PM

You may be more modest than most but I highly doubt you are dumber than most or you wouldn't be doing what you're doing! :)

Grounding does have some complexities for sure but much of it is common sense really. You need to be clear on what type of event you are trying to mitigate and then try to ensure you don't cause a side-effect that is worse than the potential event itself. For me that means AC surge protection is a no brainer but DC is more difficult to justify. That said I continue to monitor the industry best practices and the regulations and will upgrade my system if I feel it is justified in future.

Re grid failures, to be fair, in central Scotland where I live they are fairly rare in most populated areas but we live out in the countryside of North Ayrshire in a cottage where the overhead HV lines that serve most properties like ours are a bit vulnerable to lightning, trees falling down and ice loading etc. Our power company is very responsive when the grid does have a fault but it can still take a few hours to get the fault fixed. Having battery and solar backup is great in such situations, especially as I work from home as an software engineer full time.

The worst outage we had was a couple of years back when the supply cable from the transformer to our house failed underground and the power company had to come and dig a big trench in our driveway to find and repair the cable. But even in that case we were back on the grid within 36 hours or so. But yeah, I agree that it's good to have some degree of energy security ourselves because it's pretty clear that energy security in the world at large, especially in Europe right now, is less robust than it once was...

All the best.

1 ·

question