So... as you can see I've filled up my first bank of Pylon 2.4s... and I've started investing in a second bank of Pylon 3.5s.

I will parallel the two banks into my Mersen Battery Disconnect gear, and will use the first 3.5 as the master battery to drive the charge/discharge/balancing of the whole thing.

Any better ideas/comments are welcome.

I would guess you need this to connect the stacks?

http://www.pylontech.com.cn/pro_detail.aspx?id=123&cid=23

LV-HUB+Spec.pdf

18 minutes ago, Mark said:

I would guess you need this to connect the stacks?

http://www.pylontech.com.cn/pro_detail.aspx?id=123&cid=23

LV-HUB+Spec.pdf 1.11 MB · 1 download

Interesting. Nope. Haven't considered that at all. Is it a requirement?

I don’t know what amperage your system is setup to charge at.

Mine with 4 batteries is setup to charge at 49a.

I used another battery cabinet setup where the batteries stand on its side (supplied by Full Circle Solar). One of the unintended consequences of this was that I had to modify the battery cables (I was new to the game). It worked out that the batteries were connected as two sets to the isolator. When measuring the amps with my clamp meter I found that the amps were only 24a in the battery cables as opposed to 49a where it connected to the inverter. This was more acceptable to me. 

With any modification in future I would try to minimize the cable amps in this way.

29 minutes ago, PaulinNorthcliff said:

Interesting. Nope. Haven't considered that at all. Is it a requirement?

I don't know and don't have PT's (bucket list item ;)) BUT you are mixing 2000's and 3000's so I expect they need to be charged differently!

1 hour ago, PaulinNorthcliff said:

... investing ...

My 2 cents. Not sure how "investment" and "batteries" can be used in the same sentence. 😜

5 minutes ago, Mark said:

I don't know and don't have PT's (bucket list item ;)) BUT you are mixing 2000's and 3000's so I expect they need to be charged differently!

I don't know. I have been told 'not' by someone who should know. Apparently the BMS in each battery will sort itself out???

Edited May 3, 20197 yr by PaulinNorthcliff

1 minute ago, The Terrible Triplett said:

My 2 cents. Not sure how "investment" and "batteries" can be used in the same sentence. 😜

Ja, ja... 😂

13 minutes ago, PaulinNorthcliff said:

I don't know. I have been told 'not' by someone who should know. Apparently the BMS in each battery will sort itself out???

O, I'm so waiting to hear if this can be done, mixing them, as it opens a lot of doors in my mind.

11 minutes ago, The Terrible Triplett said:

My 2 cents. Not sure how "investment" and "batteries" can be used in the same sentence. 😜

In the same way you invest in a car. Everyone knows it isn't really an investment... but you still need a car 🙂

16 minutes ago, Mark said:

mixing 2000's and 3000's

Not a problem. LFP has a much better impedance response to filling up. When any one cell fills up it goes high impedance and the other cells can only charge at the amount of current the BMS can pass around. The entire bus is held at a constant voltage and the BMS plays musical coulombs until it's all done. It can take a while... 🙂

22 minutes ago, Johandup said:

Mine with 4 batteries is setup to charge at 49a.

25A per rack for the 2kwh units, 35A for the larger ones. Roughly. With 4 batteries I'd expect you can charge at 100A...

Basically, Pylontech goes with C/2 charge and discharge rates. 2500Wh divided by roughly 50V divided by 2 is 25A. Doing the same with 3500 yields 35A.

Ideally you should listen to the BMS (if you have the CAN comms) and reduce your charge current when it asks you to. That makes balancing a bit more effective. The better batteries will even adjust the charge voltage to accommodate a high cell, but it's pretty much only FreedomWon, Sony/Muratta and Victron's HE batteries that do this properly. The rest goes with a static charge voltage and attempts to use current control... which sucks to put it mildly.

Just now, plonkster said:

musical coulombs

Musical @Coulombs?

😆

Just now, Elbow said:

Musical @Coulombs?

😆

Not those kind... I meant ampere-seconds 🙂

14 minutes ago, plonkster said:

Ideally you should listen to the BMS (if you have the CAN comms) and reduce your charge current when it asks you to. That makes balancing a bit more effective. The better batteries will even adjust the charge voltage to accommodate a high cell, but it's pretty much only FreedomWon, Sony/Muratta and Victron's HE batteries that do this properly. The rest goes with a static charge voltage and attempts to use current control... which sucks to put it mildly.

OK, so (thanks for the other reassuring answers, BTW), if I monitor my charging via my Goodwe it shows that the charge voltage is NOT static, but constantly changes. Is this a good or bad thing?

24 minutes ago, PaulinNorthcliff said:

charge voltage is NOT static

By charge voltage I mean the voltage the BMS asks for via the CAN-bus. For a Pylontech that voltage is a hardcoded 53.2V. In such systems, the solar chargers and the inverter/chargers always ATTEMPT to get it up to that voltage. It may of course not be able to. So what I mean by charge voltage is the TARGET that it aims for.

Some batteries (BYD) even ask for 56.5V, or 58V (LG). Then when a high cell condition develops, they attempt to stop charging by asking the charger(s) to reduce output. My argument is simply that you get much smoother operation (less saw-tooth) by lowering the charge voltage to accommodate the high cells.

2 hours ago, Elbow said:

Musical @Coulombs?

😆

A little respect for the name, please  🧐

(I've been waiting for an opportunity to use that smiley 🙂)

3 hours ago, plonkster said:

By charge voltage I mean the voltage the BMS asks for via the CAN-bus. For a Pylontech that voltage is a hardcoded 53.2V. In such systems, the solar chargers and the inverter/chargers always ATTEMPT to get it up to that voltage. It may of course not be able to. So what I mean by charge voltage is the TARGET that it aims for.

Some batteries (BYD) even ask for 56.5V, or 58V (LG). Then when a high cell condition develops, they attempt to stop charging by asking the charger(s) to reduce output. My argument is simply that you get much smoother operation (less saw-tooth) by lowering the charge voltage to accommodate the high cells.

👌

@PaulinNorthcliff

Combining US2000 and US3000 in parallel is not a real problem, as chemistry and voltages of these two are exactly the same.

But as far as I know, connecting US2000 and US3000 together, using data link cables, has not been tested by Pylontech.
I saw the best practice document on topic of how to combine Phantom S together with US2000, and which type of the battery should be the master. But I never saw a document on whether it's okay to interconnect US2000 and US3000 with data cables.

How about to ask Pylontech techsupport first?

12 hours ago, Youda said:

@PaulinNorthcliff

Combining US2000 and US3000 in parallel is not a real problem, as chemistry and voltages of these two are exactly the same.

But as far as I know, connecting US2000 and US3000 together, using data link cables, has not been tested by Pylontech.
I saw the best practice document on topic of how to combine Phantom S together with US2000, and which type of the battery should be the master. But I never saw a document on whether it's okay to interconnect US2000 and US3000 with data cables.

How about to ask Pylontech techsupport first?

That would be like 'reading the manual'... where's the fun in that?

I spoke to the supplier of the batteries and he advised me to follow this route.

I'll let you know if the smoke escapes.

Here's a basic "mixture" guideline from Pylontech.

US2000B US2000 PLUS US3000 Difference and Mixture Installation.pdf

16 hours ago, Nicholas said:

Here's a basic "mixture" guideline from Pylontech.

US2000B US2000 PLUS US3000 Difference and Mixture Installation.pdf 410.32 kB · 11 downloads

Yep!

That's how it was explained to me, and how I have implemented the installation.

It's gonna take weeks to balance properly, but this morning it's looking good.

I used my 2 pole Mersen battery disconnect as the busbar to parallel the two battery banks and all is good.

My remaining 'issue' is now how to tell the Goodwe ES5048 about the new domestic arrangements. In the battery settings Goodwe only allows for 4xUS2000+. I haven't as yet updated the software of the Goodwe (still struggling to get it to talk to my WiFi router).

BTW, Nicholas, thanks for the doccie from Pylon. Always nice to have the official stamp of approval.

 

Still need to tidy up the battery cables. I mounted the battery disconnect to the side of the battery cabinets for ease of access (if the smoke starts escaping). You can see where I have grafted the Pylon battery plugs onto the cables.

On 2019/05/06 at 8:35 AM, PaulinNorthcliff said:

BTW, Nicholas, thanks for the doccie from Pylon. Always nice to have the official stamp of approval.

 

Hi Everyone. I'm considering adding 1x US3000 to my 2x US2000 batteries. Per the document shared by @Nicholas, I would use the 3000 as the master. Management will be done by the Pylontech BMS via the right cable to my Axpert King inverter.

Noticing the experimentation above, has mixing the 2K and 3K ranges been working for you all in the long run? Anything I should look out for?

Yep. Almost a year on and it's working just fine. Will add a second 3.5 shortly, but I've yet to discover that I run short during the night... if SWMBO can be convinced NOT to put on the dishwasher at 17h30 (2 and a half hour wash cycle).

Edited February 21, 20206 yr by PaulinNorthcliff

On 2020/02/21 at 7:49 AM, PaulinNorthcliff said:

Yep. Almost a year on and it's working just fine. Will add a second 3.5 shortly, but I've yet to discover that I run short during the night... if SWMBO can be convinced NOT to put on the dishwasher at 17h30 (2 and a half hour wash cycle).

Thanks Paul. What inverter do you use and do you rely on third party management software?

My Axpert King with the right cable is allowing the Pylontech BMS to manage my 2x US2000 batteries very well. Would want to stick to the Pylontech BMS when adding the 1x US3000 as the new master this week.

Do you find that your batteries deplete at the same rate in percentage of capacity, thus the stack as a whole, or does your 3000 run longer than your 2000's, thus batteries depleting individually at the same absolute rate?