diy solar

diy solar

Darfon Gro Watt GBLI 5001 Battery Pack- worth a punt?

That's interesting. It appears to be working now! I have put the current clamp I have on the main cable wired into 1 and 8 on the LAN cable. As the sun shone more the unit said I was importing from the grid so I reversed the cables and now it is currently charging the battery, supplying the house and exporting to the grid. I don't have a significant "phantom load" so this strikes me as correct. Weird that yours is wired to 2 and 8. You really shouldn't have opened it up for me but thank you. Suppose I need to see how it performs... Might also need to wire the current clamp up in a more permanent way.... Also would like to look at the force discharge as that strikes me as useful.

Thank you for your help!
 
I strongly suspect that on the PCB 1-4 are all joined and 4-8 are all joined. This could easily be tested with an SP2000 unplugged and a multi-meter

As I advised Jad_00 to test a clamp if you're importing, simply reverse the cables on the clamp itself, that will trick the SP2000 into thinking it's exporting - the display should change to pointing towards the Grid - and should charge the battery after a few minutes
 
joshcliffejones said:
Oh nice work! The ideal set up is clearly to get the original GBLI5001 BMS working as it was designed and paired with the SP2000! That was my initial plan when i got my first Growatt battery, but even after i charged and balance all cells, the BMS still wouldnt play ball.

Now i have two batteries and am just converting them to use Daly BMS's, then set up the SP2000 as LEAD...not ideal, but a damn sight cheaper than alternative options!

With your mention of only being able to discharge 50% when the SP20000 is configured for LEAD, this is 50% of the capacity based only on the battery voltage the SP2000 reports as there is no formal communication between SP2000 and battery.

This would mean that "50%" of a lead acid battery would be a rough voltage of 48.2v (based on 4 x 12v batteries at 12.07v each at 50%) which when compared to the typical voltage curve of a lithium ion battery would equal roughly an 80% discharge (48.2v total devided by 14 cell packs = 3.44 per cell).

View attachment 110599

I appreciate there is more to the science than what i mention above and it should all be taken with a pinch of salt, just my two cents.

More information about the cells used within the GBLI5001 batteries can be found here https://www.tme.com/Document/87b5645f730e56cad507df13706ef5b9/Swing5300.pdf
Click to expand...
I would love to find out if there is anyone out there who has managed to unlock what I believe is the lockout on the BMS for the gbli5001 battery it comes up as error code 024 which once this has happened I haven't had any chance of getting the BMS board working again hence I have 2 spare locked out boards....
There must be someone out there that can fix this
 
mike_hug said:
Interesting... also strangley my SP3000 says the battery is down to about 30% after discharge - so not sure how that has happened. I did tinker with some of the voltage settings in the software and lowered them so that may explain it

One of my SP3000's had a half bad display - otherwise fully works - the right side of the display text is missing. I did a warranty claim on it directly with Growatt, and they have sent me a new SP3000! No mention of wanting the old one back... so now I have 4!

I have another GBLI battery but 2 of the cell packs are bad. Leaked. The batteries are spot welded to a back plane within the cell pack. I did consider buying the individual cells off ebay, but the welding business is a bit of an issue
Click to expand...
Hi Mike I also have toyed with the idea of replacing some of the cells in the individual packs I have got 20 brand new Boston swing cells which I ordered from China haven't managed to weld them in yet but I'm also considering adding two extra cells to each one of the 14 battery packs as there are probably as you know 2 empty cell holders per pack which in total takes 20 and not the 18 cells as original this idea is purely to augment /boost the ageing battery
I was also toying with the idea of getting a proper battery load discharge test machine there are many videos on YouTube on how to use one this it immediately highlights a faulty or depleting battery pack
I also bought a roll of tin mesh (4 hole width fits well) and some tin tabs to spot weld back on to the new cells as well these are easily available on eBay.
Incidentally can you tell me what is the difference between the sp-2000 and the spf3000 apart from the obvious power difference I did hear that it's also something to do with the 3000 is more AC biased?
 
Tad said:
Hi Mike I also have toyed with the idea of replacing some of the cells in the individual packs I have got 20 brand new Boston swing cells which I ordered from China haven't managed to weld them in yet but I'm also considering adding two extra cells to each one of the 14 battery packs as there are probably as you know 2 empty cell holders per pack which in total takes 20 and not the 18 cells as original this idea is purely to augment /boost the ageing battery
I was also toying with the idea of getting a proper battery load discharge test machine there are many videos on YouTube on how to use one this it immediately highlights a faulty or depleting battery pack
I also bought a roll of tin mesh (4 hole width fits well) and some tin tabs to spot weld back on to the new cells as well these are easily available on eBay.
Incidentally can you tell me what is the difference between the sp-2000 and the spf3000 apart from the obvious power difference I did hear that it's also something to do with the 3000 is more AC biased?
Click to expand...
I'd be interested how that goes and I like the idea of the extra cell, but that's an awful lot of work as you can't just take apart each cell pack, it would require breaking all the welds. I have managed to get the backplane away from a few of the cells and my intention was to try and reuse the existing back plane and re-weld new batteries back onto it

The SP3000 is very nice in that it can charge off AC as well as solar. So I charge my batteries during off-peak during the night using Octopus Go @ 7.5p. Charge times are customisable. Then I have them set up to Force Discharge during the day when I'm importing

Other news - I had emails from the guy who visited me a few times from Growatt years ago. He had the software that allows you to unlock the Darfong BMS as well as monitor all the cells. I saw it running, but he wouldn't give me a copy back then. He left Growatt a few years ago - can't say I blame him - he probably had a busted knuckle after tapping the SP2000 screens so often and a bad back after having to carry so many broken GBLI batteries... Anyway, I have found where he is working now and have sent an email. Hoping he will reply so I can ask him if he will send me a copy as I'm sure he will still have it

Anyone point me to how to "sniff" RS422 comms? I have a working GBLI with original board inside. I'd like to see what it's saying. Then we could possibly use a Pi or something similar to con the SP2/3 into thinking it's connected to an unmodded GBLIU battery and get 80% DoD
 
mike_hug said:
I'd be interested how that goes and I like the idea of the extra cell, but that's an awful lot of work as you can't just take apart each cell pack, it would require breaking all the welds. I have managed to get the backplane away from a few of the cells and my intention was to try and reuse the existing back plane and re-weld new batteries back onto it

The SP3000 is very nice in that it can charge off AC as well as solar. So I charge my batteries during off-peak during the night using Octopus Go @ 7.5p. Charge times are customisable. Then I have them set up to Force Discharge during the day when I'm importing

Other news - I had emails from the guy who visited me a few times from Growatt years ago. He had the software that allows you to unlock the Darfong BMS as well as monitor all the cells. I saw it running, but he wouldn't give me a copy back then. He left Growatt a few years ago - can't say I blame him - he probably had a busted knuckle after tapping the SP2000 screens so often and a bad back after having to carry so many broken GBLI batteries... Anyway, I have found where he is working now and have sent an email. Hoping he will reply so I can ask him if he will send me a copy as I'm sure he will still have it

Anyone point me to how to "sniff" RS422 comms? I have a working GBLI with original board inside. I'd like to see what it's saying. Then we could possibly use a Pi or something similar to con the SP2/3 into thinking it's connected to an unmodded GBLIU battery and get 80% DoD
Click to expand...
That's great haha please let me know if you have any joy with the bms unlock app?
 
Tad said:
That's great haha please let me know if you have any joy with the bms unlock app?
Click to expand...
Was also hoping to squeeze the extra two cells in to the battery packs without having to take off .the whole of the back plane copper and tin
 
Also Mike Im liking the idea of being able to charge the battery with ac power as the sp3000 if you've got a good rate from your utility supplier
I'm thinking along the lines of a wind turbine feeding into my spare third input on the controller of the sp-2000 on a daytime timer for the winter months to help charge the battery problem is minimum voltage is 100 volts but not sure if that is on its own or in combination with the solar panel strings
 
Tad said:
Also Mike Im liking the idea of being able to charge the battery with ac power as the sp3000 if you've got a good rate from your utility supplier
I'm thinking along the lines of a wind turbine feeding into my spare third input on the controller of the sp-2000 on a daytime timer for the winter months to help charge the battery problem is minimum voltage is 100 volts but not sure if that is on its own or in combination with the solar panel strings
Click to expand...
You may need to ask someone more clued up about this - But all the 3 inputs on the SP2000 are the SAME string. They are NOT separate MPPTs. The SP3000 has 2 inputs which are separate so you can have 2 independant strings attached - ie East and West for example. Not sure what would happen if you had wind and solar together at the same time, but I suspect it *wouldn't* be good. But if the voltage is right and you can switch between solar and wind then - well - I *guess* it would be okay but I'm definately NO expert on wind. I'm good at producing it though!
 
Last edited:
  • Like
Reactions: Tad
Thanks for that I'm still quite a newbie to all this can anyone explain to me what mppt (maximum power Point tracking) in terms of solar panel configuration and in terms of battery charging does
 
For anyone interested this seller has packs for sale that are unwelded. Batteries will need to be re-configured at 3.6v (basically all the same way round) and then rewelded

eazyswitch on eBay

Follow eazyswitch on eBay. Buying, Selling, Collecting on eBay has never been more exciting!
www.ebay.co.uk
 
mike_hug said:
For anyone interested this seller has packs for sale that are unwelded. Batteries will need to be re-configured at 3.6v (basically all the same way round) and then rewelded

eazyswitch on eBay

Follow eazyswitch on eBay. Buying, Selling, Collecting on eBay has never been more exciting!
www.ebay.co.uk
Click to expand...
I have actually bought one of these from this seller (Tony)! I have a bad cell pack in one of my GBLI5001, so planned to rip it apart and either find and replace the bad cells and replace with good...or just throw a known good back together and swap it out.

I received the pack earlier this week and so far have only checked voltages of cells, but all appear good.
 
Yes nice find that Mike
I'm just trying to sort out my spare battery at the moment BMS says error code 04 which I think is battery imbalance i.e. the BMS board says flat battery in its memory but my spare battery has 50% charge so just trying to discharge this battery from 54 volts to around 50volts with splitting packs into 2 then using with 24 volt drill ... might be a while!
 
Tad said:
Yes nice find that Mike
I'm just trying to sort out my spare battery at the moment BMS says error code 04 which I think is battery imbalance i.e. the BMS board says flat battery in its memory but my spare battery has 50% charge so just trying to discharge this battery from 54 volts to around 50volts with splitting packs into 2 then using with 24 volt drill ... might be a while!
Click to expand...
From years ago, when I was in touch with the Growatt engineer:
"The BMS04 could be the start of a battery going faulty. This one will most likely need replacing."

Years later I discovered - to my horror - that the battery would never discharge, or even charge properly. I changed the BMS (to third party version) and it's working now

Purely for others reading this
"BMS 024 error is not good. It's the error code for cell imbalance >25mv. It's a battery swap for that one."
I think this is when the BMS is then hard-locked and is non recoverable. Again, swap to third party BMS
 
Hi Mike just a quick question if you were using the GBL i 5001 battery (design capacity at 4850w)
on the SP 3000 can you tell me what the wattage power that you are now able to draw from it I'm trying to find out if this power limiting is within the bms of the battery (ie 45 or 50amps) or within the sp2000 controller because I reckon if you're drawing more than 2000 watts using your sp3000 controller you may also want to increase the BMS amps from 50 to 75amps as for every 100 w of power drawn equates to 2.5 amps (with 48v battery) x 20 = 50amps but for the Sp3000.....2.5 amps x 30 =75 amps
I am very interested to find out where this limiting is either in the controller or in the battery BMS?
The design capacity of these gbli5001 batteries are approximate 100 amps so you might be able to get a bit more juice out of it and on the other hand maybe you should double the design capacity amperage to make it a good usable battery as well if using with the SP 3000 with a 75 amp BMS board to 150 amps or 7200w
It's all a question of where this limiting is?
 

Attachments

  • Screenshot_20220914-000926_Chrome.jpg
    Screenshot_20220914-000926_Chrome.jpg
    230.3 KB · Views: 9
Last edited:
Hi all , I've been reading this thread with great interest as i've just bought one of these (SP2000 + GBLii5001) and the red light is apparently solid on the battery (haven't picked it up yet). There's so much info in here I'm struggling to know where to start, Did you guys get round the 50% lead acid DOD issue ? With the CT clamp installed , will the SP2000 dish out the power to keep the grid close to zero ? Say panels are producing 3kw , house using 2kw , will it send 2kw to it's outputs and dump the extra 1kw into the battery ? I ask as i also got confused reading about the "forced discharge" you have to set ? . Apologies for the newby questions. Also , when I pick it up , whats the test sequence I should follow before deciding to give in and buy a 3rd party BMS ? Thanks in advance.
 
paulca said:
Hi all , I've been reading this thread with great interest as i've just bought one of these (SP2000 + GBLii5001) and the red light is apparently solid on the battery (haven't picked it up yet). There's so much info in here I'm struggling to know where to start, Did you guys get round the 50% lead acid DOD issue ? With the CT clamp installed , will the SP2000 dish out the power to keep the grid close to zero ? Say panels are producing 3kw , house using 2kw , will it send 2kw to it's outputs and dump the extra 1kw into the battery ? I ask as i also got confused reading about the "forced discharge" you have to set ? . Apologies for the newby questions. Also , when I pick it up , whats the test sequence I should follow before deciding to give in and buy a 3rd party BMS ? Thanks in advance.
Click to expand...
I haven't gotten around the 50% DoD issue. I think it's possible by either "sniffing" what is being said between the real BMS and the SP2000 RS422 and then emulating that somehow using a Raspberry Pi. Not easy. Or using a different battery inverter. Only other way would be Growatt re-writing the firmware but that isn't going to happen, or finding a way to re-write the firmware. I have various firmware for the SP2000's but it's too dangerous to attempt updating them

With the clamp installed, yes you have it mostly correct. Excess (export) is put into the battery. In standard setup mode the battery would not be released until there is ZERO solar coming into the SP2000. In summer that can be very late in the day

So you have Forced Discharge Mode, where the SP2000 will release the battery during the day when you would be importing. This is how most people would want it to work. It is not instant though and can take about a minute to react

Most SP2000 & GBLI5001's you see will have BMS 024 error displayed on the SP2000. If it has that, then the internal BMS is f**ked, and you need to get a third party BMS fitted. No which way about it

Growatt did change the firmware on later models so that it will charge the battery if the battery is getting too low (ie it will charge the battery when you're importing), which in theory will prevent the BMS 024 error. However since there are varying hardware revisions of the SP2000 not all the firmware is compatible, and if the wrong firmware/hardware revision is uploaded the SP2000 will be "bricked". I do have the 3 firmware revisions but would be very hesitant about giving this to anyone, having "bricked" two SP2000's in the past
 
Last edited:
I have also the same battery issues with the same error code, you guys have any success about it?:)
 
Michael-2022 said:
Hi Mike just a quick question if you were using the GBL i 5001 battery (design capacity at 4850w)
Click to expand...
Please read the whole post. It gives everything you need

Tad said:
on the SP 3000 can you tell me what the wattage power that you are now able to draw from it I'm trying to find out if this power limiting is within the bms of the battery (ie 45 or 50amps) or within the sp2000 controller because I reckon if you're drawing more than 2000 watts using your sp3000 controller you may also want to increase the BMS amps from 50 to 75amps as for every 100 w of power drawn equates to 2.5 amps (with 48v battery) x 20 = 50amps but for the Sp3000.....2.5 amps x 30 =75 amps
I am very interested to find out where this limiting is either in the controller or in the battery BMS?
The design capacity of these gbli5001 batteries are approximate 100 amps so you might be able to get a bit more juice out of it and on the other hand maybe you should double the design capacity amperage to make it a good usable battery as well if using with the SP 3000 with a 75 amp BMS board to 150 amps or 7200w
It's all a question of where this limiting is?
Click to expand...
I'm not sure I can answer that. You see I have 4 SP3000's all running. There is certainly a discharge limit and I hope to investigate this further, my house draws a lot of power, so I can see 1.5kw being used even if the SP3's are discharging
 
  • Like
Reactions: Tad
mike_hug said:
Please read the whole post. It gives everything you need


I'm not sure I can answer that. You see I have 4 SP3000's all running. There is certainly a discharge limit and I hope to investigate this further, my house draws a lot of power, so I can see 1.5kw being used even if the SP3's are discharging
Click to expand...
I think the test might be by draw more current from the house and see if if the the controller discharge readings go above 2000 watts
You could try some inline fuses at 45 amp just to be safe
 
Last edited:
I just read the boston 5300 battery specs , it says max continuous discharge = 13A ! So if you've got 18 of em in parallel , that's 231A , at 48V , 11 kW , presuming the internal wiring doesn't melt first :)
 
paulca said:
I just read the boston 5300 battery specs , it says max continuous discharge = 13A ! So if you've got 18 of em in parallel , that's 231A , at 48V , 11 kW , presuming the internal wiring doesn't melt first :)
Click to expand...
2000w @ 240v = 8.33 amps on the Growatt. I have 4 on them. I am not drawing anywhere near that sort of amount, so something else is a-foot
 
The battery calculation I think is 5.3a x 3.7v x 18cells x 14packs = 4912.72wh minus wire and connection resistance etc correct me if I'm wrong
Here's another thought I did hear once that if you have had an sp3000 the gbli5001 batteries that were sold with this system may have a full pack of 20 cells making the total design capacity 5490wh....check the label !
 
Last edited:
You must log in or register to reply here.

diy solar

diy solar
Back
Top