Neey active balancer equalizer hack needed

[Skypower]

I have several Neey clones that work great but can’t shut them off at a pre determined voltage and reasonable back on voltage (voltage spread too great). I know the newer ones can when using the android app.
The only way to turn it on or off is through the iOS app which isn’t going to work. I tried to disconnect a lead (not the first or last) which put it into fault and stopped it and it looked good when reconnected but it wanted a restart through the add again…damn. Has anyone found terminals inside that can halt and start the balance without app interaction.
The thing balances fast with big ah batteries but I can’t be there to turn it on above 3.4vpc and off below. Dang shame because it’s useless as is.

 

[time2roll]

I would try removing the main negative wire and see if a fault is given or shuts down for low voltage.
I see this with a similar balancer.

"Can I turn the unit off by installing a switch on main negative wire? This would bring overall voltage below 12v as voltage > 12v is one of requirements for being on

• Yes"

https://www.aliexpress.com/item/4001013797039.html

 

[Skypower]

I tried that. It comes back ready but it still wants a restart through the app. There must be thousands of the early versions out there. But they are useless unless you can be there to start it above 3.4 and stop it below. There looks to be a terminal inside. If bet there’s a way but I’ve got 8k of batteries to risk so not playing. Lol

 

[RCinFLA]

Think the old firmware did that. Turned on at your setting but then never turned off all the way down in voltage. Newer firmwire shuts off based on your setting.

 

[Skypower]

I think it also must be using the Android app.
GWbms app is the only one I can find for IOS Apple and it doesn’t have the needed line “BatteryequalizationVol” in the settings.

 

[Skypower]

For the Neey and clones there’s a new app for IOS. If you have the old app (GWbms) you must delete it before you load the new “GeeWe BMS” app. This new app gives you total control of start balance voltage and sleep voltage.

 

[MrX]

I have the original Neey active balancer. The revised HW that have the “0,3v issue” solved. I tested it for the first time today (connecting with IOS GeeWe BMS app).

EquVoltage and SleepVoltage seem to work as expected, but the StartVol (eg Voltage Diff) -parameter doesn’t seem to work at all. Balancing never stops even if cell voltage difference is below set value. It’s even balancing when the difference is zero mV (?!)

I have tested this with two different units, same result. Have anyone else experienced this?

 

[Samcat]

It is looking at average cell voltage to wake up or sleep and this why there is still issues with this design. Proper design should watch for individual cell voltage or cell voltage delta. Some other Chinese balancers are even more worse where they only look for battery total voltage to trigger operation.

 

[MrX]

I’m aware that it looks at the average, I’ve watched Andy’s videos on this thing. The issue I see is different and not like the ones in Andy’s videos. As I tried to explain, the StartVol(V) parameter doesn’t work, everything else does. As long as the conditions for EqualizationVol and SleepVol are met, the device keeps balancing even if the cell voltage difference is less than StartVol.

 

[ocular]

I had a standalone 2A JK active balancer added to the BMS + 16S 300AHr LiFePO4 offgrid setup - had cells 1-4 each 200mV less than the rest of the pack when fully charged to 55V. This active balancer also has no means of turning off <3.4V (unlike the active balancer in the JK BMS). By setting the delta V ((?EquilizationVol) to 20mV - over about 60 days the pack was back in balance at 55V with delta V = 20mV. Even with a very unbalanced pack with 4 cells each 200mV less than the rest when pack fully charged to 55V- when the pack is in the flat of the voltage graph delta V is consistently <20mV so active balancing is minimised when average cell voltage is in the flat part of graph. Adjusting this setting so the active balancer works only when "pressure difference" is greater than 20mV can be used as proxy for on above 3.4V.

I believe you should be able to use the early version of the Neey 4A active balancer and leave it connected all the time by adjusting the voltage difference parameter and using this to turn balancing off and on. Just check the definition of the terms as something may have been mixed up in translation. To my way of thinking SleepVol is when the unit goes off line when overall pack voltage is low. Start voltage is when the unit turns back on, neither of these can be used around 3.4V and EquilizationVol is when balancing comes on back on after delta V is greater than the setting.

I also have a 3rd generation 4A Neey Active Balancer (with the ability to turn off when average cell voltage is<3.4V) that I havent installed. I think it would be a good test to take a balanced pack off line and unbalance say 4 cells significantly and note the time it takes to balance when pack held at 55V. Repeat the test with another active balancer. Then rank the balancers in terms of its "balancing power".

 

[Samcat]

I’m aware that it looks at the average, I’ve watched Andy’s videos on this thing. The issue I see is different and not like the ones in Andy’s videos. As I tried to explain, the StartVol(V) parameter doesn’t work, everything else does. As long as the conditions for EqualizationVol and SleepVol are met, the device keeps balancing even if the cell voltage difference is less than StartVol.

View attachment 97617

What you pointed with red is cell deltaV and what you want to set for cells to be balanced. Lower the number more time take and less drift between cells will be.

 

[Samcat]

I had a standalone 2A JK active balancer added to the BMS + 16S 300AHr LiFePO4 offgrid setup - had cells 1-4 each 200mV less than the rest of the pack when fully charged to 55V. This active balancer also has no means of turning off <3.4V (unlike the active balancer in the JK BMS). By setting the delta V ((?EquilizationVol) to 20mV - over about 60 days the pack was back in balance at 55V with delta V = 20mV. Even with a very unbalanced pack with 4 cells each 200mV less than the rest when pack fully charged to 55V- when the pack is in the flat of the voltage graph delta V is consistently <20mV so active balancing is minimised when average cell voltage is in the flat part of graph. Adjusting this setting so the active balancer works only when "pressure difference" is greater than 20mV can be used as proxy for on above 3.4V.

I believe you should be able to use the early version of the Neey 4A active balancer and leave it connected all the time by adjusting the voltage difference parameter and using this to turn balancing off and on. Just check the definition of the terms as something may have been mixed up in translation. To my way of thinking SleepVol is when the unit goes off line when overall pack voltage is low. Start voltage is when the unit turns back on, neither of these can be used around 3.4V and EquilizationVol is when balancing comes on back on after delta V is greater than the setting.

I also have a 3rd generation 4A Neey Active Balancer (with the ability to turn off when average cell voltage is<3.4V) that I havent installed. I think it would be a good test to take a balanced pack off line and unbalance say 4 cells significantly and note the time it takes to balance when pack held at 55V. Repeat the test with another active balancer. Then rank the balancers in terms of its "balancing power".

If third revision still looks at average cell voltages to start balancing that is still issue that will going to happen and have one cell hitting over 3.65v but average still is not enough to kick in balancing. DeltaV is only used when balancing is active. When idling it is not looking at all. At least with my testing and how it pings various things from battery.

 

[ocular]

If third revision still looks at average cell voltages to start balancing that is still issue that will going to happen and have one cell hitting over 3.65v but average still is not enough to kick in balancing. DeltaV is only used when balancing is active. When idling it is not looking at all. At least with my testing and how it pings various things from battery.

I agree that using the average cell voltage will start the balancing process later, but in that scenario you describe manual adjustment of max charge voltage should help. What max charge voltage were you using? With a start balancing average cell voltage of 3.4 and a max voltage of just over 54.5V the Neey balancer should come on to catch the higher cell. As balancing proceeds then the max charge voltage can then be slowly increased. What you are describing can also happen with any active balancer if the max charge voltage is set to high which increases the charge input beyond the ability of the balancer to cope. The Neey algorithm with its slow sampling rate and using the average cell voltage is not ideal and needs closer observation with an out of balance pack.
But using delta V may be a better way to turn balancing on as voltage difference increases and turn it off in flat part of the voltage graph. What is the best delta V setting to try ? 20mV ?50mV It would be interesting to test if the Neey 4A balancer has more balancing power then the 2A JK active balancer?

 

[Samcat]

I agree that using the average cell voltage will start the balancing process later, but in that scenario you describe manual adjustment of max charge voltage should help. What max charge voltage were you using? With a start balancing average cell voltage of 3.4 and a max voltage of just over 54.5V the Neey balancer should come on to catch the higher cell. As balancing proceeds then the max charge voltage can then be slowly increased. What you are describing can also happen with any active balancer if the max charge voltage is set to high which increases the charge input beyond the ability of the balancer to cope. The Neey algorithm with its slow sampling rate and using the average cell voltage is not ideal and needs closer observation with an out of balance pack.
But using delta V may be a better way to turn balancing on as voltage difference increases and turn it off in flat part of the voltage graph. What is the best delta V setting to try ? 20mV ?50mV It would be interesting to test if the Neey 4A balancer has more balancing power then the 2A JK active balancer?

Watching after average Voltage is not going to help keeping your battery pack in check no matter how good battery pack is. It will happen over time.And this is not how balancer should be working. Even if you set max pack voltage it is possible to happen because some cells will become runners against the rest. And when this happens your balancer will be still blind in what actually happen. We need balancer that will look after Voltage delta and options where you set when to actually start balancing. For balancer to wake up from idling should be looking at set window and then start looking for pre set delta to run. And also have option to reject any sampling or run when there is discharge and there are spikes when starting high in rush loads. I have yet to get my hands on JK balancer. I have ordered 5A and 10A . But most of balancers as delta is getting down to single digits mV are not going to push full current to the cells. Higher the delta more likely it will stay to the max ratings. Resistance and balancer board temperature will also affect this .
Max voltage is not something you should be worried if all parameters are set correctly but current delivered when balancer is active and trying to bring delta down. But overall we are on same page about this beaten path with Chinese balancers.

 

[MrX]

What you pointed with red is cell deltaV and what you want to set for cells to be balanced. Lower the number more time take and less drift between cells will be.

Thats how it is supposed to work but it isn’t.
Even if set to 1v it keeps balancing the pack even if the pack is in complete alignment.

 

[Samcat]

Thats how it is supposed to work but it isn’t.
Even if set to 1v it keeps balancing the pack even if the pack is in complete alignment.

How you confirmed with calibrated volt meter readings vs app that is pulling via Bluetooth?

 

[Samcat]

Neey is one of worst balancers i have benchtop tested for accuracy. I believe it was 2nd revision. Babysitting this balancer is possible to live with but with my work and precision to my battery pack assembly neey is no go no matter how much someone wants to pay me to make it work. Please don't take my post as offense. This are my personal opinion when I have been playing with for accuracy and how topology is run with this balancer.

 

[MrX]

How you confirmed with calibrated volt meter readings vs app that is pulling via Bluetooth?

I have not verified with a clamp meter that balancing is actually happening but the app tells me so, and the cell voltage is indeed aligned

Boring screen recording of behavior

 

[Skypower]

I’m sorry that some of you guys are having problems. Since I’ve been running the new app it’s been flawless. I seem to be running the same hardware but newer version app. The only minor issue is the capacity(ah) can’t be changed but it doesn’t seem to matter.

 

[ocular]

I’m aware that it looks at the average, I’ve watched Andy’s videos on this thing. The issue I see is different and not like the ones in Andy’s videos. As I tried to explain, the StartVol(V) parameter doesn’t work, everything else does. As long as the conditions for EqualizationVol and SleepVol are met, the device keeps balancing even if the cell voltage difference is less than StartVol.

View attachment 97617

I have the original Neey active balancer. The revised HW that have the “0,3v issue” solved. I tested it for the first time today (connecting with IOS GeeWe BMS app).

EquVoltage and SleepVoltage seem to work as expected, but the StartVol (eg Voltage Diff) -parameter doesn’t seem to work at all. Balancing never stops even if cell voltage difference is below set value. It’s even balancing when the difference is zero mV (?!)

I have tested this with two different units, same result. Have anyone else experienced this?

Isn't that the problem Andy noticed in his video where StartVol or Run Balance didn't work (Neey was always balancing). Hankzor responded by saying that this was the way the firmware worked when first started up. After delta V reaches a certain value (?1mV seems a bit small ) then StartVol and Run Balance should then work. Not sure if this strange behaviour is also a feature of gen 3 Neey hardware?