Anyone had luck with these MPPT boards? It blew up on me.

[nzoomed]

Im needing to charge a 6S lithium battery bank comprised of 12 x 21700 lithium cells from a 50W panel.
Panel voltage is about 18V or 22V open circuit.
I had one of these MPPT boards floating around ( https://www.aliexpress.com/item/4000552321010.html ) and when I went to set it up using my benchtop power supply starting with about 8V, I tried to adjust the output voltage but couldnt get much more than 15V out of it no matter how much I tweaked the pots.
Increasing the benchtop power supply voltage would help, but I could not get much close to the battery voltage without cranking up the supply voltage of the power supply to that which exceeded the solar panel voltage.
Anyway, I thought I might have had to have added a load while testing, so connected the battery pack and I think it killed the board. I could smell a burnt smell but no smoke.
Specs say 26V is the maximum output voltage, so this should have worked I thought, even though my battery voltage is about 25V

I dont know if i did something wrong, the poorly translated instructions seem vague, but from what I could make out if it, you adjust the MPPT pot until the LED starts to flicker, and adjust the other pot to suit your output voltage.
Is that right?
Either way, I need to find a charge controller that can charge this bank.
TIA

 

[rjenkinsgb]

Looking at the specs in that link, it cannot increase the output voltage, only limit it -

The lowest pressure difference: 1V. Ultra low dropout design is used to achieve low power consumption and higher conversion efficiency,

 

[nzoomed]

Looking at the specs in that link, it cannot increase the output voltage, only limit it -

Ok, so its likely a buck unit? Technically it's not really an mppt unit then?
Could I just find a generic boost converter and hook that up to my BMS instead?

 

[rjenkinsgb]

MPPT does not imply boost, only that the load on the solar panels will be adjusted for optimum output.
The input voltage to output voltage range is a separate feature, depending on the system.

You could run one of those MPPT modules in to a fixed boost converter, to get the voltage you need without having a higher voltage solar panel setup?

ps. Be wary of "BMS" modules - they are not charge controllers, only voltage protection and [hopefully] cell balance.

Conventional lithium cells need a charge controller with current and voltage control, plus a total cut-off at full charge if using the cells to full capacity.


One of these may be suitable? 2A out at around 24V, so close to 50W:

https://www.aliexpress.com/item/1005003189876514.html

It does still need a BMS with balance circuit, to protect the cells in the battery pack.

 

[nzoomed]

MPPT does not imply boost, only that the load on the solar panels will be adjusted for optimum output.
The input voltage to output voltage range is a separate feature, depending on the system.

You could run one of those MPPT modules in to a fixed boost converter, to get the voltage you need without having a higher voltage solar panel setup?

ps. Be wary of "BMS" modules - they are not charge controllers, only voltage protection and [hopefully] cell balance.

Conventional lithium cells need a charge controller with current and voltage control, plus a total cut-off at full charge if using the cells to full capacity.


One of these may be suitable? 2A out at around 24V, so close to 50W:

https://www.aliexpress.com/item/1005003189876514.html

It does still need a BMS with balance circuit, to protect the cells in the battery pack.

Yes that might work. I'm assuming it detects cell voltage?
I installed the BMS to offer protection and balancing, it doesn't appear to offer overvoltage protection when charging, but I would assume it should prevent discharge.
I charged the pack with my lab power supply and set the float voltage to that of a full charge and watch the current load drop off as it charged.
There is no risk of this panel overcharging the batteries, as it can't supply enough current to overload them, only thing that is critical would be the voltage I would imagine.
The cells measured balanced while it was charging on my power supply, so the BMS must be doing something right?

 

[rjenkinsgb]

If you want maximum capacity in the battery, you must use a charge controller that cuts off completely at full charge.

The MPPT one I linked to says it does that, and it combines the boost system to work from a panel like you have.

You can get away with permanent voltage limited charge if you keep it down to around 4V per cell, but that gives a lot lower battery capacity and most balance systems won't cut in at that.

 

[nzoomed]

If you want maximum capacity in the battery, you must use a charge controller that cuts off completely at full charge.

The MPPT one I linked to says it does that, and it combines the boost system to work from a panel like you have.

You can get away with permanent voltage limited charge if you keep it down to around 4V per cell, but that gives a lot lower battery capacity and most balance systems won't cut in at that.

Is that essentially what placing a constant power supply across the battery does?
For instance, I can have it hooked up to my power supply and once charged, the load goes to zero amps, which means its charged, there is nothing that really needs to cut off? Essentially I would be trying to do the same thing with my solar charger.

 

[rjenkinsgb]

With the maximum cell voltage connected, the "trickle" never drops to zero and the cells can be damaged.

With the cells you are using (2 x ~5AH in parallel, so 10AH) the cutoff should probably be when the current drops to something like 200 - 400mA, with full voltage.

Charging can restart when the battery voltage drops by eg. 0.1 - 0.2V per cell.

(You have to consider lithium cells as rather like balloons - any trickle of charge / air beyond the safe rating means they are increasingly likely to burst, creating a serious fire risk).

 

[nzoomed]

With the maximum cell voltage connected, the "trickle" never drops to zero and the cells can be damaged.

With the cells you are using (2 x ~5AH in parallel, so 10AH) the cutoff should probably be when the current drops to something like 200 - 400mA, with full voltage.

Charging can restart when the battery voltage drops by eg. 0.1 - 0.2V per cell.

(You have to consider lithium cells as rather like balloons - any trickle of charge / air beyond the safe rating means they are increasingly likely to burst, creating a serious fire risk).

I guess that also depends on how good my BMS is? Either way, I have a 50w panel and have about a half an amp load I want to keep running 24/7, so unless it's a clear summers day, the battery will take most of the day to charge if we only have about 3 amps or so available power to play with, so any period that it's potentially trickle charging is likely limited.
But I'm happy to install a better charging unit, my predicament right now is I need to get this thing up and running to test out until I get a better charger from out of China which will take weeks.

 

[rjenkinsgb]

OK, with a permanent load it's not going to hold the battery at full charge, so it could work for a while..

I'd try it, and carefully monitor the battery voltage for a few days?

 

[nzoomed]

OK, with a permanent load it's not going to hold the battery at full charge, so it could work for a while..

I'd try it, and carefully monitor the battery voltage for a few days?

Yeah, I should be able to get a decent charger for this eventually.
I was also thinking I can just set the voltage to a lower voltage than max capacity anyway?
About 4 volts per cell is roughly 80 percent charge and would also mean that the battery wouldn't risk over charging?

Would also have the added benefits of extending cell lifespan.
This also brings me to question whether I've over engineered this thing.
Ideally I want it to run 24/7 uninterrupted.
Looking at its power usage it's only using about half an amp at 12 volts. I had built this bank for a 12 watt load.
The 4g router only adds about 0.6A load during standby which is fine.
I have 12x 5000mah 21700 cells, in 6s configuration, 25.2v fully charged x 10ah = 252Wh am I right?
Divide that by 12 watts and I get about 21 hours run time if fully charged?
Could I afford to drop some cells to make it a 4s or 5s bank?
I'm not sure what the average amount of night time operation will work out to so went with 12 hours to be conservative, winter nights will be longer obviously.
I was also wondering whether it's best I leave this number of cells and use a smaller 30w panel and charge the battery bank to a lower percentage instead to extend the lifespan?
Any suggestions would be good, but I want to keep the footprint as small as possible but keep a bit of room for cloudy periods.

 

[rjenkinsgb]

A lower voltage will extend the cell life, however a normal BMS balance board will never do anything, so you are likely to experience problems unless you set the charge to 4.2V occasionally for the system to re-balance.

Watt-hours are calculated from average cell voltage x amp-hours, so eg. 3.6 or 3.7V per cell.

I'd definitely keep the larger solar panel - from the other thread, the long term average is probably 16% rated output, so 8W.

That's not going to last very long with a 12W ?? load.

 

[nzoomed]

A lower voltage will extend the cell life, however a normal BMS balance board will never do anything, so you are likely to experience problems unless you set the charge to 4.2V occasionally for the system to re-balance.

Watt-hours are calculated from average cell voltage x amp-hours, so eg. 3.6 or 3.7V per cell.

I'd definitely keep the larger solar panel - from the other thread, the long term average is probably 16% rated output, so 8W.

That's not going to last very long with a 12W ?? load.

Ok, well i plan to do some real world tests anyway.
Would be nice if I can use a smaller panel as it's quite large.
Am doing a load test on the battery pack right now from a full charge, seems to be holding up well so far.
Been going for 7 hours so far and voltage is still over 24V

 

[nzoomed]

Ok so it's really doing my head in right now.
Just tried using one of these boost converters and the whole thing died on me.

Trade Me

www.trademe.co.nz

What's weird was it was trying to draw a ton of current (my solar panel only can supply 3A, but I cranked up my power supply to 12A and it would put in only 1.5 amps into the battery when connected. The voltage also dropped to match the battery voltage when connected to the board.
At 3A on my power supply, my meter was only reading half an amp going into the battery and still drawing a ton of current around 10 amps.

 

[ChrisP58]

Please post a connection drawing of all parts of your system, and voltages at each point.

 

[nzoomed]

Please post a connection drawing of all parts of your system, and voltages at each point.

I can draw up something later, but I had made a video before it blew up.
Literally it's input was connected to my lab power supply set to 18v and 3 amps to simulate the power supply from the solar panel and them connected directly across my battery bank with my meter in series.
Video below:

 

[tepalia02]

You can make an MPPT solar controller and power supply motherboard like this by yourself:

MPPT Solar Controller and Power Supply Motherboard - Share Project - PCBWay

This is an MPPT Solar Controller and Power Supply and is part of a project to create a wireless home weather station. I plan to use mine to collect meteorological data and send it via wi-fi to my loca...

www.pcbway.com

This is a PCB that accommodates one CN3795 MPPT solar controller module and up to three Mini360 power regulator modules for a wx station.

 

[rjenkinsgb]

Literally it's input was connected to my lab power supply set to 18v and 3 amps to simulate the power supply from the solar panel and them connected directly across my battery bank with my meter in series.

The current limit on the boost regulator should have been set well down, so it could not take excess power - no more than available. With no current limit, it will not protect itself.

It should be in current limit until the correct charge voltage across the battery is reached, then the current start to drop as the battery finishes charging.

Also, there is something seriously wrong with the PSU, if it's capable of 10A but the voltage was dropping way down at relatively low current!

 

[nzoomed]

You can make an MPPT solar controller and power supply motherboard like this by yourself:

MPPT Solar Controller and Power Supply Motherboard - Share Project - PCBWay

This is an MPPT Solar Controller and Power Supply and is part of a project to create a wireless home weather station. I plan to use mine to collect meteorological data and send it via wi-fi to my loca...

www.pcbway.com

This is a PCB that accommodates one CN3795 MPPT solar controller module and up to three Mini360 power regulator modules for a wx station.

That's exactly the sort of thing I'm looking for, I was looking for something simple that can be made with off the shelf parts on some veroboard so I can get this unit working, can't wait to get something out of China. I only need 2 amps output

 

[nzoomed]

The current limit on the boost regulator should have been set well down, so it could not take excess power - no more than available. With no current limit, it will not protect itself.

It should be in current limit until the correct charge voltage across the battery is reached, then the current start to drop as the battery finishes charging.

Also, there is something seriously wrong with the PSU, if it's capable of 10A but the voltage was dropping way down at relatively low current!

Turns out my PSU has a constant voltage and current setting. Took me a while to figure out what was happening and how to change it. The interface was not that intuitive.
Anyway I did have the current set at minimum at first and then cranked up.
The biggest issue is the surge current when connecting to the battery.
If you crank it down and slowly up, it appears OK, but still getting hot with only half an amp output. I tested with the PV cell and it's no performing great. I definitely will need some kind of mppt controller.