AGM Battery

For help with the bus itself (not a specific product).
rpu_bus
Posts: 165
Joined: Sun Sep 18, 2016 6:07 pm

AGM Battery

Postby rpu_bus » Fri Jul 21, 2017 3:08 pm

My 10AHr AGM SLA battery charges at a higher voltage, so the thermistor I used is not allowing much charging during the day. I have removed the thermistor and will let the charger go all the way to its max 14.3V output. This may cause some gassing of hydrogen but the AGM should recover it (right?).

RPUno^6+RPUadpt^5WithAgmBat+4LedString.jpg
RPUno^6+RPUadpt^5WithAgmBat+4LedString.jpg (162.18 KiB) Viewed 150 times

rpu_bus
Posts: 165
Joined: Sun Sep 18, 2016 6:07 pm

Re: AGM Battery

Postby rpu_bus » Sun Jul 23, 2017 2:40 pm

After a few days, the 10AHr sealed AGM seems to be doing well. The RPUno used 473mAHr blinking the LED strings during the first night (7/21/2017 -7/22/2017) and charged 773mAHr from the 3W PV (7/22/17). On the second night (7/22/2017 -7/23/2017), the RPUno has used 501.87mAHr. At noon it was charging at a rate of nearly 4mAHr per minute.

With the thermistor removed the highest voltage I have seen is about 13.5V (i.e. it has not got to the 14.3V float setting yet), so I don't think it has even used the AGM to recover hydrogen. I was using an SLA charger to keep the battery charged but in retrospect, I am starting to think that these AGM may actually do better with some hydrogen liberation. This could add tremendous longevity to the battery because I know for a fact that sulfation is what kills sealed lead acid batteries.

rpu_bus
Posts: 165
Joined: Sun Sep 18, 2016 6:07 pm

Re: AGM Battery

Postby rpu_bus » Sat Sep 02, 2017 1:27 pm

After running the [NightLight] for a few weeks ([RPUno^6NightLight_log]) I pulled the battery just before the sun went down and tried to charge it with a Duracell SLC10003 charger, the charger decided the battery was at a full charge within a few minutes. The SLC10003 goes to 14.5V with 1.5A so I guess the 150mA is doing the job even though I don't see it even reach 14.3V.

[NightLight]: https://github.com/epccs/RPUno/tree/master/NightLight
[RPUno^6NightLight_log]: https://github.com/epccs/RPUno/blob/master/Hardware/Evaluation/RPUno%5E6NightLight_log.txt

One thing that is clear is that the buck converter is not doing much of anything useful with this AGM 12V battery, it works as advertised but when the battery voltage is so close to the MPP the converter is harming the power storage more than helping. The buck converter turns 85% of the 16V power into 14V power, but if I just run the PV at 14V, then I have only lost 12.5% (e.g. 87.5% of the Maximum Power is put in the battery). Also as the PV panel heats up the Maximum Power Point is lower which makes the situation even worse.

With a 3W PV on a 10AHr battery, the voltage never got high enough to reduce the charging current, this effectively means I only needed a diode between the PV and the battery. At higher charge rates the battery voltage should increase above 14.5V, and then some way to throttle the power flow is needed (at which point MPPT is useless).

There are a few options to throttle the power flow from the PV. One is too short circuit the PV with a MOSFET. Since the PV is acting like a current source below its power point this means the current flow is nearly the same when shorted as it is when charging, therefore noise from magnetic fields is actually less than if the current flow was abruptly stopped. Capacitively coupled fields are a toss up since either shorting or opening the PV results in a substantial discontinuity in voltage.

Another option to consider is a regulating diversion load. The idea is to switch on the diversion load when the voltage reaches 14.7V and then have it draw a linear amount of current between 14.0 and 14.7V. For example, a 3.5A 12V diversion may sink .5A at 14.1V and 3A at 14.6V. The problem is that the diversion has to sink the power, and will need a way to dissipate heat. The diversion can also fail and then the PV either needs to be shorted or opened to prevent the battery from over charging.

Anyway, I no longer think it is a good idea to integrate the solar PV and battery power onto the microcontroller board.


Return to “Installation, Troubleshooting, and Usage”

Who is online

Users browsing this forum: No registered users and 1 guest