2
\$\begingroup\$

I have been put onto a project that needs to be powered by a small (1-2W) solar panel trickle charging a battery pack. The product will be located on rooftops. We have reviewed similar products and have found them to be using NiMH batteries so are planning on following suit. I have reviewed many similar posts, particularly the below, but still have some unanswered questions, apologies if I am repeating the questions of others.

The battery needs to supply power to a 12VDC motor and a 24VDC Actuator (planning on boosting 12V to 24V) but these will only be active for a small amount of time each day so overall energy requirement is quite low.

At this stage we are planning on using 10xAA NiMH 3600mAh in series to get a 12V output. I plan on using a Zener diode between the battery and panel to prevent back discharge at night.

The solar panel will be something like the 12V model at the bottom of this page: https://www.futurlec.com/Solar_Cell.shtml

I understand that trickle charging NiMH batteries is not ideal, but it seems to be commonly done and NiMH seems like the ideal chemistry for the application.

My questions are:

  • Is a 12V panel the best choice for trickle charging the 10 batteries in series?
  • The panel linked outputs a peak of 80mA, 80/3600 = 1/45C, my understanding is that this is well within the safe limits for solar charging of NiMH, is there a benefit in regulating this to a constant current (with LM317 for example)?
  • Should we be considering including an IC such as the DS2715 to monitor battery level and prevent overcharge, or is this completely unnecessary given the low output of the panel?

Happy to provide any more information that would be useful.

Cheers.

\$\endgroup\$
  • \$\begingroup\$ Welcome to EE.SE! I changed your "power requirement" to "energy requirement" since that's what it is. \$\endgroup\$ – winny Apr 1 at 8:03
  • \$\begingroup\$ "NiMH seems like the ideal chemistry for the application" - Why? Lithiums are easier to charge and keep charged, have higher energy density and possibly higher power density. On the downside they don't like temperature extremes and may or may not have higher cost per energy unit. \$\endgroup\$ – JimmyB Apr 1 at 11:03
  • \$\begingroup\$ @ Winny, thanks for the welcome and correction! @ JimmyB, I should have mention this product will be sold in Australia, so will potentially be exposed to some pretty high extremes of temperature. We considered using a lithium a potential hazard because of this. Their ability to self-limit overcharge as Winny mentioned is another benefit \$\endgroup\$ – Jacob Rodda Apr 1 at 23:16
0
\$\begingroup\$

You need to know the short circuit current and open circuit voltage of your solar panel to be able to tell for sure, but NiMH is a good option here since they self-limit any oversharge well given that the current is low enough, and at 80 mA, it should be. NiCd is even better in this regard, but they are terrible otherwise. If your open circuit voltage is much higher than 1.6 V/cell, I would at least think about it and test it first.

Perhaps you could put a bit less stress on the batteries by using an LDO like your suggested LM317 to limit the peak voltage. If it's a one-off, I would not bother. If it's mass production, you need to find and test the worst case scenario in this regard and check the datasheet of your batteries.

Normal NiMH charging is done with either negative delta V detection or temperature sensor to terminate the charge. The former is possible in your case but the current is too low for the latter.

NiMH charging

If you want to keep it simple still and limit the voltage and balance the cells, something like this would be a safe bet. 1.5 V Zeners don't exist, hence two cells in series per Zener (~3.1 V):

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
  • \$\begingroup\$ "temperature sensor to terminate the charge" - Will probably not work on a rooftop where ambient temperature can vary significantly within seconds. \$\endgroup\$ – JimmyB Apr 1 at 11:00
  • \$\begingroup\$ @JimmyB which is one more reason on top of what I already wrote. \$\endgroup\$ – winny Apr 1 at 11:49
  • \$\begingroup\$ Thanks for the great response! We are tasked with making a proof of concept prototype so I think we will keep it simple and then recommend the circuit be revised before production to maximize battery lifespan. Am I correct in thinking that simply connecting the panel too the 10S battery pack with a blocking diode (or not) will slowly be able to charge all the batteries (albeit at some detriment to their lifespan), or is charge balancing on some level essential? \$\endgroup\$ – Jacob Rodda Apr 1 at 23:19
  • \$\begingroup\$ @JacobRodda Not essential. It would only minimize the risk somewhat and possibly increase the lifespan of the battery pack. \$\endgroup\$ – winny Apr 2 at 6:27
0
\$\begingroup\$

I have a panel of this current level and the reverse current is less than 1 mA so I don't use a blocking diode. The panel will need an open circuit voltage of 18 to 20 V and should be able to supply a reasonable current at 16V. The panel is a current source so adding an external one is redundant. Determining end of charge is very difficult here because the battery voltage will vary with current which will vary with sun intensity. An undervoltage lockout might be desirable. 3.3V zeners have a very high conductance at lower voltage and will be a significant drain. When NiMH were first introduced they could not tolerate a trickle charge but they seem to make ones that can now.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.