Saturday, December 03, 2016

Battery Blues

Our stairwell now has a functional solar light that gets triggered with an automatic twilight switch. Yet, the circuit is not really working as I had expected, and I wonder if the problem lies in the solar panel, the charger, or the battery.

Initially the panels had been rigged to a Sealed Maintenance Free (SMF) battery I got from a friend. It was an Exide 26Ah SMF - compact,and relatively lightweight at ~8 kg. The trouble was that even after getting charged by the 60Wa panel in over five hours of clear sunshine, it would not get the ~12W LED panel to run for more than a few hours.

Perhaps this was because the battery had got discharged after lying unused for more than a year. So, based on an online tutorial, I pried open the sealed valves and poured in a bit of distilled water in all its compartments. This did not seem to work either, with the LEDs going blank after about 3-4 hours every evening.

At this stage I purchased a dedicated solar battery from Okaya. This one was a 40Ah tubular battery, weighing more than 15kg. Big, heavy and designed for the outdoors.  In theory it was expected to hold about 480 Wh of juice (40Ah x 12V). It worked just fine the first night. The lights switched on automatically at dusk (5:30PM) and stayed on until daylight turned the twilight switch off (6:30AM). In effect, during this 13-hour duration it should have used up about 156Wh (12W x 13 hrs).

Again, in theory, there should have been about 324 Wh remaining in the battery (480 Wh - 156 Wh). Yet, within two days, the battery turned out to be completely discharged.

Okaya was a bit tardy when it came to servicing warranties but a service engineer turned up and checked the battery. It was showing an output of 12.31V on the multimeter, but after checking all the six compartmetns with a hydrometer, he declared that the  battery had indeed got discharged. Aparantly, the density (specific gravity) of the electrolyte had dropped way below 1.265 for a fully charged battery.

The next day, he came with a charger (16V), plugged the battery to the mains and said that it would get recharged in the nex 24 hours. All the caps were left oepend during charging because he expected it to bubble during the charging period.

Now the solar lighting system seems to be working perfectly fine, even though I continue to be nervous about the voltage drops. A dusk-to-dawn operations brings down the battery voltage to 11.86V, and it looks like the 60Wa panel is not really doing its job of topping up the battery even when there is bright sunshine for ~5 hours every day.

So, finally, the key questions that need answers are -
  • How does one measure the juice remaining in a battery? 
  • If the 60Wa panel is not doing its job, should I get an additional panel?
  • Is the LED panel consuming more than the 12W, as claimed by the vendor?


8 Dec,, 2016 - Update:

So far, my understanding of solar systems has been guided by the WWW, shopkeepers and technicians who had many years of installing systems. Yesterday, I finally met somebody who was not only a hands-on expert but also one of the pioneers in the solar ecosystem in India -- Yeshwant Thakur, ED of Surya Kiran Technologies Inc.

One look at a pic of my hastily assembled unit and he pronounced that I had been wasting my time and money. The "Tata BP" solar panels were cheap imitations that could never deliver 60Wa; the charge controller was a rip-off, and, I did not have to buy a separate twilight switch in the first place!

A dusk-to-dawn sensor, built into the CC was all that was needed. It would simply sense the incoming current from the panels to decide that the ambient light was not enough, and to turn on the light circuit.

Also, if the battery needed 10 "chappatis" of sunlight, it the panels had not been giving it even the bare minimum of two (6Amps), even while the LED lights were making it work overtime. So, in the absence of sufficient current coming in from the sunlight, the battery had gone back into deep-discharge mode. The only solution was to get the battery charged on the mains, once gain, and to get a new panel that feeds in at least 100Wa (6A) to the battery.

Wish I had met YT before seeking advise from the quacks at Bhagirath Palace!

23 Feb., 2017 - Update:

By stumbling, I am learning to walk.

For the third time in two months my Okaya battery got discharged for mysterious reasons. Earlier, it was the amperage of the solar panels that was causing the discharge. The vendor had sold me a 20Wa panel claiming that it was an "original" 60Wa panel from TataBP. I had got this one replaced with a 100Wa panel from a Noida-based company.

Since I had been travelling during Dec-Jan., I was able to fix the new panel about a month after I had got the battery charged on the mains. When the battery went kaput after three days, I had to get it charged on the mains again. During the charging process (20+ hours) the acid levels had dipped to half, so I had topped it up with distilled water after charging before fitting it back to the solar unit. So now I had a 100Wa panel, a 40Ah battery, a solar charge controller (Phocos) and a twilight switch (Electrocine).

Now I thought I could forget about the lighting system for the next one year, at least. Yet, the charge lasted for less than a week. What had gone wrong? Winter was over, so there had been sharp sunshine for more than 6 hours every day (100Wa * 6A = 600Wh), the battery was getting charged on a 6A current, so it should have had about 500Wh of juice to spare (40Ah*12V = 480Wh), an the load was less than 150Wh/night (10W * 12hrs = 120Wh). So there was no reason for the system to fail.

The Okaya engineer - Ranjit -  visited today to do some serious trouble shooting. He loaded the battery to the home inverter and added load (5 lights, 2 fans) until the voltage dipped to less than 10.5V. Then he tested each compartment and found that all 6 were working just fine. The only possibilites were -  the solar panel was not able to charge the battery, or the SCC was cutting-off the charging process. Neither seemed logical.

Ranjit then had a brainwave. The mistake had been to top-up the battery water after it had been charged on the mains for two days. This had brought down the specific gravity, and, in turn rendered the whole exercise useless.

Lessons learnt:

  • Never all top-up the battery water levels after it has been charged on the mains
  • Lead Acid batteries self-discharge fast. So avoid leaving them idle for more than a week.


* All about lead acid batteries -
* Lead acid battery charging basics -
* DV Voltage Drop Calculator -

* Video - Using a hydrometer to test batteries -
- A lead acid battery cell is fully charged with a specific gravity of 1.265 at 80° F.

* Specific Gravity of LA Battery acid -

* Maintaining SMF batteries -

No comments: