Performance Issues with Compatible Batteries

Exactly in the 13th month of my purchase, the Xolo Q700 mobile phone started showing severe battery issues. It would take about 2 hours to charge but hardly last for 30 minutes when using Youtube. In “Idle / Standby mode”, the phone would last 3 – 4 hours before switching off. Was I upset that a pricey product started showing faults the day after the warranty expired?

Opening the back cover revealed a really nasty surprise. The battery had bulged like a pillow and was getting burning hot during charging. Clearly the battery had reached end-of-life and needed to be replaced. Thankfully, the phone has a removable battery and I could spot the issue easily. In a sealed phone, the issue may have remained undetected and a consequent battery explosion could have proven deadly.

Like a typical Indian electronics brand, the spares for products are just not available in the market.

• Xolo does not sell spares on their website or through another ecommerce partner.
• Xolo does not have brick and mortar mobile phone shops.
• Xolo spare parts are not available with other mobile phone brick and mortar shops.
• Emails to Xolo are met with boilerplate answers like “Please visit our nearest customer care center”. Specific questions such as “Can you confirm availability before I make the trip?” are always left unanswered.
• Internet is rife with stories of Xolo customers waiting forever to receive spare batteries from the service center or being overcharged for them.

Since the phone is out of warranty in any case and it is now a standby phone for me, I decided to take the plunge and buy a compatible battery. These batteries are popularly known as “China made” and also sold on ecommerce sites as “OEM” batteries. For all I know, the batteries could be made in Timbuktu. The hallmark of these products seems to be cheap prices and short life.

Product for Testing:

The battery I bought, was titled “XOLO Q700 Mobile Phone Battery OEM” and being sold on ShopClues (Product ID: 5313502) by seller “Kartvalue, New Delhi, Delhi”. I paid  Rs.275/- for it while prices for the same product were up-to Rs.900/- on other ecommerce websites.

The battery looks like the original product from Xolo. No spelling mistakes, no printing mistakes, no packaging mistakes. I had high expectations from it. In fact, the battery worked out of the box with the phone reporting 44% charge.

I decided to put the battery through the grind in an unscientific method to determine the idiosyncrasies of the product.

Tools for Testing:

The test was conducted purely using software that is available for Android phones. No industrial-grade test instruments were used.

I installed the following apps on the phone:

• Battery Log by Hwang Tae-rok (Link) : This application runs in the background and makes a log of the battery voltage, battery temperature and date-time whenever the battery-life percentage changes. The logs can be exported to CSV format for analysis.
• StabilityTest by intostudios (Link) : This application features a “CPU+GPU Stability Test” mode which is effective in consuming battery power without making a ruckus. The tests run in loop mode till the phone turns off.

Procedure for Testing:

I executed the tests for 3 cycles, spread over a span of 2 days. The testing procedure was identical in each case.

1. Charge battery to 100% level in Offline mode (Phone is switched off) using a 5V 2A USB Wall Charger.
2. Keep charger connected and switch on phone, wait for boot process to be completed (essentially, wait for 2 minutes before next step).
3. Start Battery Log app and minimize app.
4. Start StabilityTest app and start CPU+GPU stability test. Wait for testing to begin.
5. Switch Off Wall Charger so that testing may continue on battery.
6. Leave device alone till it switches off due to low-battery.

After testing, I exported the data from Battery Log app in CSV format and transferred it to the computer for analysis. I deleted those rows from Battery Log which were created in the interim stages of testing (such phone boot up, app start etc.). The clean CSV file was imported into Plotly and a graphical plot was created of the data. The plotted data was exported in PNG format and uploaded to this website for presentation. You can view the plot here.

Inferences:

If you study the graphical plot of the data, you will notice the following:

• The battery lasted between 5 – 5.5 hours in each test. We can average it out to 5 hours and 15 minutes of runtime in a new battery. This comparable to the manufacturers original part if his part were to be put under the same intensive test.
• The battery charge level drop was non-linear. After approx. 47% battery-life, the reported battery-level dropped like a stone till it reached approx. 12%. Though shocking, it can be explained. The battery charge levels are computed by Android OS based on historical full-charge-voltage and minimum-charge-voltage. If the reported voltage level drops non-linearly, the reported battery charge level too drops non-linearly.
• The battery voltage level drop was reasonably linear and became non-linear at 25% battery charge level reading. This too can be explained. Batteries rarely discharge linearly and some geometric progression is expected. The integrated circuits that computer battery voltage are consumer-grade too and can make errors in determining the true levels.

Caveats and Conclusions:

Unscientific tests such as these may be used only to draw generalized conclusions about product quality and life. In the tests that were conducted, a number of factors can contribute to errors or variance in data. Factors such as:

• Consumer-grade electronics that are responsible for reading and reporting data to the OS.
• In accuracies in software code.
• Complexities of the non-realtime Android OS where many background processes may interfere with the tests.

Based on the test data, I will move to the following assumptions:

• I expect the battery life (runtime) to reduce over the next 3 – 6 months with 50% battery charge levels by end of year. This is also depended on the full-charge-discharge cycles that the battery is exposed too. Full-discharge is bad for batteries.
• I expect the vertical drop in battery charge levels to move upwards from present 47% to 60% level. i.e. the battery may show rapid discharge once the battery level reaches 60%.
• Towards end-of-life of this battery (probably a year), this battery too will swell like a pillow.