Earlier this summer, we began investigating our Twitter followers' experiences with public electric-car chargers in our weekly polls.

In June, we learned that more electric cars get charged at work than at public chargers.

Last month, we learned that the type of public charger that electric-car drivers rely on most often is still 240-volt Level 2—even with more than 200,000 Teslas on the road that have access to Superchargers.

Last week, we explored the question of how often those public chargers (both  Level 2, or Level 3 DC fast chargers), are out of order when our electric-car-driving Twitter followers try to use them.

Specifically, we asked, our followers: "How often do you encounter a public electric-car charger that is out of order?"

Significantly, just over half of our respondents said it happens to them more than 10 percent of the time.

The other half, 48 percent said they only encounter chargers not working "Infrequently."

Among those who said they encounter public chargers that are out of order more often, a quarter said it happens just 10 percent of the time.

Another 18 percent said every fourth time they go to charge up they find the charger doesn't work, a number that seems like it would truly get frustrating quickly.

In addition, 9 percent said it happens half the time.

2017 Chevrolet Bolt EV electric car at EVgo fast-charging station, Newport Centre, Jersey City, NJ

2017 Chevrolet Bolt EV electric car at EVgo fast-charging station, Newport Centre, Jersey City, NJ

Taken together, nearly three-quarters of our respondents said they encounter a public charger that isn't working 25 percent of the time or less.

We're not sure what constitutes realistic and adequate up-time for a public charger, but for those trying to get home or to a destination who find one not working regularly, the situation seems fraught with frustration.

We also didn't ask how often electric-car drivers encounter gas cars blocking electric car charging spaces. (And we won't even ask about plug-in hybrids, so as not to spark a Twitter war about charging etiquette.)

All of these questions may be worth exploring further in the future.

In the meantime, please note that our Twitter polls are unscientific, because they have neither the sample size nor the random responses required to achieve a representative sample.