Memory backup capacitors: why a capacitor?

Question

Why would an engineer choose a memory backup capacitor instead of any of the types of rechargeable batteries available?

Their energy density seems far poorer, and they can't hold their charge very long.

The only advantage I can think of is it's slightly simpler to charge? (Just needs a resistor), but then again, the same can be said for trickle charging many batteries.

Answer 1

They're cheaper and (rightly or not) in product design that's often an argument which overshadows a lot of technical arguments.
And, like you said, capacitors don't require special charging electronics. Also, batteries contain chemicals for which you have to comply to special regulations. (Sometimes this is the reason why toys and such are delivered without batteries.)

edit
Another thing: batteries are not suited for wave-soldering. You'll either have to solder them manually afterward (cost), or, like for coin cells, use a battery holder (cost) and add the battery manually (cost).

Answer 2

Rechargeable batteries for memory/RTC backup power isn't a good solution today. The batteries will eventually die due to the many charge/discharge cycles that it can be put through. And many rechargeables will self-discharge in a month or two. There are also regulations regarding the metals in these batteries that might come into play.

Non-rechargeable lithium batteries can have a lifespan of 10+ years in a RTC/memory backup use case. This is often longer than the expected life of the product it is in. So, it has become common to see these batteries soldered down to a PCB-- because it's more rugged than a battery holder and it is never expected to be replaced. Of course, this makes many people nervous. These batteries also have the regulatory issues common with the rechargeables.

Supercaps are a nice alternative. They are not always cheaper than the non-rechargeable lithiums, and don't operate as long on a charge as the rechargeables. But they are maintenance free and don't have the regulatory issues.

So, none of the solutions are a perfect mix of cost, regulatory, lifespan, and maintenance free. The designers of a product just have to weigh the pro's and con's and pick the one that they like best.

BTW: One product I did used a Supercap as an RTC backup. It would run about 9 months before loosing it's charge. Of course, turning on the unit would recharge that cap in about a minute.

Answer 3

Capacitors are nicer for the end user because they don't require any maintenance. They charge up quickly the first time the product gets turned on, instead of taking a day or two to get fully trickle charged. They work for thousands of times more charge cycles and don't leak caustic chemicals onto the circuit board when they stop working. Capacitors are easy to charge, the circuitry around them is simpler, just limit the voltage and current going into them and they'll stop drawing energy when they're full. Batteries generally either need a chip to monitor charging or will tend to get somewhat overcharged if left on trickle charge indefinitely without being discharged, which is detrimental to their lifespan. Batteries also have to be disconnected from the load when they get discharged down to their minimum voltage, which requires extra circuitry, whereas capacitors can be discharged to zero volts.

Batteries hold more energy in the same volume/weight, but otherwise are a pain to manage, all the other advantages are with capacitors. A battery makes the most sense in a small/portable product, or one that is so cost-sensitive that it's worth inconveniencing the end user to save a few pennies in design cost.