4
\$\begingroup\$

I am using this linear voltage regulator. The datasheet indicates the input and output values for the capacitance to use, 1uF and 10uF respectively.

Should these capacitors be or a particular type, or does it not matter?

\$\endgroup\$
  • \$\begingroup\$ This answer is related: electronics.stackexchange.com/a/426181/202270 \$\endgroup\$ – Edgar Brown Mar 16 at 23:44
  • 2
    \$\begingroup\$ I agree with the general consensus. Another option is to use ceramic caps but put a resistor in series to insure stability. Another option is to contact ST micro and just ask them. \$\endgroup\$ – mkeith Mar 17 at 0:44
6
\$\begingroup\$

It doesn't usually matter, but be aware that some linear regulators--the popular LM2940 series, for example--may be unstable if the output capacitor's ESR is too high or too low. As the datasheet for your regulator doesn't seem to say anything about that at a glance, it should be fine with any capacitors you pick, but see the edit below for a warning.

Non-polarized capacitors more than about a microfarad used to be rare and expensive, which is probably why the datasheet shows polarized capacitors being used. Today, you can get 10μF ceramic capacitors for less than $0.30 each.


Edit: As @ThePhoton points out, this regulator may be so old that multi-microfarad ceramic capacitors, with their inherent low ESR, may have been a far-off pipe dream to the engineers writing the datasheet. So this may still be unstable with too low an ESR on its output, so unless you want to test its stability under different operating conditions with the ceramic caps, it may be best to stick to aluminum electrolytics. After all, that's probably what the IC's designers had in mind.

\$\endgroup\$
  • 4
    \$\begingroup\$ The chip might just be so old that when they wrote the datasheet, they didn't consider the possibility that someone would want to use a low-ESR ceramic capacitor for such high values (1 and 10 uF). I'd stick with electrolytic unless I had time to experiment and make sure it stays stable with ceramic over all operating conditions (temperature, input voltage, etc). \$\endgroup\$ – The Photon Mar 16 at 23:40
  • 2
    \$\begingroup\$ That's a good point, @ThePhoton. Then again, MLCCs are relatively high ESR as ceramics go, and you can get pretty low ESR electrolytics--I'm not sure how they compare, but you do make a good point and I'll add a note to that effect in the answer. \$\endgroup\$ – Hearth Mar 17 at 0:10
1
\$\begingroup\$

The datasheet application circuit example schematic shows a 1 microfarad polarized capacitor on the input and a 10 microfarad polarized capacitor on the output. Since the values are in the 1 plus microfarad range and the capacitors are shown as polarized, I would guess that the manufacturer (ST) wants you to use electrolytic caps. I guess you could use a tantalum caps, but unless the datasheet specifies tantalum, that would be a needless expense.

The polarized caps shown on the datasheet circuit lead me to believe that electrolytic caps are what are intended. Very few ceramic caps are over 1 microfarad and very few are polarized.

\$\endgroup\$
  • 1
    \$\begingroup\$ Actually, ceramic capacitors of up to hundreds of μF are, while not common, certainly readily available. They're not terribly expensive, either. Ceramic capacitor technology has improved dramatically in the past decade or so. \$\endgroup\$ – Hearth Mar 16 at 23:21
  • 2
    \$\begingroup\$ I agree with Hearth that you're wrong to say 1 uF and up are rare as ceramics. But I'd still advise OP to stick with electrolytics since older regulator designs depend on a reasonably high ESR in the capacitor to maintain stability. If the datasheet doesn't promise the regulator is stable with low-ESR or ceramic output capacitors, it's not wise to assume it will be. \$\endgroup\$ – The Photon Mar 16 at 23:44
1
\$\begingroup\$

What matters most is a temperature stable ESR with a nominal value of 0.3 Ohms +-25% for good stability and C>22uF. This must be your search criteria for ganged output caps.

\$\endgroup\$
0
\$\begingroup\$

In contrary to the previous answers, it does matter.

The electrolytic cap is very good at storing a large amount of energy, but bad for high transient. Having large transients on an electrolytic cap will shorten its lifespan.

Ceramic capacitor store less energy but are good at coping with large transients.

That is why, a good implementation, as shown on the datasheet, is a large electrolytic cap that is there to store the energy, and a ceramic cap, as close as possible to the switching element, to take the switching transients.

If you only put an electrolytic cap, the lifetime of your circuit will be impacted, and only a ceramic cap, your circuit might be susceptible to issue when you have rapid power draws or input line unstable feed.

\$\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.