Author Topic: Choosing the best capacitor for switch mode power supply  (Read 5227 times)

0 Members and 2 Guests are viewing this topic.

Offline arildj78Topic starter

  • Regular Contributor
  • *
  • Posts: 67
  • Country: no
Choosing the best capacitor for switch mode power supply
« on: September 04, 2019, 08:40:08 am »
I'm designing a SMPS and for bulk capacitance I'm considering a bank of 8 to 10 22µF ceramic caps. This is to achieve low ESR which in turn may keep the ripple low.

The switching frequency is 1MHz, and attached is a plot of capacitor Z vs frequency from the cap datasheet. Now, is it good or bad that the part seems resonant at 1MHz?

Arild
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21226
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Choosing the best capacitor for switch mode power supply
« Reply #1 on: September 04, 2019, 08:49:56 am »
This is a very complex subject to get right.

For starters, make sure you understand the consequences of the C-vs-Vdc spec. Continue by considering failure modes.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Choosing the best capacitor for switch mode power supply
« Reply #2 on: September 04, 2019, 10:39:19 am »
It's a good and bad thing, though mostly bad. The capacitor will look like a dead short* over a very narrow frequency centered around its series resonance frequency, and if the switching frequency coincides with such any reflected ripple will be all but eliminated. However, this may cause way more ripple current to flow through the capacitor than what it is rated for.

Ideally you want to use capacitors which have an SRF higher than the switching frequency, but at 1MHz that is going to be hard to do, so a combination of elkos and ceramics or film types is your next best bet. The elkos provide bulk filtering of low frequency ripple (from load changes and the inevitable rolloff of the converter bandwidth well below fsw) and damping for the high-frequency filtering provided by the smaller value/package size ceramic/film caps.

* - actually, the impedance will drop to the ESR, but as that is vanishingly low for an MLCC you can treat it as a dead short.
 
The following users thanked this post: arildj78

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 4033
  • Country: us
Re: Choosing the best capacitor for switch mode power supply
« Reply #3 on: September 04, 2019, 05:24:20 pm »
You also can't trust that resonance.  The real capacitor is approximately a series LCR.  L and r don't vary too much but C has a huge variation with temperature and especially voltage which will shift the resonance to higher frequency.
 
The following users thanked this post: TimNJ

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Choosing the best capacitor for switch mode power supply
« Reply #4 on: September 04, 2019, 07:11:28 pm »
It's a good and bad thing, though mostly bad. The capacitor will look like a dead short* over a very narrow frequency centered around its series resonance frequency, and if the switching frequency coincides with such any reflected ripple will be all but eliminated. However, this may cause way more ripple current to flow through the capacitor than what it is rated for.

Rest assured the AC current flowing in a switching converter is defined by the switching and the inductor, and not the capacitor. :)


Quote
Ideally you want to use capacitors which have an SRF higher than the switching frequency, but at 1MHz that is going to be hard to do, so a combination of elkos and ceramics or film types is your next best bet. The elkos provide bulk filtering of low frequency ripple (from load changes and the inevitable rolloff of the converter bandwidth well below fsw) and damping for the high-frequency filtering provided by the smaller value/package size ceramic/film caps.

* - actually, the impedance will drop to the ESR, but as that is vanishingly low for an MLCC you can treat it as a dead short.

The general strategy is this: treat the filter as an RLC network.  Consider the Q factor of each stage, and introduce resistance as needed to avoid peaks in impedance or transmission.

OP: there's nothing wrong with ceramic caps, but using them for bulk tends to be expensive, and you should be aware of their pitfalls: namely, DC bias, tempco, aging, and sensitivity to cracking.  (For these reasons, you would typically choose an X7R dielectric, check the datasheet to make sure you have enough C at DC bias, and use 1210, maybe 1812, or smaller, chips.  Or THT or leadframe parts, but those are even more expensive, so probably not.)

Where ripple reduction is key, an LC filter stage is far more effective than bulk capacitance.  1MHz shouldn't need very much filtering, but we have no sense of scale (is this at 1W or 1kW?).  Obviously you need enough capacitor to handle the ripple current; any more beyond that, may be reasonable to provide with aluminum polymer or electrolytic types.  (Not that polymers are much cheaper, but they are available in much larger values.  Both do suffer from reduced lifetime at high temperature, something to consider.  Tantalum and ceramic are the alternatives, and tantalum may be undesirable because of its poor handling of surge current (tendency to deflagrate) or conflict mineral use.  It's not clear if you've considered other types, so this fills out the list just in case.)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Choosing the best capacitor for switch mode power supply
« Reply #5 on: September 04, 2019, 07:39:06 pm »
It's a good and bad thing, though mostly bad. The capacitor will look like a dead short* over a very narrow frequency centered around its series resonance frequency, and if the switching frequency coincides with such any reflected ripple will be all but eliminated. However, this may cause way more ripple current to flow through the capacitor than what it is rated for.

Rest assured the AC current flowing in a switching converter is defined by the switching and the inductor, and not the capacitor. :)

Right, but the OP wants to use a bank of 8-10 22uF caps in parallel, and the likelihood that they all end up at exactly the same SRF is vanishingly small, and because they will have a high Q - being MLCCs, after all - even a slight variation in SRF will radically shift current from one to another.

That said, I didn't exactly clarify this to be the case in the response I penned at 6:30AM...  :P

 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Choosing the best capacitor for switch mode power supply
« Reply #6 on: September 04, 2019, 07:53:02 pm »
Ah, sharing between caps, sure.

Though -- depending on how they're wired, of course -- it tends to be that there's more stray inductance in traces than in the package itself, and the inductance between capacitors is as large as each one.

That is, draw the ladder network with each shunt branch being ESL + C, and each series branch being the L_stray between caps: L_stray ~= ESL, so the resonant circuits are very closely coupled if nothing else, pushing the resonant peaks far away from 1/sqrt(LC) (just as an overcoupled resonant transformer has two resonant peaks).

The ESLs do tend to be well enough matched compared to the Q, that they all tend to act in parallel, when L_stray is small enough to ignore of course.  That is, if ESL = 3nH, ESR = 10mΩ and C = 10uF, then sqrt(L/C) ~= 17mΩ, so the Q is a paltry 1.7.

Whereas to have one hog all the current at a nasty resonance, you'd need a Q of, well, about 10 to get a 10:1 hogging ratio.  So the large C value helps in this case, but ESL needs to be kept small as you can see.

Conversely, when L_stray is large, current tends to concentrate in the nearest capacitor, because the ripple is simply being filtered out as you go along the network.

All good reasons to use a tight layout -- group the caps together in a row, leaving enough space for vias between caps.  Connect the supply terminals with a polygon (minimizing L_stray), and connect the ground terminals with many vias (for caps of this size, two vias per pad should be fine). :-+

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline arildj78Topic starter

  • Regular Contributor
  • *
  • Posts: 67
  • Country: no
Re: Choosing the best capacitor for switch mode power supply
« Reply #7 on: September 04, 2019, 08:40:27 pm »
I'm designing around a TPS61178 from TI. Input is 3.7V and output is 10V@740mA with additonal 9A for 50µs at 1Hz. The average should be around 12W.

The power supply is for LED lighting but it is mounted close to a high sensitivity, low frequency receiver, so it is crucial to keep the noise at a minimum.

Quote from: TPS61178 Datasheet
9.2.4.2 Selecting the Output Capacitors
The output capacitor is mainly selected to meet the requirements at load transient or steady state. Then the loop
is compensated for the output capacitor selected. The output ripple voltage is related to the equivalent series
resistance (ESR) of the capacitor and its capacitance

As I understood it, choosing multiple ceramic caps would help me keep the ESR down.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Choosing the best capacitor for switch mode power supply
« Reply #8 on: September 04, 2019, 09:01:23 pm »
The switching frequency is 1MHz, and attached is a plot of capacitor Z vs frequency from the cap datasheet. Now, is it good or bad that the part seems resonant at 1MHz?

It does not matter; a low impedance is a low impedance in this case.

The same applies to decoupling capacitors which always have a self resonate frequency lower than some of the switching harmonics.
 

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 7521
  • Country: ca
  • Non-expert
Re: Choosing the best capacitor for switch mode power supply
« Reply #9 on: September 04, 2019, 09:34:17 pm »
This is a very complex subject to get right.

For starters, make sure you understand the consequences of the C-vs-Vdc spec. Continue by considering failure modes.

This is probably one of the most important, considering the parts out there. Some ridiculously "optimized" caps have less capacitance at 50% voltage rating than ones with a much lower uF rating :palm: Plus they cost a fortune.

Hell look at OPs part: https://product.tdk.com/info/en/documents/chara_sheet/C3216JB1E226M160AB.pdf
Using it at 10V means its a 7uF capacitor, so you are paying ~$5 for 70uF. What a joke.

I think you could be better off with something like a solid electro in parallel with some ceramics (and LC filter as mentioned above).


I'm designing around a TPS61178 from TI. Input is 3.7V and output is 10V@740mA with additonal 9A for 50µs at 1Hz. The average should be around 12W.

The power supply is for LED lighting but it is mounted close to a high sensitivity, low frequency receiver, so it is crucial to keep the noise at a minimum.

Quote from: TPS61178 Datasheet
9.2.4.2 Selecting the Output Capacitors
The output capacitor is mainly selected to meet the requirements at load transient or steady state. Then the loop
is compensated for the output capacitor selected. The output ripple voltage is related to the equivalent series
resistance (ESR) of the capacitor and its capacitance

As I understood it, choosing multiple ceramic caps would help me keep the ESR down.

Simulate it, see for yourself what works best.
But low ripple is not necessarily going to be the lowest noise/EMI, frequency/switching speed/etc might be important as well.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Choosing the best capacitor for switch mode power supply
« Reply #10 on: September 04, 2019, 10:33:58 pm »
Heh, nice, LED strobe.  That explains the bulk!

Yeah, don't worry about the ESR, "good enough" is just whatever you need for the LEDs, not even the switcher.  Again, if it's making a lot of noise, you'll most likely need filtering anyway.  And with all those caps, you have, well let's see, about nine opportunities to make one heck of a strong CLCLC... filter. ;) (Mind, you'd use quite small L in this case, to keep the filter impedance sqrt(L/C) low.)

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 2077
  • Country: us
    • The Messy Basement
Re: Choosing the best capacitor for switch mode power supply
« Reply #11 on: September 05, 2019, 12:08:00 am »
The numbers may be academic until you figure in trace length and the size of the loop.
 
The following users thanked this post: Someone

Offline arildj78Topic starter

  • Regular Contributor
  • *
  • Posts: 67
  • Country: no
Re: Choosing the best capacitor for switch mode power supply
« Reply #12 on: September 05, 2019, 10:55:42 am »
If I'm adding an inductor in series with the output (as in an LC filter), wouldn't this contribute a lot to the emitted magnetic field? Also, would such an LC filter need me to adjust components in the SMPS such as the compensation network?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Choosing the best capacitor for switch mode power supply
« Reply #13 on: September 05, 2019, 11:34:34 am »
Inductors are small, and often shielded.  No problem.  If you wanted to use planar magnetics (PCB traces as inductors, with or without cores), that would tend to have more external field, and better shielding may be a concern.

Indeed it may affect the compensation.  You can do a few things:
- Keep the cutoff relatively high (only somewhat below Fsw) so the filter still tends to act as one big capacitor as far as the control loop is concerned.
- Draw AC feedback from the front of the filter, and DC feedback from the output (basically, regarding the voltage sense divider resistors: split the top one in half, and connect a relatively large capacitor (10nF+?) from the front of the filter, to the new resistor midpoint).  Or really, since your load is pulsed, AC regulation doesn't really matter (there's no way the regulator can keep up with the load anyway, during a pulse), so you can just take the divider from the front in the first place.
- Tweak the loop components for best results.  These are typically a bit experimental, anyway -- it's hard to solve for these values in a first pass, but easy to dial in with an oscilloscope.

And again, yeah, definitely make sure the filter is well damped.  As noted, you don't have much room for inductance to remain reasonably well damped from capacitor ESR alone, so you'll probably need to add some ESR, either as a chip resistor on some of the capacitors, or using a selected ESR+C value of another chemistry.

I did a project last year, with similar ratings (well, at lower output voltage), but continuous output, and with a low noise specification.  I used a 5 pole filter, using a mix of ceramic and polymer capacitors, and achieved ~0.1mV output noise over a wide bandwidth (not just the 20MHz bandwidth that all the PS mfg's use).  At this level, a carefully considered layout is as important as the filter order and values are.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf