14
\$\begingroup\$

I'm using a few Case R tantalum capacitors (10uF 6.3v) as bulk decoupling on the 3.3v power of a 100mbps Ethernet phy. I am also using 0.1uF ceramics closer to the pins.

Ethernet phy PCB

As usual, I'm terribly pushed for space on the PCB, and so I would like to replace them with 0603 size caps. Problem is that they're only rated for 4v. Normally I would always rate my capacitors for twice the voltage that they're ever going to see.

Is it likely to be a problem if I use a 4v capacitor on a regulated 3.3v line?

\$\endgroup\$
  • \$\begingroup\$ I take it you mean 10uF instead of 10uV? Out of curiosity, why not use ceramic capacitors? Digikey has a variety rated for 6.3V or 10V in the 0603 and even a few in 0402 sizes. \$\endgroup\$ – helloworld922 Oct 8 '12 at 23:02
  • \$\begingroup\$ @helloworld922 - Yes I did! It's funny. I was looking at the uV and thinking 'something looks wrong about that', but I couldn't put my finger on it. \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 8:16
  • \$\begingroup\$ Wow, that 3D is great! \$\endgroup\$ – abdullah kahraman Oct 11 '12 at 18:33
  • \$\begingroup\$ @abdullahkahraman - Yes, the joys of Altium. \$\endgroup\$ – Rocketmagnet Oct 11 '12 at 19:29
11
\$\begingroup\$

Using tantalum caps for decoupling is silly with today's options. 1 µF 6 V ceramic caps in 0603 package are cheap and readily available. That would be better than 10 µF for decoupling anyway. A 10 µF at the power supply connection points makes sense, but not for decoupling. Take a look at the impedance plots of 100 nF, 1 µF, and 10 µF ceramic caps in the datasheets of any of the reputable manufacturers.

\$\endgroup\$
  • \$\begingroup\$ Thanks. That's good if I don't need to use tantalums. The only reason I was using them was because the ET1200 connected to it says tantalums in its datasheet. Does this mean I could just replace all of the tantalums with ceramics? (Except where linear regulators specifically require them). \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 8:26
  • 2
    \$\begingroup\$ @Rocket: Old datasheets sometimes say tantalum because back then that was the only way to get reasonable frequency response at larger capacitances like 1 uF and above. Nowadays ceramic can do this better up to a few 10s of uF for most cases. Tantalum caps are pretty much obsolete. No regulator requires a tantalum cap, although some old ones require a minimum ESR, which tantalum would provide natively. Today there are plenty of regulators specified for 0 ESR output caps, so I would just use one of them. \$\endgroup\$ – Olin Lathrop Oct 9 '12 at 12:25
  • \$\begingroup\$ OK, I'll try it. \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 13:00
15
\$\begingroup\$

Be super careful about your caps specs! They might not be what they claim to be.

One problem with ceramic caps is that they loose capacitance as the voltage increases. For example, this X5R ceramic cap from Venkel is rated at 10 uF and 6.3v. The problem is that at 6.3v, the actual capacitance goes down to only 2 uF! That's an 80% drop! Even at 2.0v, you are missing 30% of the rated value.

This is documented here, but it is in no way limited to Venkel. X7R, X5R, and probably other ceramic caps suffer from this to some extent. COG/NPO does not seem to have this issue. I should point out that this problem was not listed in the overall capacitor datasheet, but only in a supplemental "technical data" document that was not located in the same place on their web page.

Your basic question, however, is "how much should I derate" capacitors. Of course if you ask 10 EE's about this then you will probably get 15 different answers. Here is my very rough rules of thumb or guidelines.

  • Aluminum Electrolytic caps will be derated to at least 50% if the capacitance is critical. Meaning, I will use a 25v cap on a 12.5v rail. If the capacitance is not critical then I will allow for less voltage margin.

  • Tantalum caps will be derated to at least 50% on most rails, reguardless of use.

  • Ceramic caps above 10v will be derated to 50%. As the voltage drops below 10v, I will allow for less margin. I will run a 4.6v cap on a 3.3v rail, for example. But keep in mind what I've already said about ceramic caps.

  • The lighter the stress on a cap the less margin I will allow. For example, if a signal only very occasionally goes to 90v but is mostly below 50v then I might use a 100v ceramic cap. Temperature, voltage, current, and ESL/ESR effects all play into this. A cap with lots of stress will get much more of a margin.

  • Derating less than 50% only happens if there are other reasons to do so. If cost, size, and other factors do not come into play then I always derate at least 50%.

  • If a circuit is well behaved and issues of voltage spikes or other bad behavior is not an issue then I might leave less margin.

But these are only rough rules of thumb. You have to consider each case separately and weight the pros and cons.

Update:

Here is some documentation from AVX. It is on page 3, under the heading of "DC Bias Dependency". Note that their voltage vs capacitance curves for X7R are much "nicer" than what Venkel has-- so if the cap value is really important then do the research and get the cap that you intended.

\$\endgroup\$
  • \$\begingroup\$ Thanks David. I think that's pretty much settled then. I won't go down to 4v tantalums. I might consider ceramics instead. \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 8:29
7
\$\begingroup\$

In additon to Olin's comments re size - note well that a Tantalum capacitor on a circuit with any sort of "power" present is a total disaster waiting to happen.

Tantalum capacitors are EXCEEDINGLY prone to insulation layer punch through at voltages only slightly in excess of rated voltage, and they are NOT self healing. Once the layer is breached the capacitor will dump available energy and self destruct. Normal failure mode is a hard metallic short. Optional alongh the way are smoke, smell, flame , sound and explosion. I have seen heard and smelt all these in a single exciting event on one occasion.

A very very short voltage spike on a power rail which exceeds voltage rating can puncture the insulation layer and then allow the power rail energy to finish the task.

Ceramic: Note that temperature effect varies with grade. Also mechanical microphony is worse on wide temperatre variation grades. Not usually an issue for power decoupling.

A low grade ceramic (poor temperature tolerance, so also poor mechanically) at the input to a regulator can "ring" when a step voltage is applied (such as at power turn on) and can destroy the regulator. This is unusual and easily guarded against but needs to be known about.

If ceramic does not meet your needs for some reason look at solid Aluminum capacitors (ie not Al electrolytic). They are competitive with Tantalum in capability size and cost but do not have Tantalums fatal failure mode.

\$\endgroup\$
  • \$\begingroup\$ See also en.wikipedia.org/wiki/… I like how they describe it's self destruction as being similar to a thermite reaction heh \$\endgroup\$ – Earlz Oct 9 '12 at 4:25
  • \$\begingroup\$ I've seen that happen with tantalums too many times, usually just sitting there glowing hot. One advantage this has is that they can act as sensitive sacrificial voltage fuses, tripping out the power supply, and protecting other, harder to replace chips. \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 8:21
  • 2
    \$\begingroup\$ Also, if tantalums are so bad around power, why do some linear regulators specifically require them? E.G. MIC5205 \$\endgroup\$ – Rocketmagnet Oct 9 '12 at 8:27
  • 2
    \$\begingroup\$ @Rocketmagnet - (1) People like to live dangerously. ? (2) The applicatioj engineer is a recent college graduate ans not on talking terms withe the design team? (3) Likes pyrotechnics. (4) ...??? \$\endgroup\$ – Russell McMahon Oct 9 '12 at 9:06
1
\$\begingroup\$

The rule on tantalums is to avoid them unless there is good current limiting. Or at least derate the voltage by at least 3:1 (some people say 4:1), not always possible. I've used them a lot on power supplies on the output of a small regulator but not on the input.

\$\endgroup\$
  • 1
    \$\begingroup\$ What do you mean by good current limiting? I would readily understand voltage limiting. But I've never heard of current limiting in conjunction with tantalum capacitors. \$\endgroup\$ – Nick Alexeev Nov 22 '14 at 3:50
  • \$\begingroup\$ I think he simply means a fuse (or a more elaborate protection like a current "diode" etc.) simply because tantalum's failure mode is short-circuit. \$\endgroup\$ – Fizz Nov 28 '14 at 3:57

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.