3
\$\begingroup\$

My focus is on surface-mount ceramic capacitors. The obvious failure extremes are open and short. Other than these extremes, can the failure be that the effective value of the capacitor is either higher or lower than the spec? Can this failure be intermittent in that it disappears when the technician eventually comes around to examining the circuit? I don't mean intermittent in the sense that it lasts seconds or less but that the failure lasts minutes but only at random time of day and random day of week.

If yes then I'm curious as to what the failure mechanism could be. Also, is there a test to confirm such a bad installed capacitor?

\$\endgroup\$
  • \$\begingroup\$ Tests at elevated temperature sometimes flush out failures like this. \$\endgroup\$ – Nick Alexeev Jun 4 '18 at 1:28
  • \$\begingroup\$ I have seen on very rare occasion a board with a charred hole in it where a SMD capacitor used to be. The 3.3 volt and 5 volt feeds can supply lots of current on some boards. It had no pattern other than the capacitor manufacture was shipping parts with latent defects. \$\endgroup\$ – Sparky256 Jun 4 '18 at 4:10
  • 1
    \$\begingroup\$ Partial/intermittent failure? Oh yas. Think cold solder joints. Connection is made depending on temperature, vibration, relative humidity, and for all I know phases of the moon. Note that in this case the cap itself is still good - but the failures are the same as if it were bad. \$\endgroup\$ – WhatRoughBeast Jun 4 '18 at 4:29
1
\$\begingroup\$

3 possible conditions I can think of that could cause psudo-random appearing intermittent issues.

  1. Temperature. Ambient and operational temperatures can cause dramatic changes in performance depending on the specific component specs and tolerances.
  2. Physical stress. Bending of the mounting surface can cause deformation and therefore changes in capacitance. They usually don't have much give in them before they crack and fail open/short, but a light flex or moderate pressure during operation can cause unpredictable results.
  3. Piezoelectric Effects. Certain electrical conditions can cause multilayer ceramic capacitors to vibrate, and any vibration can case them to generate a voltage back into the circuit. If your circuit only enters a state where it causes a piezoelectric vibration under certain conditions, or the usage conditions create random bumps and vibrations by the user, then you could have undesired and intermittent issues.
\$\endgroup\$
1
\$\begingroup\$

There are bunch of different failure modes that basically boil down to cracks in the ceramic (including micro-cracks) or termination failures. Such things can certainly be intermittent and sensitive to flexing of the PCB or other seemingly random factors.

Here is an AVX document with a fairly detailed description of various causes.

Not mentioned in the AVX document, but some makers have "soft" terminations that are resistant to some sources of problems.

\$\endgroup\$
1
\$\begingroup\$

If the capacitance changes are significant, it's likely one of the endcaps is broken off. This can happen if the board flexes with the cap attached. The result is that both sides of the cap are materially separate, but held in place mechanically with enough precision to allow the cap to work. Subsequently, slight twisting or bending will result in total or partial failure of the component.

If you unsolder such a component, it will fall apart immediately.

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

SMT ceramic capacitors are very sensitive to strain. Class II dielectrics (everything except NP0) are also piezoelectric. Stresses on the board, including thermal stress can change the capacitance significantly.

\$\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.