| Electronics > Projects, Designs, and Technical Stuff |
| "MLCCs ... do not require any voltage derating" |
| (1/5) > >> |
| 741:
Here is TDK's paper https://product.tdk.com/en/contact/faq/20_mlcc_voltage_strength.pdf That seems useful information, I think 6.3V is therefore fine for 5V, and that makes some cost difference for 10uF units. |
| wraper:
You should derate voltage for MLCC because of huge capacitance drop under voltage, not because of reliability issues. Likely there will be 10-25% of capacitance left for 10uF cap when used under rated voltage. |
| Rerouter:
To make things more complex, you really do need to look at the datasheet due to how the capacitance vs voltage curves change for every different series and manufacturer, sometime a 10V MLCC can be better than a 25V MLCC of similar size, because they then use a different method to allow for the higher voltage rating, which in turn makes the voltage curve tank much harder. |
| 741:
That is a very good point, applying to many MLCC types. For problematic ('ferroelectric') dielectrics at given 'rating' like 6.3V, I was wondering how much 'product range' variation there can be - do the dielectric, voltage rating, capacitance and size/shape all affect this issue? The reply just now by "Rerouter" shows it is indeed down to the individual datasheet. I'm looking at a Multicomp (Farnell/CPC) branded item, MCCA000593, 'X5R', CPC code CA08032. The 10uF capacitor is at the output of an MCP1623 switcher, giving 5V output. I would imagine it will spend most of the time at about 5V, so at least any voltage dependence is a fixed factor. There are no plots for voltage dependence I can see on the DS. I am committed to 0805 for now - maybe I should have assumed a 1206 size, that might be better "voltage rating for voltage rating"? |
| Rerouter:
Larger sizes with the same voltage and capacitance are "usually" better, however it comes down to exactly what dielectric material they use, Sadly there are multiple flavors of materials that still meet the specifications for X7R, which they use in different capacitance vs voltage groupings, so you may find the 0805 and the 1206 actually still use the same material, just with less layers. yet drop from a 10V 0805 to a 6.3V 1206 and suddenly its much better because they used a different composition to make up that 1206 X7R dielectric, as the density was now low enough that it made more sense to use a different material, Yes this stuff drives me mad, because they go out of there way to make it almost impossible to do parametric comparisons across manufacturers, even different series from the same manufacturer! |
| Navigation |
| Message Index |
| Next page |