"104" caps testing as 50-60 nF and their use as decoupling caps

Question

I have a bag of capacitors labeled "104." (I don't recall their origin; probably an old AliExpress order.) However, they all test, on both my multimeter (a UNI-T UT61E) and a little microcontroller-based tester, as between 50 and 65 nF. I have another (clearly different in both colour and shape) "104" cap from another source that does test at around 93 nF, which makes me believe that I am testing them properly. (I do short the leads for a moment before checking the value.) But perhaps I'm not; am I missing anything here?

In case it helps, here's a picture of the caps in question: the two brown ones on the left are the seemingly 56 nF caps (front and back), and the one on the right is the actually-100 nF cap.

If I am measuring the value right, are these caps damaged in some way or just mislabeled. If the latter, is this sort of mislabeling common when buying jellybean parts?

I no doubt bought these as decoupling caps for things like old 8-bit CPUs and their peripheral chips. What should I do with these?

• Use a pair of them at each decoupling location to get the seemingly-standard 0.1 μf decoupling value.
• Just use a single one because 50-60 nF will work fine? (They do seem effective at cleaning up, e.g. a 1 MHz waveforms when I check on my oscilloscope.)
• Not use them at all because they're too dodgy?

Answer 1

I do short the leads for a moment before checking the value

Right start

---

If it is a breadboard PCB, i would put one or two in parallel and still go with it for a low voltage application 5V, 3,3 V.

Always buy components from reputed shops/sites.

1. The tolerances will be +/- 20% for the general purpose capacitors
2. for a decoupling application, the exact 100nF is not a must requirement
3. if it is for a RF application or a safety application, you would not have even considered using the capacitors without the datasheet

Answer 2

Readings that low - 50% of the marked value? Dodgy caps. Not worth using frankly speaking, unless you don't care so much about the value.

Something to be aware of. Ceramic caps have an issue called bias effect: the greater the DC bias, the lower the capacitance. The effect can be considerable, depending on the size and voltage rating of the cap. If your meter uses a large DC bias that may be part of why you get the low reading.

More about bias effect here: https://www.eeworldonline.com/the-effect-of-dc-bias-on-mlcc-class-2-capacitors/

Be aware of the bias effect when you use ceramics and plan accordingly.