3
\$\begingroup\$

In the product I analyse (an optical fork sensor, rated 10V-35V), there is a sizewise big capacitor between ground and chassis.

caliper measurement

I measured its value with an LCR meter, it is 60nF. I also broke one by accident, which revealed a liquid from inside. Looking at its size and considering the liquid inside, I think it is a film capacitor.

Questions:

1) Why is there a capacitor between ground and chassis? For EMI purposes? (Note: the cables used for this product are usually not shielded - chassis is floating if there is no capacitor)

2) Why is the capacitor a metallized film type? (if it is) It seemed to me like a ceramic capacitor may also have been used in this scenario - it has a smaller footprint.

\$\endgroup\$
1
  • \$\begingroup\$ it looks like a short to high frequencies, but won't pass DC if something pops loose. \$\endgroup\$
    – dandavis
    Commented Feb 8, 2019 at 18:54

1 Answer 1

Reset to default
5
\$\begingroup\$

The sensor is a PNP output with 2x potentiometers, I presume as coarse and fine with act as some optical comparator with a load resistor to 0V for measuring an optical gap. In noisy environments a metallize film cap shunts Stray RF noise between 0Vdc and the metal body to improve SNR from injecting noise with your hand.

The hands can inject high impedance line noise picked up easily by the body. thus cap shunts this line noise to 0Vdc.

Film is often the best choice for very low ESR and big enough to reducing errors from handling.

Zc=1/2piCf

\$\endgroup\$
2
  • 2
    \$\begingroup\$ But the product is not connected to earth ground. Your answer sparked a hypotheses though: because chassis is not grounded, RF noise may be coupled to other elements on PCB (like a photodiode output, for example). To prevent this, the capacitor forms a short circuit to signal ground for RF signals. Could this be correct? \$\endgroup\$
    – C K
    Commented Feb 8, 2019 at 7:00
  • 1
    \$\begingroup\$ the attenuation can be called a “short circuit” relative the high stray impedance from body to power line. But the formula tells you it is not 0 Ohms. \$\endgroup\$
    – D.A.S.
    Commented Feb 8, 2019 at 7:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.