Electronics > Projects, Designs, and Technical Stuff

Residual current device

(1/2) > >>

Red_Micro:
I'm just checking this device from ABB: https://library.e.abb.com/public/56f56bae475a5325c1257d50002d806f/2CSC444001B0201.pdf

The document says:


--- Quote ---High accuracy of measurement
Both RD3 range and RCQ020 devices perform the RMS measure of earth leakage current. This means that the residual current devices measure any type of signal, calculating the actual RMS value, averaged according to the frequency filtering.
--- End quote ---

Since this is an earth leakage device, it must respond extremely quick. But I'm wondering if this can be quick enough as it does RMS. Usually, if one wants to capture both 50Hz and 60Hz, we would want to capture at least 5/6 cycles (100 ms) of the signal to get good RMS. Isn't this too slow for this kind of protective?

uer166:
Define "extremely quick". The required timing ranges massively depending on standard, anywhere from 20ms to 10 seconds. Consider that even at 20ms one can measure the RMS current, maybe not accurately, but at those high leakage currents levels nobody cares.

Some other food for thought is Type B RCCBs that have to respond to DC, 60Hz AC, and even high-frequency 2kHz AC leakage.

Gyro:
The required trip time for a Type B RCD at 5 times rated leakage current (i.e. 150mA) is a maximum of 40ms, which is two mains cycles at 50Hz. In practice, they trip in about half that. Tripping at 30mA is typically pretty similar. That doesn't leave much time for RMS conversion.

Industrial stuff is no doubt different - from the look of the brochure you linked they seem to have an adjustment for instantaneous to 5s.

Red_Micro:

--- Quote from: Gyro on September 10, 2020, 11:15:35 am ---The required trip time for a Type B RCD at 5 times rated leakage current (i.e. 150mA) is a maximum of 40ms, which is two mains cycles at 50Hz. In practice, they trip in about half that. Tripping at 30mA is typically pretty similar. That doesn't leave much time for RMS conversion.

Industrial stuff is no doubt different - from the look of the brochure you linked they seem to have an adjustment for instantaneous to 5s.

--- End quote ---

Instantaneous often mean ~30 ms or less in type A&AC  RCDs. That's why I wonder how they perform RMS for both frequencies in that short time.

Gyro:
I assume at that timescale it's an a mean/peak approximation. A fair chunk of that time is going to be taken up mechanically opening the contacts. They probably get into more clever RMS stuff for the longer trip times.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod