Timeline for Termination capacitor value & grounding
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 10, 2023 at 9:17 | comment | added | Tim Williams | Not surprised there's no (or at least nebulous) difference on the linked answer. It wouldn't affect radiated, the signal source is already balanced. It does however affect unbalanced conditions, like induced current on the cable (CM) which is therefore unbalanced between pairs (two otherwise open-circuit pairs when connected to 10/100). Termination impedance doesn't matter much because it just needs to be dampened well enough to absorb energy that would otherwise resonate between pairs. The real question is: given the CMCs inside the magnetics, what is the "correct" or ideal value? | |
| Oct 10, 2023 at 8:52 | comment | added | Rohat Kılıç | @Freshman it's actually out of your question's scope, but here it is: You normally don't have to connect them. But, as a counter example, ATX PSU's have their DC ground tied to chassis ground, and this makes the PHY-side on the MoBo earth-referenced. If your PHY and cable grounds are isolated then the value of the capacitance should be determined by the tests and requirements (e.g. hi-pot tests, max allowed leakage, etc). Usually, 1~2 nF / 2 kV ceramic between the grounds is fine. That cap also provides a low-res shunt for RF currents and EMI (e.g. 1n @ 100 MHz shows 1.6 Ohms resistance). | |
| Oct 10, 2023 at 8:39 | comment | added | Freshman | Thank you very much for the answer. 1. Do you have any advice on connecting the two grounds/references,i.e., the cable side and the PHY side? If so, the values and the reason, please? | |
| Oct 10, 2023 at 8:35 | history | edited | Rohat Kılıç | CC BY-SA 4.0 |
added 28 characters in body
|
| Oct 10, 2023 at 8:29 | history | answered | Rohat Kılıç | CC BY-SA 4.0 |