微波EDA网,见证研发工程师的成长!
首页 > 研发问答 > 微波和射频技术 > 电磁仿真讨论 > magnetic field from a relay

magnetic field from a relay

时间:03-22 整理:3721RD 点击:
hi, I have a cicuit that has a few sensors and a few 5V relays. each relay takes about 75 mA current and voltage from sensors are read with 16 bit ADC. whenever there is a toggle in relays there is a spike in sensors voltage. my question is how can I alleviate this by reducing electromagnetic interference. as far as I know magnetic field forms between the wire that has a current and its nearest return path(please confirm that). so should I thinking about creating a return path around the relay? thanks.

Some measures to be done :
- connect sensors to adc via differential lines
- check for adc input impedance and decrease if possible
- adc pcb part can be shielded
- sensor wires to connect can be shielded
- sensor signal can be filtered before analog conversion
- diverse relays from analog part far away
- check spike on power supply for relay and put spike suppressors
- design pcb as current loop return path for relay will not affect adc power supply
- use different power source for relay
- put filters to relay supply
- ...

try to identify noise intrusion point(s) in your product and of course post schematics before asking assistance .

I don't think you will be able to easily overcome this problem: on one hand you have 16-bit ADC (≈75μV resolution) switched by a relay contact, on the other hand there is the 75mA swithing current ..

On option will be to replace relays with CMOS switches, for example MAX322, CD4066, or others, under condition, that you can re-design PCBs to accomodate these changes ..

If you can't go this path, and I assume that you have diodes in parallel with RLYs winding, bypass caps and all of that stuff, then I would suggest that yo re-arange the way you read the ADCs: apply software delay(s) after you switch relay(s) and before you start AD conversion ..

Regards,
IanP

thanks both of you. IanP I would love to use CMOS switches but the current that passes through relay contact is about +/-1A(+/-15V). I also thought about using mosfet switches, but problem is that leakage current in off state should be less than 1nA and leakage of mosfets would be a problem.
PS: sensor voltages are read with a 4051 analog mux. the interfereing relays carry voltages that stimulate a circuit and sensors read various parameters after stimulation.

As the switching current is ≈1A/dc maybe part of your problem is contact arcing ..
With both AC and DC, contact arcing can be minimized with the addition of a "snubber" circuit (a capacitor and resistor wired in series) in parallel with the contact (see picture below) ..

Regards,
IanP

Copyright © 2017-2020 微波EDA网 版权所有

网站地图

Top