Lightning protection for signal loop

[Ron M]

i would like to add lightning protection to my Automower 450X. The ground loop, which consists of five wires that run around the perimeter of the property and subdivide it, has to be disconnected from the transmitter/charger if there is the threat of an electrical storm. And of course the power disconnected. I would like to build a switch system that Disconnects and grounds the loop wires and cuts the power to the transmitter/charger.

With my very basic knowledge, I have an idea but I don't know if it's any good and I don't know what components would be best.

My idea is to use double throw relays to switch the signal wires to earth ground when the coil is not energized (NC) and to the appropriate transmitter terminal when the coil is energized (NO). A separate relay would be used to switch off the power to the transmitter.

In my case, a z-wave switch would be used to actuate the relay coils via a web enabled controller. This would provide the ability to control the system remotely and integrate with IFTTT Maker for automatic, weather specific operation.

I would greatly appreciate input from others on the idea and how it might be implemented to provide the best protection.

This is similar to the pet containment loop signal, for which there are lightening arresters available, except they only use two wires.

 

[Tony Stewart]

A line filter with an MOV is the usual protection for raising the immunity to lightning from line transients.

Next protect the TX output.

If you had a power interruption during a lightning storm, your system may still be connected. I would suggest the RF connects via miniature relay with NO contact and activated if power is available a low Voltage inside the unit could be used to enable the output and when powered off the DPDT could connect the external wires to gnd. or to transmitter when power is turned on. Thus only one toggle control.

So 2 wires to POLES and Tx to NO and GND to NC.

The ground should not be shared by the equipment but rather to a nearby ground rod to avoid gnd shift induced by lightning on the equipment.

 

[MikeMl]

After traveling several miles (cloud to ground or vice versa), do you think that a lightning bolt cares a whit about jumping across a 1/8" gap???

 

[Tony Stewart]

After traveling several miles (cloud to ground or vice versa), do you think that a lightning bolt cares a whit about jumping across a 1/8" gap???

Haha. I've seen experienced senior designers use MOV's and gas tube protection in low AC powered RF repeaters. One came back with molten crater in the middle of the AC -DC section of the PCB. He wasn't aware that gas tubes are negative resistance clamps like crowbar SCR's. I had to prove it to him in the lab with an ignition coil, and 1ook resistor with a tiny cap with a low voltage , current limited in parallel to make it glow.

Series current limiter is a must. Line filters help,attenuate the peak.
I've verified most old power meters are gapped to 6kV yet most consumer products are protected to 2 to 3 kV. Meanwhile I have 14Vdc perimeter wire to power various garden LED's that blew the supply in the basement for a very close lightning strike recently.

 

[alec_t]

I think blade switches, with their wider gap, would offer better protection than a conventional relay.

 

[Ron M]

This a the sacrificial product currently available for pet containment systems. There are a couple models that look about the same. They cost about $25.

This is typically what they look like when spent. From the product reviews they seem to work very well.

I'm surprised that there is not a similar product for the Automower systems, which have from 3 to 5 wires.

 

[Tony Stewart]

I think blade switches, with their wider gap, would offer better protection than a conventional relay.

I was thinking that the relay NC contacts should be closed in the power off condition and shunted to earth gnd so the voltage drop to the transmitter would be minimal on the NO contacts

 

[Ron M]

I was thinking that the relay NC contacts should be closed in the power off condition and shunted to earth gnd so the voltage drop to the transmitter would be minimal on the NO contacts

Would varistors be used in addition on either the loop side and/or the NO side between the transmitter?

 

[Tony Stewart]

I think the capacitance is too high for the RF. But suitable TVS's might be possible but if the Tx can handle no load on startup, for 50ms or so , it shouldn't be necessary. I'm not familiar with the output power and transients of open circuit during shutdown either and have not tested this. But If I saw more details I could tell.

 

[Misterbenn]

I was thinking that the relay NC contacts should be closed in the power off condition and shunted to earth gnd so the voltage drop to the transmitter would be minimal on the NO contacts

Wont matter if the contacts are closed or not, unless its a huge relay with a very significant ground connection in the order or 16mm^2 it will just vaporise the contacts and everything will see a significant potential.

For aerospace the standard practice is to put a series filter impedance in the path followed by transient voltage suppression diodes.

 

[Tony Stewart]

Wont matter if the contacts are closed or not, unless its a huge relay with a very significant ground connection in the order or 16mm^2 it will just vaporise the contacts and everything will see a significant potential.

For aerospace the standard practice is to put a series filter impedance in the path followed by transient voltage suppression diodes.

This is an RF loop antenna. A line filter would block the signal.
The relay shunts stray lightning pulses effectively to a good ground rod.
Normally a tree or a higher object gets hit.

Perhaps a series tuned LC filter in the Loop could improve a direct loop hit immunity, but a line filter is only good for direct hits to the line not the antenna.

 

[Ron M]

Could a separate ground rod be used for the loop antenna only, without being bonded to the main service ground?

 

[Tony Stewart]

Could a separate ground rod be used for the loop antenna only, without being bonded to the main service ground?

only for small grounded loop antenna is lightning interference suppressed significantly.
This is a large area antenna, unless you dont expect hits near the property.

 

[Ron M]

I am sorry but I didn't follow that explanation.

 

[Tony Stewart]

The antenna may be grounded internally already. Diverting that ground externally won't help as the loop will pick up significant levels on the signal side with a yard sized loop.

 

[Ron M]

I would still isolate the loop from the transmitter with the double throw relay and shunt to ground. Just wondering if this ground needs to be bonded to the service ground if all I am grounding is the loop. I think if I were grounding other line voltage equipment it would need to be bonded to the service ground. In my case the ground would only be for the loop when it is isolated from the transmitter.

 

[Ron M]

Wondering if it would be recommended to twist the 5 RF leads between the transmitter and relay 10 to 12 times per foot as is done with the pet containment loop when using a lightning arrester? I assume this somehow helps protect the transmitter when the arrester is fried. Any thoughts?

Thanks!