Surge damage repeating

[oh6hfx]

I have installed cameras on a site 4 years ago. There has been several devices damaged during these years. Poe switches, cameras, wifi access points and injectors.
There are cameras in 3 buildings, and internet / lan connection to NVR is done with 2.4GHz wifi outdoor TP-link access points. In all other buildings everything has been working fine since installed, but one, a barn for sheeps is a problem place.

First I thought it was the thunder which was killing the devices, then I thought it is some kind of inducted ESD in outdoor access point. Two weeks ago, there was again something
gone broken, and I found it is the poe switch and camera. As here we don't have thunder this late of year, I just thought, well the switch was bad quality and I replace it with poe injector and put a new camera.
It worked well for 2 days and then it happened again. There had been just a normal power outage in power network, and this time everything in the camera system was damaged.
48V poe injector for camera, Dahua camera itself, 24V injector for wifi access point, and wifi access point of course.

I decided to replace the U/UTP cables with F/UTP and put an extra surge protector for access point, grounding that to buildings yellow-green safety ground. So far it has been working fine.

Later I opened the damaged devices. Injectors and camera are burnt bad! In access point I can not see any damage or black components, it even works with poe, just the lan connection does not work anymore.

The question is: where is the surge coming? Is it a voltage spike from AC network hitting the injectors and going via ethernet cables to other devices burning them also? In this one picture you can see cameras poe injector and it is burnt
from the side where LAN connectors are. It's just the same with wifi poe injector. In wifi poe injector even solder has been melted in LAN connector and I could pull it out with my fingers. In both devices the most bad place is around LAN connector, not poe connector.

 

[BORIStheBLADE]

Is the power for the barn fed from the same source of the building? Basically what I'm getting at is there a separate transformer for the barn?

Did you ever try or consider a UPS to clean up the power for the network hardware?

They have devices that can monitor A/C power and give you readings. Have you tried anything like that?

 

[oh6hfx]

Is the power for the barn fed from the same source of the building? Basically what I'm getting at is there a separate transformer for the barn?

Did you ever try or consider a UPS to clean up the power for the network hardware?

They have devices that can monitor A/C power and give you readings. Have you tried anything like that?

I'm not sure about the source, I was thinking the same that it might be separate as there has been big power consumers in the past. Nowadays just some lights and drinkwater-defroster.

I have been thinking about UPS too, if the power is too harsh. However, I have advised them to get electrician the check the AC network and groundings, because I'm thinking this kind of damage can not
be from ESD or inductive currents?

 

[BORIStheBLADE]

I'm not sure about the source, I was thinking the same that it might be separate as there has been big power consumers in the past. Nowadays just some lights and drinkwater-defroster.

I have been thinking about UPS too, if the power is too harsh. However, I have advised them to get electrician the check the AC network and groundings, because I'm thinking this kind of damage can not
be from ESD or inductive currents?

I dont have a lot of knowledge regarding ESD and inductive currents.

Another factor could be water damage and damaged wire (not visible to the naked eye)
underground in a pipe. Is any of the wiring ran underground?

 

[TVille]

A while back, 30 years ago, on my birthday no less, I had a lightning strike near my house. It came in over the power lines, went into the TV, jumped to the coax, to another building where it physically blew apart a coax splitter. Also got into the phone, and stereo, but didn't appear to jump from there. Point is, that it can easily jump through a device and out the other side. It is possible that it goes through and does not damage a particular piece, and takes out a weaker link down stream.

 

[user8963]

looks and sound like main voltage has contact to network cables..

 

[garycrist]

Can the problem be that the cams @ the sheep barn are at "local ground potential"? With such a long
run and everything is at "local ground potential", one may see a big "earth battery". The blue MOV looks ok,
so then the ground is now passed via the twisted pairs to case ground.

Have you measured how much voltage is on the ground line at the sheep barn? Is the "buck" caused by
equipment in the barn getting shut down or at start-up?

 

[Teken]

As it relates to equipment protection the first thing to understand is basic Ohms law. Once that is understood the next step is to validate the homes electrical system is properly functioning in all aspects. Whatever the POCO needs to do is on them but you can surely make it easier on yourself by offering some facts to them.

In all cases its their job to affirm the incoming line voltage is correct and the system is properly connected to the service entrance. Its also their responsibility to confirm the system is properly grounded. It does not mean they will be able to change any earth grounding issues that may be found.

Meaning if the earth ground is above 25 ohms its on you to correct it by adding more grounding rods in parallel.

Next, is moving inside of the building and validating the entire service panel is properly wired and all connections are tight and not corroded. This again leads to validating the service panel is properly earthed grounded to the service panel and water pipe.

Moving forward everything that is supplied power must use a single point ground to insure the same voltage potential. If there is a difference in potential this is one way things get smoked. Voltage always moved toward the path of least resistance first and migrates to other secondary systems if available. The next step is to employ a tiered SPD / TVSS protection method that span Type 1~4 where it makes sense.

Type 1 SPD's are installed at the service entrance (utility meter). Type 2 SPD's are installed at the service panel (breaker panel). Type 3 SPD's encompass many varieties but are known as Point of Use (POU) at the electrical outlet. Type 4 SPD's are installed inline (series) prior to the device like a furnace, hot water tank, dryer, washer, dishwasher, etc.

Low resistance earth grounding is the first line of defense and everything is built upon this first line. SPD / TVSS are the second line of defense and their primary goal is to absorb and shunt to ground incoming voltage rise (surge). AVR / UPS offer something a SPD / TVSS can't and that is protection against voltage sags (brown out).

A brown out is more dangerous than a voltage rise (surge) in heavily regulated electronics.

Another very dangerous line condition is creeping voltage rise / voltage sag. That is when the line voltage rises above the expected 120 / 240 VAC in North America. The average (single voltage) electronics is designed to operate from 90~130 VAC. When a slow creeping voltage rise exceeds 130 VAC the magic smoke will be released. The same is true when the voltage sags slowly below 90 VAC. As stated up above when a voltage sag (brown out) is present many highly regulated electronics will continue to increase in current draw while the voltage drops. This is counter to Ohms law but Ohms law only applies to simple resistive circuits and not complex ones that incorporate RLC networks that use boost / buck circuits.

Regardless of the above a AVR / UPS should be used and deployed to maintain not just the line voltage but the frequency in concert with a layered SPD / TVSS protection system.

Things to consider which everyone forgets or understands every time an surge event arises impacts every electronics just a little bit. So if something died immediately and something next to it didn't - that device is simply living on borrowed time!

You'll hear endless Internet stories where someone indicates a huge boomer destroyed X. Only later to say You know last month my dishwasher was fried from that last storm. Now a month later the microwave died, washer, furnace, HVAC, etc.

The above isn't bad luck its simply the fact they were damaged prior but had enough design robustness to sustain a period of time before dying. One thing that should be made clear for those who haven't seen or endured such an event and that is Short Cycling. People will often simply call this a surge event but in reality this is called a spike event. The difference is based on time and the amount of current flow (amperage).

When you see the lights for example come on and off very quickly this is short cycling. This again is one of the most dangerous conditions that can exist when power is about to be restored or where the POCO system is trying to come back up.

When people lack the finances or the ability to deploy a solid electrical protection system the easiest thing a person can do is unplug devices. Next is to turn off breakers at the service panel and for those who live in modern homes most large devices have dedicated switches that should also be turned off like the furnace, AC, HWT, etc.

Lastly, its important to understand that resistance increases when its dry and cold. So in the dead of winter if your soil is less than ideal seeing 50 ohms and above is easy to achieve. As stated many times here and in other forums if you have the ability to purchase multi voltage / multi frequency devices it make sense as they offer a wider operating range and thus offer in a round about way built in surge / sag protection because they are designed to operate in a wider range than 120 VAC etc.

 

[user8963]

As it relates to equipment protection the first thing to understand is basic Ohms law. Once that is understood the next step is to validate the homes electrical system is properly functioning in all aspects. Whatever the POCO needs to do is on them but you can surely make it easier on yourself by offering some facts to them.

In all cases its their job to affirm the incoming line voltage is correct and the system is properly connected to the service entrance. Its also their responsibility to confirm the system is properly grounded. It does not mean they will be able to change any earth grounding issues that may be found.

Meaning if the earth ground is above 25 ohms its on you to correct it by adding more grounding rods in parallel.

Next, is moving inside of the building and validating the entire service panel is properly wired and all connections are tight and not corroded. This again leads to validating the service panel is properly earthed grounded to the service panel and water pipe.

Moving forward everything that is supplied power must use a single point ground to insure the same voltage potential. If there is a difference in potential this is one way things get smoked. Voltage always moved toward the path of least resistance first and migrates to other secondary systems if available. The next step is to employ a tiered SPD / TVSS protection method that span Type 1~4 where it makes sense.

Type 1 SPD's are installed at the service entrance (utility meter). Type 2 SPD's are installed at the service panel (breaker panel). Type 3 SPD's encompass many varieties but are known as Point of Use (POU) at the electrical outlet. Type 4 SPD's are installed inline (series) prior to the device like a furnace, hot water tank, dryer, washer, dishwasher, etc.

Low resistance earth grounding is the first line of defense and everything is built upon this first line. SPD / TVSS are the second line of defense and their primary goal is to absorb and shunt to ground incoming voltage rise (surge). AVR / UPS offer something a SPD / TVSS can't and that is protection against voltage sags (brown out).

A brown out is more dangerous than a voltage rise (surge) in heavily regulated electronics.

Another very dangerous line condition is creeping voltage rise / voltage sag. That is when the line voltage rises above the expected 120 / 240 VAC in North America. The average (single voltage) electronics is designed to operate from 90~130 VAC. When a slow creeping voltage rise exceeds 130 VAC the magic smoke will be released. The same is true when the voltage sags slowly below 90 VAC. As stated up above when a voltage sag (brown out) is present many highly regulated electronics will continue to increase in current draw while the voltage drops. This is counter to Ohms law but Ohms law only applies to simple resistive circuits and not complex ones that incorporate RLC networks that use boost / buck circuits.

Regardless of the above a AVR / UPS should be used and deployed to maintain not just the line voltage but the frequency in concert with a layered SPD / TVSS protection system.

Things to consider which everyone forgets or understands every time an surge event arises impacts every electronics just a little bit. So if something died immediately and something next to it didn't - that device is simply living on borrowed time!

You'll hear endless Internet stories where someone indicates a huge boomer destroyed X. Only later to say You know last month my dishwasher was fried from that last storm. Now a month later the microwave died, washer, furnace, HVAC, etc.

The above isn't bad luck its simply the fact they were damaged prior but had enough design robustness to sustain a period of time before dying. One thing that should be made clear for those who haven't seen or endured such an event and that is Short Cycling. People will often simply call this a surge event but in reality this is called a spike event. The difference is based on time and the amount of current flow (amperage).

When you see the lights for example come on and off very quickly this is short cycling. This again is one of the most dangerous conditions that can exist when power is about to be restored or where the POCO system is trying to come back up.

When people lack the finances or the ability to deploy a solid electrical protection system the easiest thing a person can do is unplug devices. Next is to turn off breakers at the service panel and for those who live in modern homes most large devices have dedicated switches that should also be turned off like the furnace, AC, HWT, etc.

Lastly, its important to understand that resistance increases when its dry and cold. So in the dead of winter if your soil is less than ideal seeing 50 ohms and above is easy to achieve. As stated many times here and in other forums if you have the ability to purchase multi voltage / multi frequency devices it make sense as they offer a wider operating range and thus offer in a round about way built in surge / sag protection because they are designed to operate in a wider range than 120 VAC etc.

as shown on the picture the secondary circuit seems to be burned. this could be a lightning damage (but then it would not burn like this , more just pop like a pimple, if it was real on fire the whole pcb would be damaged and the case melted) or when main voltage is connected to the 2nd circuit.

if the main voltage is the problem (spikes), then there should be damage on the primary circuit.
in US spikes should be no prob, because many devices are rated for 110-240V AC. if main voltages go above 240V for longer then you might have other problems...

 

[TonyR]

What sticks in my mind is that the AC power input side to the POE injector doesn't look physically damaged but the network side (LAN & POE) are blown up and blackened.
Perhaps the Ethernet cables that are strung inside that building are picking up the ESD from nearby lightning like an antenna and the ESD is looking for ground and the first one / best one it finds is via the POE injector.

I've had equipment-damaging ESD from nearby lighting strikes come in via satellite coax, via CAT-5 in an attic (not a metal roof), and subsequent surges via the incoming utility power line and go through a Cyberpower UPS like it was a straight copper conductor.

In 7 years I've lost 3 TV's, a satellite receiver, 2 switches, 2 routers, an AP, a Blue Iris server motherboard and 3 cameras to ESD and powerline surges caused by severe area lightning.

In the SE states of the US, it's not a matter of IF you'll get lightning damage, it's WHEN. Lightning pretty much goes where it wants to; you can lessen the chances of damage by following proper grounding and bonding procedures and having appropriate surge protection and UPS's in all the right places, but......when I can I unplug, a practice that can't happen when a surprise storm comes up or I forget. But when I can, I do it.

 

[oh6hfx]

What sticks in my mind is that the AC power input side to the POE injector doesn't look physically damaged but the network side (LAN & POE) are blown up and blackened.
Perhaps the Ethernet cables that are strung inside that building are picking up the ESD from nearby lightning like an antenna and the ESD is looking for ground and the first one / best one it finds is via the POE injector.

That's exactly what I was thinking also. But the damage in these devices look so big that could it be ESD? I tried to find pictures of ESD damage in poe injectors / power supplies, but could not find any. In this time of year there is no thunderstorms in Finland.

 

[oh6hfx]

As it relates to equipment protection the first thing to understand is basic Ohms law. Once that is understood the next step is to validate the homes electrical system is properly functioning in all aspects. Whatever the POCO needs to do is on them but you can surely make it easier on yourself by offering some facts to them.
Lastly, its important to understand that resistance increases when its dry and cold. So in the dead of winter if your soil is less than ideal seeing 50 ohms and above is easy to achieve. As stated many times here and in other forums if you have the ability to purchase multi voltage / multi frequency devices it make sense as they offer a wider operating range and thus offer in a round about way built in surge / sag protection because they are designed to operate in a wider range than 120 VAC etc.

Thanks Teken, very interesting information. I will let electrician do the measurements and checks, let's see what he founds.

 

[oh6hfx]

Can the problem be that the cams @ the sheep barn are at "local ground potential"? With such a long
run and everything is at "local ground potential", one may see a big "earth battery". The blue MOV looks ok,
so then the ground is now passed via the twisted pairs to case ground.

Have you measured how much voltage is on the ground line at the sheep barn? Is the "buck" caused by
equipment in the barn getting shut down or at start-up?

There is no very long runs, just around 20m to the outdoor access point and <10m to the camera. Camera is attached to metal inside ceiling. With ohmmeter it seems that ceiling is not grounded to buildings safety groundwire. It is "grounded" via steel support bars. Now thinking, there might be potential difference.

I don't know if the damage happens in shut down or start up. Last time the outage was longer, around 15 minutes, so it was not a cycling.

 

[sebastiantombs]

First thing I'd do is get a high impedance meter and find a real, solid, true ground in that building. Then I'd start looking for voltage between that ground and ground in the equipment and even between pieces of the building. Something is definitely wrong in the electrical system.

Years ago I had a neighbor come to me complaining that every time he urinated he got a shock. I traced it out to a bad, not up to code, installation that was feeding a subpanel in his basement. That panel had the water heater and all the basement lighting and outlets on it. The ground had broken in an LB (probably pulled too hard and nicked when installed), corroded and arced, so ground was coming through the plumbing system and that ground had a different resistance than the rest of the electrical system, Only took me about an hour to trace it out to the source of the problem. I replaced the whole feed with a new run of 8/3 to bring it up to code, the original was a 40 amp breaker feeding through 12 gauge wire to the basement enclosed in conduit that was running outside. Needless to say, he was "relieved".

 

[BORIStheBLADE]

Another possible issue which I don't recall being brought up is bonding of water pipes, ground rods, ground rings and building steel together . If they aren't all bonded you have the potential of voltage between two grounds and create a high impedance path to ground.

I actually have been shocked by a ground when touching another ground because of this..

 

[oh6hfx]

Another possible issue which I don't recall being brought up is bonding of water pipes, ground rods, ground rings and building steel together . If they aren't all bonded you have the potential of voltage between two grounds and create a high impedance path to ground.
I actually have been shocked by a ground when touching another ground because of this..

That's what I thought also when I realized that camera chassis is attached to steel ceiling which is not grounded in the safety ground.

To me it feels that there has been LOT of energy gone trough those fried devices. It needs >200C to get solder to melt. Could a voltage between two grounds do such a damage?

 

[garycrist]

Could a voltage between two grounds do such a damage? YES!

Think about 2 grounds separated by 100 feet just for fun. Lightning hits by ground rod 1. @ 1 foot per nanosecond (speed of light),
You already know about coax propagation speeds. So the electrons start to flow and smoke comes out. Not necessarily from the strike
itself, but from the leaders. The unit might have burned up before the strike!

 

[garycrist]

Leaders and step leaders. Here is a pic that shows the lightning one hears and see (the main flash in pic).
But all around one sees because of camera (eyes are not fast enough) all the leaders already sucking the life
out of someone's equipment, not the main blast!

Perfect Timing: Photog Captures Lightning Striking the Tallest Building on Earth

Michael Shainblum captured this perfectly timed (call it luck if you want, but luck favors the prepared, so we're going with perfectly timed) shot of

petapixel.com

 

[oh6hfx]

Could a voltage between two grounds do such a damage? YES!

Think about 2 grounds separated by 100 feet just for fun. Lightning hits by ground rod 1. @ 1 foot per nanosecond (speed of light),
You already know about coax propagation speeds. So the electrons start to flow and smoke comes out. Not necessarily from the strike
itself, but from the leaders. The unit might have burned up before the strike!

Sorry, I might sound stupid, but if there is no lightning, where the energy comes from?

 

[oh6hfx]

That's the 24V access point PoE injector. LAN connector solder melted and foils blown out.