I’ve recently lost 4 3000L with sector antennas at the same time which I must suspect due to lightning.
On internal inspection no visible damaged inside the 3000L but definetely something was fried on the electronics side, seems some sort of short on the low voltage power rails after the PoE circuit, PoE circuit appears OK.
Radios, PoE 30V inline step down and switches which each 3000L was connected to weren’t damaged. Hence suspect some sort of EMP killed the 3000Ls.
Have no lightning arestors installed, units aren’t earth and using UTP.
Would earthing via the lug on the rear of the 3000Ls offer any further protection?
Can’t find any information in the manual on recommendations either.
thanks,
Rob.
At the VERY LEAST you need to be using shielded cable. This will provide a path to ground for the static buildup that is happening on the radios and cable. Bonding the radio and using a lightning arrestor are both best practices as well.
ive never lost a radio to lightening and I don’t follow best practice either.
I do this: use good cable with the added drain wire and solder it to the metal rj45s at both ends. Put a ground spike in by tower and bond your POE switch to this spike, bond your mains power to the spike and bond your antenna’s mast to it, and bond all antennas to mast. Use grunty af earth cable from antenna mast to earth spike, and decent stuff from radio’s to mast. Also use grunty stuff from POE switch to spike.
If mains power comes from far away, like a house that’s 50m from tower, cut the mains earth wire completely and only use your local earth spike at tower for all grounding.
Doing nothing will result in more losses. Losing radios sucks.
As a coincidence - we lost two 3000L and 4 300c radios (same basic hardware) early yesterday when there were nearby lightning strikes. It reminded me that last time there were nearby strikes in the same area we lost some of the same 300c radios. We have never lost any other Cambium radios in a similar situation, or in the same area.
The radios that failed were grounded and had surge arrestors inline.
I wonder if there is some component in the 3000l/300c that is especially sensitive?
Maybe co-incidence. Just that we’ve rarely lost radios in the past.
The 4 x 3000L that failed were mounted on the same tower with 2 x V3000, 1x V2000 and 1 x V5000. All mounted at the same height and randomly spaced around the tower approx 2 meters from the top. (It’s a rather large existing tower). They didn’t miss a beat .
Two difference I see between the devices, the XV2 and V series of radios are by default grounded via the standard supplied brackets.
3000Ls have a large sector antennas compared to the V series, hence providign a greater surface area to absorb EMP/ESD??
Anyway had another site go down, but it appears to be a significant hit. Anyway nothing like a 4000km round trip service call to start the year off!
Keep you posted on my findings.
I wonder the same thing. I have a customer that is on his 5th radio in 18 months. After he lost the first 2 we installed a lightning arrestor and grounded it, and he still loses a radio after every storm. We also use shielded cabling.
Yeah, we got rid of every 3000L radio. They blew up left and right with lightning even with shielded cable and shielded flexible conduit running to them. Think the L stands for lightning. Went back to regular 3000 units. The 3000 units take a direct strike to die, a 3000L will die with a close hit. This is in Arkansas though, so it is lightning central. The old 1000 2.4 radios and the 3000 non “L” radios are solid equipment, but we finally stopped using 3000L units after 12 of them died in about 6 months. Also, the regular 3000 units just have way better QOS and service. You get what you pay for I guess!
Years ago, I hosted a webinar with RF Elements on this topic, which might be helpful:
In short, when an Electrostatic Discharge (ESD) event occurs, such as on a tower, mast, or connected environment, the discharge spreads through all available paths based on each path’s impedance.
The first step to protect against ESD is ensuring compliance with NEC 810.21. This includes bonding the antenna mast and antenna discharge unit to the intersystem bonding terminal. Here is an Article 810 - Radio and Television Equipment guide in case that’s helpful. In practice, this means creating a unified grounding system by properly bonding the electrical ground, tower or mast ground, surge protector ground, and earthing/ground rods. This minimizes impedance and ensures an effective path for surges.
For protection against direct strikes, ensure your equipment is installed within the protected zones defined by the rolling ball method:
Finally, always follow the manufacturer’s installation instructions, including the correct installation of recommended surge protection. In my experience, properly bonded and grounded systems rarely sustain any form of damage from lightning or ESD. When challenges arise, the issue is often traced to improper bonding to the utility grid powering the equipment (NEC 820.21(J)). This can create a condition where your equipment becomes the lowest impedance path between two grounded systems, inviting damage.
I had a few customers over the years that lost a radio almost every single time it stormed. Most we resolved this way -
(1) Installed surge suppressor very close (inches) to the radio.
(2) Use a F200 / F300 dish and run a ground wire from the metal of the dish/mount to the ground lug on the surge suppressor (we were never able to make this work using F180s or 300-16 or 300-19 where there was no way to ground the case/reflector/antenna).
(3) Use very short ungrounded (no shielded/metal ends and cable no longer than it has to be) patch cable between the radio and the suppressor.
Ground, shield, drain wire everything normally from the surge suppressor by the radio to the surge suppressor where the cable enters the house/structure. That is assuming you are using a surge suppressor where the cable enters house/structure and it is properly grounded.
I had 1 customer though that, long long story spanning 10+ years and electrical engineers and crazy rigged electrical antics by the customer (hundreds of feet of extension cords plugged into extension cords and buried between outbuildings 300+ feet apart from each other daisy chained to the tower) and thousands spent on grounding systems/arrays and tower climbs and blown radios and surge protectors… one radio got fried while hanging from the climbers harness as he climbed the tower… just beyond bizarre . If the radios (our ePMP and their ubiquiti bridge radio) were not physically disconnected at the base of their tower before a storm comes within 10 miles of them, the radios would be fried every single time, no exceptions.
Some places are just cursed and no radio was ever meant to be installed there.
I don’t know if I ever took any pictures or where they might be now if I did. The Wisp I ran was sold about a year ago and now I just do some consulting and backend stuff for the Wisp that bought it (they are an Ubiquiti Wisp). I don’t go out and do customer installs or repairs any more at all.
Found a really old one. This is an early attempt, later we would try to mount the surge suppressor (little white box there is a Mimosa NID / surge suppressor) even closer to the radio. No need to run ground from the dish to the surge suppressor here because the suppressor is bonded to the metal post it’s mounted on there and so is the dish because it’s all metal clamped to metal. The ethernet cable connecting the radio there to the little white box has a regular plastic (non-shielded, no metal jacket) RJ45 connector on the end of the cable plugged into the radio.
The cat# cable coming out of the little white Mimosa NID and going underground to over there by the AC unit has a drain wire in it and the metal jacketed RJ45 connectors with the little nubs for connecting the drain wire to them. You can’t see it but on the other side of the electric disconnect box for the AC over there the cable comes out of the ground and plugs into another Mimosa NID and that NID is grounded to the ground bar inside the AC disconnect box (or maybe that’s a ground rod I see there, if so it would have been grounded to that).
On most installs we did not do this and in general only did it on installs were the dish was installed on a post in the yard , in tree or had a very long cable run between the radio and the house ( e.g. some customers had 100ft tv antenna towers or build towers for us to install the radio on). However even when we did not use the surge suppressor by the radio we did not use a shielded RJ45 connector on the radio end of the cable. We did always install/ground a surge suppressor as close as possible to where the cable goes into the house and would use a shielded RJ45 connector with the drain wire on that end of the cable.
Sorry to break in but don’t forget a lightning arrestor is designed to protect equipment downstream (behind it in the circuit), rather than the device directly connected to the antenna or radio that might be struck by lightning or near lighting. Of course it dies.
With the climate change happening expect more frighted radio’s in the future!
So as I mentioned lost 4 3000Ls on a tower at the start of the post.
And mentioned that had to attend another site…heres what I found.
This tower had 2 x 3000Ls with sectors, 1 x 3000L with Omni and 1 x E500
Arrangement was two sector at very top of tower and OMNI probably 2meters below and E500 say 5meters further down. All connected directly to the same switch which was also fried.
E500 and OMNI survived! The fact that the 3000L did survive, I don’t think they’re as delicate as some have mentioned.
Since these locations are extremely remote all equipment was swapped out, not worth the risk.
All equipment was installed with no shielded wires, no grounding etc. Tower is naturally grounded. ie roof and frame is steel all guy wires etc… secured to roof.
I will review your links @Joshaven_Potter. But what I struggle to understand is by earthing your equipment your creating a electrical path. Doesn’t this make a path for surges to follow? that is if the surge is somewhere else in your earth cabling it then has a path to the radio?
The fact that I’m not earthing the radios makes them to some degree isolated and a less attractive path for surges??
Anyway heres some pics of the damage caused, other equipment was lost. Word was you could hear and smell the electricity in the air. Community was cut of for almost 2 weeks by roads on accessible via helicopters.
Clearly a direct hit and no earthing etc… would protect in this instance.
Just a bit of background, I’ve got probably 50 sites in remote parts of OZ.
Maybe I’ve been lucky but over the years we’ve been doing this, have lost maybe 2-3 radios in that time.
Maybe I was lucky and my luck has run out or I’ve just having some bad luck!
The issue is that a lower-impedance paths through your equipment allowed excessive current during an ESD event (lightning strike).
Think of it like interconnected lakes: when water is rapidly added to one lake, it flows through all paths to equalize. Similarly, in a surge, electricity flows through your tower and utility ground system based on their impedance until an equilibrium is achieved. Bonding systems with low-impedance connections prevents sensitive equipment from becoming the primary route.
It’s helpful to remember that larger connections have lower impedance, while electronics inherently add impedance. Bonding your tower to the electrical ground rod with proper cable and connections helps prevent ESD from flowing through your equipment. Using an affordable “Earth ground resistance tester” ensures your grounding is effective and reduces the risk of equipment failure. Without bonding, high-impedance grounding in either the tower or electrical system forces current through your equipment, acting as a bridge. This can cause interface errors during normal operation and severe damage during ESD events. Based on your description and photos, it appears a surge caused excessive current between the power system and the radio mast.
Surge protection devices and proper inter-system bonding (minimum AWG 6 cables) mitigate damage by directing excess energy to a common ground. However, neither grounding nor surge protection alone suffices—a unified system equalizing potential across all components is essential to lightning survival in a grounded system.
To protect your system, bond your electrical ground rod to the tower ground rod per NEC 810.21. This ensures all components rise and fall in potential together, preventing damaging current flow through your equipment.
It might seem logical to remove ground connections to isolate your system, but the utility power system is inherently grounded and referenced to ground (both hot and neutral are connected to ground at the transformer, at least in the U.S.A.). Achieving true isolation requires isolating both your equipment and power, which is complex and impractical. Proper grounding is a simpler, more effective solution. With correctly installed equipment, lightning damage becomes a rare occurrence—this is based on my experience assisting hundreds of operators worldwide in addressing similar issues.
Surge suppression depends on a proper ground reference, which requires a well-bonded system to function effectively. In environments where electronic equipment is frequently damaged, addressing the root cause is far more effective than relying solely on surge protection.
For instance, I assisted a customer whose equipment on a water tower sustained damage during nearly every storm, often burning out surge protectors. They suspected insufficient grounding of the water tower and were asking how to enhance it, believing that water towers inherently attract lightning.
Using a 3-pole earth ground resistance tester, I found the water tower had an excellent connection to earth, and the utility ground also had a good connection. The real issue was high resistance between the water tower ground and the utility ground. With the equipment connected to both, the equipment became the unintended ground-bridge for surges—causing the damage.
The solution was simple: bonding the utility ground rod to the water tower using 2x AWG 4 conductors we had on hand. This immediately and permanently resolved the Ethernet errors and storm-related damage. The system has operated flawlessly for over five years. Proper bonding between grounded systems is the fundamental key to long-term reliability and damage prevention. Surge protection, while important, serves as a secondary line of defense behind properly installed systems.
This approach generally works, but cutting the ground wire on long connections is counter-productive and violates the NEC requirements for proper bonding and grounding. A 110V power connection is established by bonding the center point of the transformer’s 220V coil to neutral, which is also earth-bonded. This ensures all conductors are consistently referenced to earth.
Cutting the ground wire increases the impedance between neutral and ground, but it does not isolate your connection, which seems to be the intended purpose of this approach. Both the neutral and load conductors remain referenced to ground, increasing the likelihood of an ESD event between equipment connected to the tower and the utility mains.
The ground wire provides a low-impedance return path for fault currents and is bonded to the neutral at the main fuse box and power meter for safety and reliability. Removing this connection undermines safety, reduces surge protection effectiveness, and raises the risk of dangerous voltage differentials and equipment damage. NEC 250.4(A)(5) explicitly requires that all grounding systems eliminate potential differences through proper bonding, and NEC 250.50 mandates that all grounding electrodes at a structure be bonded into a single grounding electrode system.
Using a local earth spike at the tower as an inter-system bonding terminal is highly beneficial, but it must comply with NEC 250.32(A), which requires a grounding electrode system at the remote structure to be bonded to the equipment grounding conductor (EGC) of the feeder. Additionally, NEC 250.32(B)(1) mandates running a bonding conductor (EGC) alongside the feeder conductors to create a unified grounding system between the structures.
Instead of cutting the mains earth, the proper solution is to run an equipment grounding conductor alongside the mains power and install additional ground rods connected to this conductor at the tower. This creates a well-grounded bonding path between the tower and utility, ensuring compliance with NEC requirements while improving both safety and system reliability.
Here is a good explanation of how transformers are grounded for further study that helps understand why cutting the ground is actually counter-productive:
Here’s another helpful video (more about code) that talks about grounding with supplemental ground rods (tower spike) and how to properly connect that to the utility ground system:
