Quickly Fluctuating Voltages WS-8-150-AC

[Chris@edgarhighspeed.com]

Morning,
After running all night the swapped switch is now showing the voltage fluctuations. Same as the one I just took out. Perhaps a screen shot of the main page will provide any clues.

The switch isn't rebooting, it is throwing 3.5V warnings but so far no other warnings.

The switch is connected directly to the halo ground. grounding goes. Halo--->buss bar ---> AC plug ground
---> WS-8-150-AC ground lug
Radio and mounts are grounded directly to the halo.


Firmware 1.4.2 Final.


https://drive.google.com/open?id=0BxVaW ... lNVdFhId1E

The cable distance to both AF5X backhauls is 60 feet, I am grasping at straws now but perhaps it has something to do with the AF5X being powered on the 24V instead of 48V.

[sirhc]

I do not know, there has to be something environmental here?

But the odds of you getting (2) defective switches is so high it is un-real.

My guess is you could put switch after switch in here and the same thing would occur?

I still suspect grounding of some sort but from what you're telling us your doing it right down where you are not using the electric service ground "at all" but only the HALO ground and the HALO ground is now connected to the AC socket ground?

What size wire runs from the roof to the boiler room for the HALO ground and how long is it?

[Chris@edgarhighspeed.com]

Seems environmental to me also.

I agree that I could continue to put in switch after switch and have the same thing happen. I thought I had it solved with redoing the grounding.

Just like in the photo from the post you sent me, I am not using the Utility ground at all. The 110 plug I have wired in my cabinet has the hot and neutral from utility and the ground from HALO.

The run from the HALO to the cabinet is 30 feet with 6 AWG ground wire, then it goes to buss bar and using 14 AWG bare copper to the switch 18 inches long and a/c plug 12 inches long

HALO is 2/0 cable

[sirhc]

So the first switch lasted how long before it started this behavior?

This switch was less than 24 hours right?

I sent a Skype to Dave to ask what he thinks is getting damaged to cause this.

Are you using any Ethernet surge suppressors? (Personally I hate them and do not use them)

Once on the roof how are your antennas mounted and to what?

Do you have a ground wire running from the HALO up there to each antenna?

Also do you have a sufficient service loop to increase the path of the Ethernet so it would be longer than the intended ground path to carry away Static and ESD charges?

I am starting to possibly suspect Static charges maybe?

Static charges are generated by wind blowing past the antennas at the right relative humidity which might explain why it ran for awhile until it cooled down changing the humidity and the wind picked up with the temperature change.

Static charges can be very high voltages.

Is this rooftop windy?

I know we are grasping at straw here but they are all valid.

[Chris@edgarhighspeed.com]

So the first switch lasted how long before it started this behavior?
- about 60 days

This switch was less than 24 hours right?
- yes

I sent a Skype to Dave to ask what he thinks is getting damaged to cause this.

Are you using any Ethernet surge suppressors? (Personally I hate them and do not use them)
- none, I hate them too and the ones I hade from UBNT were very looser

Once on the roof how are your antennas mounted and to what?
The antenna are mounted to a galvanized 2.5 inch pole, which is anchored into the side of the cement building. Antennas are 1 RD-34, 1 AF30, 1 RD-30 and 1 LBE23AC on a single pole

Do you have a ground wire running from the HALO up there to each antenna?
- Ground wire from HALO to pole and from HALO to each antenna

Also do you have a sufficient service loop to increase the path of the Ethernet so it would be longer than the intended ground path to carry away Static and ESD charges?
- Each antenna has a service loop of about 12-14 inches


I am starting to possibly suspect Static charges maybe?
- Maybe

Static charges are generated by wind blowing past the antennas at the right relative humidity which might explain why it ran for awhile until it cooled down changing the humidity and the wind picked up with the temperature change.
- we get warm days 80F here this time of year and we are currently sitting at 73% relative humidity

Static charges can be very high voltages.

Is this rooftop windy?
- I was on the roof yesterday and there was little wind, I didn't notice any large amounts of wind last night.

I know we are grasping at straw here but they are all valid.
- true

I will continue to run this switch like this until it dies or we figure out what to do, so far it hasn't rebooted or anything but I am beginning to not trust it. I also thought maybe the UPS it is on was giving issue but I check the Mikrotik voltage and it seems stable at 23.7V

[Dave]

I can't add to much to this story..our ws-8-150-ac design is practically the same as the ws-12-250-ac....just a smaller board & smaller power supply.

[sirhc]

I can't add to much to this story..our ws-8-150-ac design is practically the same as the ws-12-250-ac....just a smaller board & smaller power supply.

Do not think he mentioned the WS-12-250-AC
I am asking what could be damaged to cause this.

[Dave]

Hi Guys,
I was wondering if someone else can confirm this as normal or abnormal. It isn't happening on my WS-12-250-AC.

He did mention the 12-250-ac....but that wasn't even my point..i was just stating that the new ws-8-150-ac is basically the same design as our ws-12-250-ac, so I don't have any ideas what might be causing the failures.