WS-6-MINI Keeps crashing/rebooting?

[sbyrd]

I would test all the port POE options then test each port to achieve 1G and if it runs fine in office it is OK.

Temperature "could" still be the culprit but my bet is on the fact that your service ground was different then the tower ground.

Ok so how long should I bench test with each port? If it links up and powers up is it fine, or should I make sure it runs without issue for 24hrs on the bench? Remember the switch ran fine in the field for 22hrs the other day.

Also me and the tower owner got out a volt meter and ran a resistance test on the grounds. Not sure if it means anything or not, but when I went from tower ground lug to the ground clamp I put on the tower we got 0 resistance. When we went from the tower ground lug to the AC ground wire we got -222.00 resistance.

[sirhc]

The WS-6-MINI is good up to 90C board temp
I am saying just test all the ports and then let it run for a day or 2 if you like on the bench.

I still think it is your grounding but if not that then we are down to insufficient power which you could test by using a WISP switch to power it or get the larger POE brick from streakwave that I suggested which is 54V 80W

[sbyrd]

I have tried 3 switches at this site. Switches 1 and 2 both were running before Sirhc told me to disconnect the PoE ground from the AC ground and attach it instead of the tower. Switch 3 was a fresh switch never connected to improper grounding.

I have bench tested switch 2 (it was only running with improper ground for less than 1hr) in the office and I was able to get it to reboot.

Here is my bench setup

AF24 POE powering MINI in port 1
Port 2 - NO power, Uplink port to Laptop -1G
Port 3 - 24v powering nanobridge - 100F
Port 4 - 24v powering Rocket M365 - 100F
Port 5 - 24v powering Powerbeam M5 - 1G

This ran in my office for 12 hours with no issue. I then decided to add a Rocket TI to Port 6 at 48V. As soon as power was applied to the Rocket TI the switch locked up and then went into a reboot cycle. The switch did not come back up until i removed all the radios from the switch.

I have powered off the switch and then turned off POE to each port. I then applied power one port at at time. Switch is up and running but only lasted 2 minutes. I will run the same test with switch #1

When switch #2 rebooted the power load was around 21 watts and the board temp 49C CPU temp 74C.

[sbyrd]

While the switch was locked up the power to each radio was fine and I could reach them wirelessly. They just had no ethernet link.

[sirhc]

Seth is switch #3 at the site with the grounding fixed running OK still or are you saying it rebooted as well now?

[sbyrd]

Seth is switch #3 at the site with the grounding fixed running OK still or are you saying it rebooted as well now?

No switch 3 is currently running fine. Will know better later today if it is truly stable.

[sbyrd]

I have tried 3 switches at this site. Switches 1 and 2 both were running before Sirhc told me to disconnect the PoE ground from the AC ground and attach it instead of the tower. Switch 3 was a fresh switch never connected to improper grounding.

I have bench tested switch 2 (it was only running with improper ground for less than 1hr) in the office and I was able to get it to reboot.

Here is my bench setup

AF24 POE powering MINI in port 1
Port 2 - NO power, Uplink port to Laptop -1G
Port 3 - 24v powering nanobridge - 100F
Port 4 - 24v powering Rocket M365 - 100F
Port 5 - 24v powering Powerbeam M5 - 1G

This ran in my office for 12 hours with no issue. I then decided to add a Rocket TI to Port 6 at 48V. As soon as power was applied to the Rocket TI the switch locked up and then went into a reboot cycle. The switch did not come back up until i removed all the radios from the switch.

I have powered off the switch and then turned off POE to each port. I then applied power one port at at time. Switch is up and running but only lasted 2 minutes. I will run the same test with switch #1

When switch #2 rebooted the power load was around 21 watts and the board temp 49C CPU temp 74C.

Just tested switch #1 with the same setup as above. With devices in ports 2-5 the swith ran for 18 minutes. I then added the 48V device is port 6 and the switch locked up 7 minutes later.

Both switch #1 and #2 have high variance on the reported Board Voltages on the Status page. Switch #3 (the new and working switch) has stable voltages.

Sirhc - Let me know if it is worth it to have these repaired or if I should just buy new ones.

I am still waiting till later today to see if switch #3 (at the tower) is stable.

[sirhc]

Well lets first see what happens to switch #3 with the grounding fixed after 4PM.

But if we determine that #3 is fine and we "assume" 1 and 2 were damaged from the ground potential difference then worst case is $75 each for new replacement boards but well talk about that after 4PM when we know more.

What is your board temp running now on #3?

[sbyrd]

Outdoor temp -74F
Board - 60C
CPU - 86C
PHY - 92C

All Board voltages are stable with less than 1V of variance.

[sirhc]

Outdoor temp -74F
Board - 60C
CPU - 86C
PHY - 92C

All Board voltages are stable with less than 1V of variance.

Yea, your pretty safe up to about 90C +/- for board temp.

CPU and PHY are safe up to 125C