power issue?

[Mike]

Should I replace it or is this fixable? ws6 mini 1.4.8rc5 firmware. Never seen it happen in a year on this one. Nothing changed.

Mike

Sep 11 21:47:58 switch[747]: 24V is out of range (22.7V)
Sep 13 07:51:38 switch[747]: 24V is out of range (22.7V)
Sep 18 07:14:00 switch[747]: 24V is out of range (22.8V)
Sep 19 07:38:17 switch[747]: 24V is out of range (22.7V)
Sep 19 12:34:45 switch[747]: 24V is out of range (22.7V)
Sep 19 17:41:36 switch[747]: 24V is out of range (22.7V)
Sep 21 08:11:38 switch[747]: 24V is out of range (22.7V)

[Julian]

It's a good candidate for an RMA, should be fixable once we get it in-house. RMA procedure here: http://forum.netonix.com/viewtopic.php? ... =30#p20516

[Mike]

I can RMA it back but one question I have is why cant these be like 25 volts which is plenty safe for anything rated 24 in my opinion and that will allow for long runs and voltage drops etc being closer to 24. I have seen dozens of Ubiquiti equipment running off Motorola POE's at 29.5 volts running for years without a hitch. Need to reverse power wires on one end but works fine. In old days on long runs we would use a Moto POE so voltage would still be there on far end but Ive seen them on 10 foot runs running years so the equipment can handle 25 all easy not sure why make it 24 exact knowing there will be some loss and drops. 24.5 would be better then 24 and be perfectly safe.

Mike

[Dave]

I can RMA it back but one question I have is why cant these be like 25 volts which is plenty safe for anything rated 24 in my opinion and that will allow for long runs and voltage drops etc being closer to 24. I have seen dozens of Ubiquiti equipment running off Motorola POE's at 29.5 volts running for years without a hitch. Need to reverse power wires on one end but works fine. In old days on long runs we would use a Moto POE so voltage would still be there on far end but Ive seen them on 10 foot runs running years so the equipment can handle 25 all easy not sure why make it 24 exact knowing there will be some loss and drops. 24.5 would be better then 24 and be perfectly safe.

Mike

Mike, thanks for your comments. Our internal voltage board 24 volts is set higher than 24 volts. However, you failure indication is that the on board 24 volt circuit is at 22 volts. This is not the far end of the cable where the voltage is expected to be low (cable loss), but rather at the source. So something is wrong with the on board regulator, or your load is really heavy & dragging the regulator down.

[Mike]

I understand its at the source making it even more scary because I would hate to see what the voltage is 310 feet out on a radio. Im saying bump it higher to allow for some drop in load and length. Your on the edge to begin with. If I had a dial I would set them all to 25 to 26 to start at source.
I checked 10 switches of mine just now and here are the results at source. Another volt wouldn't hurt to help on voltage drops on long run up towers etc..

23.2 the one that gave me errors earlier at 22.7
24.6
24.2
23.8
24.4
24.1
24.0
24.2
23.9
24.5

Mike

[sirhc]

UBNT radios will work all the way down to around 12V or even less but 24V is pushing the upper limit if we set to 25V and people use a short cable it could damage the device.

We could allow the voltage to be software selectable but that would increase cost and if people turned it up too high on a short cable and burned up radios they would also be upset

Allowing software selectable on a per port basis would require a LOT more stuff and would be very expensive and much larger unit.

[lligetfa]

We could allow the voltage to be software selectable...

I would be happy with a trim pot. Actually it is the 50V side that I would want to increase to 54V.

[jakematic]

We could allow the voltage to be software selectable...

I would be happy with a trim pot. Actually it is the 50V side that I would want to increase to 54V.

The problem with trim pots is when people find them:

eleven.JPG (20.96 KiB) Viewed 5057 times

[Mike]

UBNT radios will work all the way down to around 12V or even less but 24V is pushing the upper limit if we set to 25V and people use a short cable it could damage the device.

We could allow the voltage to be software selectable but that would increase cost and if people turned it up too high on a short cable and burned up radios they would also be upset

Allowing software selectable on a per port basis would require a LOT more stuff and would be very expensive and much larger unit.

Like I mentioned earlier I've seen Ubiquiti radios go years on a 29.5 volt poe so I don't believe 24 volts is pushing the upper limits by any means.
But Im glad to know they work at much less voltage.

Mike