DC-Switches -- Reboot upon fast voltage drop !

[yahel]

We've recently noticed switches reboot when grid power drops.
Batteries are connected in parallel to the switch, and upon loss of grid voltage drops from ~ 27-28v to 24-25v and results in a reboot of the switch.
The switch comes back seconds later, and works just fine on the lower voltage.

This happens also with 48v systems (where normal voltage is in the 55-56v area and drops to 48-50v upon loss of grid power).

I'm tempted to say this is due to recent firmware upgrades, as we had years of service without this problem, with the same switches...
Obviously, these occurrences are not common (thankfully grid power is pretty stable), so can't say with certainty what's going on.
We did had about 10 such cases in August alone, and none before, at least not that we've noticed (August did not had particularly many power failures).

This is probably also related to another observation:
We've recently purchased a number of these chargers (Mean-Well PB-1000-48):
http://download.siliconexpert.com/pdfs/ ... asheet.pdf
The reasoning was to have good chargers for our large sites, to prolong the life of our (costly) batteries.
These sites use 4x200AH AGM batteries in series.
We liked the reputable 8-stage charging intelligence of these chargers.
Sadly, the Netonix switches (tested only with WS-8-150-DC models), reboot occasionally when the charger is in it's 8-stage mode.
They work fine when the charger is in the 3-stage mode, which is the solution we're using for now, at the cost of losing the benefits of these costly chargers.

Attached is an image of the Netonix's voltage graph upon connection of the smart charger... The switch would reboot when the charger is in it's float/trickle stage, especially in the 5am time frames (being the coldest hour - the chargers are temperature compensated {we're in coastal California - so it's never very cold}).

I think all the above are symptoms of the same problem - which I'm guessing got introduced into recent firmware of the DC-DC converter.

Thanks,

Yahel.

[Julian]

Yahel, I am setting up to duplicate this - In your experience, would a drop from 48 to 41v trigger this?

What's the power supply firmware version and hardware revision of the affected units? The tale is there, but now I need the details.

Thanks!

[yahel]

Confirmed to happen at these switch models/versions at least:

WS-8-250-DC:
Firmware Version 1.4.8rc7
Board Rev D
Power Supply Firmware Version 53
Power Supply Board Rev A

WS-8-150-DC:
Firmware Version 1.4.8rc10
Board Rev F

========

Don't know about dropping to 41v ... but I'm guessing that yes -- it would trigger this.
Let me know...

Thanks!

Yahel.

[sirhc]

Is the switch connected "directly" to the batteries bank or is your power supply in between the batteries and the switch?

Our SMART DC switches are designed to hook directly to the battery bank not a power supply.

If you are going to hook the switch to a power supply then you should use the WS-12-DC and power it with a 48V power supply

[yahel]

All of our switches are DIRECTLY connected to the batteries (there's a fuse in between, but nothing else).
They are in parallel to the battery charger.

These reboots upon sudden voltage drop are recent - it was fine for years, but some recent firmware upgrade damaged it.

We do NOT see this problem with the WS-12-DC (strangely enough), but I have not tested extensively yet, and we're slowly phasing out all Netonix switches without a DC-DC converter. Ironically, so that we'll have better stability/tolerance to voltage changes and low batteries situation.

Yahel.

[sirhc]

Julian said he tested this in a LAB and could not duplicate?

So switch is connected to batteries.
Charger is connected to batteries.

To confirm this would allow for the charger to be removed from the batteries and the switch is still connected to the batteries? (just making sure the charger is not involved)

Post up your these TABs
Status
Ports
Power
Device/Status

[Dave]

The reasoning was to have good chargers for our large sites, to prolong the life of our (costly) batteries.
These sites use 4x200AH AGM batteries in series.
We liked the reputable 8-stage charging intelligence of these chargers.
Sadly, the Netonix switches (tested only with WS-8-150-DC models), reboot occasionally when the charger is in it's 8-stage mode.
They work fine when the charger is in the 3-stage mode, which is the solution we're using for now, at the cost of losing the benefits of these costly chargers.

It is rather strange that the reboots happen in Stage 8, as Stage 8 is the maintain low charge stage........

[Dave]

We've recently noticed switches reboot when grid power drops.
Batteries are connected in parallel to the switch, and upon loss of grid voltage drops from ~ 27-28v to 24-25v and results in a reboot of the switch.
The switch comes back seconds later, and works just fine on the lower voltage.

Yahel

What do you have set for the shut down voltage of the switch?

[yahel]

Attached requested screenshots of (one of) the switches that rebooted earlier today when the area lost power.

Yes - we can power-off the charger and the switch (used to) remain working - now it reboots and comes back (with the charger still off).

The voltage for hibernate, as shown in the screenshot, is set to 23.5 for this switch - but voltage never dropped that low.

Another switch that rebooted today is a WS-8-150-DC which does not have hibernation feature.

Yahel.

[Dave]

have some ideas, I have asked Eric to prepare some test code, we will test in our lab as best we can & then send out for you guys to try...