Netonix Forums

Hello. We deployed our first WS-8-150-DC switch and is is being powered by a Meanwell AD-155B power supply.

We have the load voltage set to 27.0 volts as shown in the screenshot but when I look at the input current graph, I see it dip down from about 1.5A to less than 0.5 A about every 3 minutes. Is this just a bug in the graph or something more I should be worried about? The reason I think it may be a bug is when it dips way down, it has a flat line instead of showing more detail like when it is operating normally.

Also, when the graph shows a flat line, the input current shown at the top with the gauge shows the normal 1.4 to 1.5 A. I've attached a screenshot showing this behavior as well with the graph not showing what the other indicator shows on the same page.

Let me know if this is a bug in this RC firmware. Thanks

I am at a loss as to why people are buying the WS-8-150-DC or WS-12-250-DC if they are going to power them with a power supply.

I do see you are using a 27V power supply which the WS-12-DC would not work with but you could use a 48V-53V power supply and use the WS-12-DC which is less expensive.

The idea of the WS-12-250-DC and WS-8-150-DC is that you hook them to the battery bank not a power supply and can condition the power as the battery bank drops or the voltage goes over due to charging.

If you are going to use a power supply then buy the WS-12-DC as that is what that product is designed for.

Also if you going to power a WS-8-150-DC the power supply feeding it needs to be 300 watts +/- do deal with the inrush properly and does not damage the power supply over time.

And if you going to power a WS-12-250-DC the power supply feeding it needs to be 500 watts +/- do deal with the inrush properly and does not damage the power supply over time.

I will ask Julian to LAB this and see if we can see this behavior.

27V input and powering a similar load

We are using the WS-8-150-DC because we only have room for TWO 12v batteries in our box and because it is the most affordable DC switch in your product line. The WS-12-DC is $30 more with more ports than we need and it would require 4 batteries which increases cost and the box size. We're using the WS-8-150-DC for micropop sites with only 2 or 3 backhauls and a Mimosa A5. Our power usage should never be higher than 65-70 watts total and the power supply is rated for 155W continuous.

The AD-155B power supply is also a UPS so when grid power fails, it transfers the Netonix load to the 24v battery bank. This is a very popular setup from your users from people I've talked to.

Does this make sense to you why we went this route? You seem offended that we use this setup even though it makes the most sense for us financially and with the box space we have available to us. Your suggestion would be more expensive and require more space and we wouldn't gain anything that we need for these MicroPops.

No not offended just confused.

We like to see the SMART DC units hooked direct to the batteries and then a charger unit supplying / charging the batteries.

Also we get people trying to power the WS-8-150-DC and WS-12-250-DC with an insufficient power supplies.

If the power supply is our WS-12-250-DC SMART 250 watt then you should have a 500+/- watt power supply feeding it to deal with inrush demand of the switch power supply for longevity.

But we also see people trying to power a WS-12-250-DC with a 180 watts DC power supply then wondering why it will not work properly or even at all.

Julian reproduced this in v1.4.2 now he is trying v1.4.6rc14
He will then confirm it there and if it is there he will get Eric to fix in next release

We are trying to follow people's advice not to have a load more than 50% of the rated capacity of our power supply. So while the switch is capable of 150W, we are capping our usage at 75W to keep it under the rated 155W of the power supply. I understand where you're coming from about underpowering the switch.

The other thing people have told me is it's not best to have a load on the battery bank since the charger can be confused or not properly charge the batteries since it will provide more current than necessary since the load is on there. Thus the batteries might not float charge properly since the charger thinks the battery is never fully charged. Is this correct? Multiple threads on WISPA list suggest this which is why we went with the UPS/charger in one unit so the batteries are charged separate from the load.

Well when you do it this way the switches ability to pull your battery bank down as low as need be in an emergency is by-passed.

Many people off grid are using the switch connected directly to the battery bank and then a solar or wind charger connected to the bank. I would assume it would work the same way if your using an AC charging unit.

Personally I like a multi stage AC charger that can trickle all the way up to 10A per battery like the Minnkota charger which charges each battery separately so it does not care if the batteries are in series or parallel.

BUT back to your initial post, you have found a bug and this has been passed to Eric and will get fixed in the next release....thanks and sorry for the issue.

In my opinion it is fine to push a switch power rating to 80% of MAX constantly I feel 50% is way too conservative.

So a WS-8-150-DC with a constant load of 120 watts is fine and a WS-12-250-DC a constant load of 200 watts is fine too.

Okay, so couple observations:

When the low points on the graph are being drawn, the stated value for input current does not change, which tells me it's a graph-scaling issue rather than a sensing issue; In simpler terms, the picture isn't an accurate reflection of the power state. Will try to get a screen capture to show this, but the good part is that SNMP monitoring will return the correct value of ~1.2a.



I've heard from customers who use both setups, and from what I gather, they both have their high points and low points, complexity and redundancy being the double-edged swords they are. Your power design makes sense; 50% utilization is a general best practice when designing for reliability.