Solar Setup - WS-12-250-DC - Need Help!

[brendan@webperception.com]

Hello,

So I need a bit of help. We have had a Solar site up for about 1 year now. We originally had some major issues, but finally found a setup that worked for us.... until now.. We are located in Northern California and recently had fires in the area followed by rain. Because of the heavy smoke for a week, and then the 3 dark days filled with rain, our solar site went down in the middle of the night. It came back up the next morning when the sun came out, but has not been the same since.

Before I get into the story, please note there are pictures attached for reference. They are images of everything in our setup.

There is a Suniva solar panel (https://www.ecodirect.com/Suniva-OPT330 ... -4-100.htm) that connects to a Tristar TS-MPPT-60 controller. The controller then connects to the battery bank. Originally we had 3 batteries in our bank, 1 acid and two gel filled. We had them wired in series and ran electrical wire directly from the battery bank to a Netonix Switch (WS-12-250-DC). This switch powers 4 UBNT radios: A Powerbeam 400, a NanoStation 900Mhz, a RocketM5 and a Rocket Titanium M5.

With this setup, the Netonix switch would register the input voltage at around 40v throughout the day, and then at night it would drop down to 34-36v and when the Sun would come back up in the morning, it would bounce back to 40v. Since the heavy and smoke, we have seen the input voltage drop all the way down from 40v to 10v at night and then it would charge back up the next day to 40v once the sun started shining. This was obviously not enough volts for use to power our equipment and keep it up throughout the night, so we needed to do something.

We went out and bought 2 gel filled batteries and brought them up to the site and removed the acid filled battery that was there. We also ran a new cable from the 900Mhz radio that we thought might be the culprit (short or something). So now we have 4 gel filled batteries in serial. Last night we saw the voltage drop from 43v to 16v.

Why are the volts dropping so quickly now? What would cause this or where should I look first?

[sirhc]

Your firmware NEEDS upgraded.

Either v1.5.0 FINAL or v1.5.1rc8

v1.5.0rcX had issue we were working out.

Also please provide a wiring diagram how everything is hooked up.

Looks like your switch is connected to your charging controller and not directly connected to the batteries as suggested.


Also you cut off the top of all your images, best to show the Header of the web US.

Also the Status TAB showing watts per port draw.

[brendan@webperception.com]

Hi sirhc,

Thanks for your prompt response. I have upgraded the switch to v1.5.1rc8.

Below you will find the wiring diagram. I apologize in advance for my lack of drawing skills.

[brendan@webperception.com]

I also can not seem to get NTP working on this switch. This switch is on a private network (172.16.xxx.xxx) so I setup an NTP server on our private network as well. The NTP server is working and I have several hundred devices on our network that connect and use it.

[brendan@webperception.com]

It is 4:10pm. Over the last hour the switch has reported a drain of 8 volts.... It is a bit dark and rainy this afternoon...

[sirhc]

Well your diagram looks right, the switch is directly connected to battery bank and charger separately.

Could be you have a bad battery which is discharging the others.

Or possibly a bad charging unit?

So that solar charger is 48V correct?

Have you measure the voltage out of the controller during a clear sunny day, should be > 48V?

I would have each battery tested or if you do not have a tester you can swap them out 1 at a time and see.

[brendan@webperception.com]

So we ended up going up there last night with a generator and brought down all 4 batteries. Indeed 2 of the batteries tested bad and luckily were still under warranty. So good call on that one. We are going back up this morning to put the batteries back in place. One of my techs was doing some research last night and came across this idea and I wanted to run it by you to see what your thoughts were... This is what he sent me below:


4 Optima Blue top batteries 12v 55AH
We can wire them up in Series for 48 volt / 55AH= 2,640 Total Watts
Or wire up on pair in parallel and connect them all in series for 36 volt / 110AH= 3,960 Total Watts
Our panel is 36 volt 330 watts of total power.
But we can only drain the batteries down to a Total of 9V. So two batteries would be acting as one in the second example.

The image attached is from the link here: https://www.batterystuff.com/kb/article ... orial.html

[sirhc]

What is your purpose in this?

I would run all series 48V.

Keep in mind if you drain most 12V batteries below a certain level (see manufacturer spec sheets but a lot are 10V) you may need 10A per battery charger to recover them properly whereas most solar controllers are no where near this capacity.

So if you run a 48V bank down to 9V that is 9V / 4 batteries = 2.25V each battery which is DEAD DEAD DEAD!!!

You can prevent the switch from running them down this far by setting limits on the Power TAB but in emergencies you can chose to keep the site at up all costs knowing you will need to roll a truck with a heavy duty charger to recover.

Running GOOD AGM batteries (which OPTIMA is) down all the way every now and then is OK but if you have been repeatedly running them down and not properly recovering them this could damage them.

Discharging a battery so low then not properly recovering them can cause shorts in the battery.

[erikwebper]

Hey sirhc,

I'm the Tech who went up and replaced the batteries today.

It does look as if the other two batteries were dead and draining our system rapidly.

After running our generator for 17 hours (Predator 4000 from harbor freight) , we got 4 new Optima batteries in.

Our switch is holding high up above 49 volts, and only dropping .1 volt per hour so far.

We still have 4 Ubiquiti radios plugged in.

(one rocket m5, one rocket titanium, one powerbeam 400, and one nanostion 900m)

Looks like we may have solved the problem! We will know forsure in the next couple days as it is still rainy.

The next steps would be to test all cables, and radios, and or replace radios.

[slimey]

Just a couple of observations on your use of the Tristar controller...

- Use the cut-outs to route your cables into it and get the lid back on it - I wouldn't be happy without, as it's easy to accidentally touch the DC terminals.

- When you had 3x12V batteries connected to it, you were trying to run it at 36V. The Tristar will (if I remember correctly) run at 24V or 48V, so depending on which it was attempting, you were either over charging or under charging your batteries.

- It looks like you've connected the temperature probe to the earth connector on the back of the metal case! It should be connected to one of the green connectors on the circuit board.