Switch from 100 W to 50 W Panel

Razed11

Our current IoT power plant consists of a 100 W Renogy panel, a Morningstar SS-6L-12V (6 A limit, PWM with load disconnect) charge controller, and a 12 V flooded lead acid battery (90 A @ 20 hr). Our system draw is about 4.2 W and is pretty consistent.

Under these conditions and with a relatively healthy battery I can observe the three charge phases. On a typical unit the absorption phase is reached in the early morning. Before I get to my question I thought I'd share a plot as it might be interesting to those learning about charge controllers. This plot is from yesterday and it's centered at 12 noon (forgive my typo):

This unit is in the midwest and has a fixed tilt. It's also on top of a 10 story building. The charge controller has temperature compensation and you can see the set point voltages for absorption and float dropping as temperature increases. Note that it keeps dropping at the end of the float stage despite the drop in temperature. I believe this is because the controller is enclosed in a black waterproof box and its temperature lags that of the air.

I wanted to get a feel for how a 50 W panel from the same vendor would perform so I setup things at my home near 24/680 in the San Francisco bay area. Not having anything to mount it to I placed it on my driveway and shimmed it until it was horizontal. I'm using a battery from a different store front but with the same rating (could be the same battery for all I know). Rather than attach one of our systems I used a programmable load configured to constant power and dialed in 4.2 W.

The absorption and float stages never seem to be reached. I don't have data logging with this setup but I'd check it often through out the morning and afternoon. At best it gets to just over 13.0 V. The date sticker on the battery is 5/17 and I'll admit that I have not checked its specific gravity (I don't have a hydrometer but will order one shortly).

The IV curves under 1 Sun irradiance have similar open-circuit voltages but the short-circuit current is approximately half.

I knew that my horizontal panel had a disadvantage but given that its fixed tilt brother was charged by 9 AM I figured it would easily be charged by the end of the day. By this I mean it had completed the absorption phase of charging. Any thoughts?

mvas

How many amp(s) was the new setup (in the driveway) pushing into the battery?
Maybe, just over 1 amp?
What is the Voc of the battery, after 24 hours rest with no load?
Can tell if the new charge controller was in Bulk mode?
What is the voltage on the PV Panel when "charging"?

In the original setup, is the flooded battery inside the black box?

Estragon

Also interested in the answers to mvas's questions, especially whether the battery is also in the box. If not, there's a good chance the controller is dialing back absorb based on the temp in the hot box, but the cooler battery needs higher voltage to charge properly.

As well as the tilt, another factor may be location. Midwest could be 1000' feet or so higher in elevation with dry air. The bay at sea level with haze may not get 1000w/sq.m the STC ratings are based on. Most of us don't.

Razed11

At the moment I don't have a good tool to measure current in to the battery but I'm in the process of ordering something. I have a multimeter but I'd like to get a clamp ammeter. I'll use the meter to measure Isc and report back. Quite an overcast day today which is unusual this time of year.

Voc of the battery was 12.6 V. I measured this when I received the battery so presumably it had been sitting around for more than 24 hours.

The LEDs flash at different rates if in absorb, float, or equalize mode. The states I've observed are solid green. It's now solid yellow because I intentionally placed a 50 W load on it for a while.

At the moment I'm measuring 12.4 V at both the panel and battery terminals.

Regarding the black box: The field units have the controller in a small black box and batteries in their own black boxes. My home experiment has the battery and charge controller in my garage and the panel just outside on my driveway. It can become a hot box of its own on the warm summer days. Morningstar publishes the temperature compensation coefficients so I'll review those. It's a good theory but I think the LEDs will indicate the absorption and float phases.

myocardia

San Francisco is at 38° latitude. Your solar panel(s) in San Francisco should then be mounted at 53° from horizontal, facing the south. At the 0° from horizontal that you said in the first post that you have the 50 watt panel, you are costing yourself a considerable amount of power.

Razed11

I received my DC clamp ammeter yesterday and took some measurements. It was only putting 0.2-0.3 A out of the panel and into the battery (though the charge controller). Measured the panel Isc to be like 3.1 A in pretty good sun. Took it apart, checked cables, reassembled and now all is well (I think the battery negative was loose). Fairly dark clouds today with full sun in between. With the battery at 12.6 V I'm getting 0.5 A with the clouds and 1.8 A when the sun reveals itself (that was at 10:45 AM). Nice to have the right tools.

@myocardia
I'll get that direction an inclination straightened out. Thanks for the numbers.

Razed11

Hm. I don't think this was a wiring problem after all. I'm wondering if I'm having a thermal issue. After some time I returned to it to find it only pushing 0.3 A or so in to the battery. I brought it out of the sun, let it cool off, and returned it, and I got much higher currents.

Any thoughts on measuring the panel temperature? Can I can shoot the back of it with an IR gun? I was reading 50 C on the backside. Rather than lay it flat on some cardboard I propped it up at an angle to allow some airflow.

mvas

Yes, heat will lower the Vmp. Solar Panels prefer to be cool 25°C

Estragon

0.3a may be fine if that's all the battery will take. It will only take the max current the panel will produce if the battery is at less than ~75-80% full. As the battery gets closer to full, it will take progressively less current at a given voltage.

Razed11

Disconnected everything and connected it to the programmable load set at 12.6 V constant voltage mode. When enabling this load it starts off just under 20 W and then slowly drops off to 6 W. If I disconnect and reconnect the load it starts off at ~20 W again. This was around noon and a perfectly clear sky. This is same behavior observed when connected to the charge controller and the battery charged to around 12.6 V. I'm assuming the panel has been damaged in some way. While I don't expect 50 W I would expect 35-40 W under these conditions.

At this point it is somewhat academic. I'm just curious what could cause this kind of failure mode (and I've sent an email to Renogy technical support). Measuring Isc and Voc under theses conditions gives values very close to the spec. But placing a load on it causes odd things to happen.

solarvic

Is your driveway blacktop?  They had a story on the tv news from Youngstown ohio about how hot blacktop is. It was about pets and you should not let your dogs walk on Blacktop. They said your dog could burn it,s feet and have a heat stroke. Cement is supposed to be cooler than black top.  Maybe your smaller output has something to do with a heat problem???

Estragon

I'm a bit confused by the load at "12.6v constant voltage mode". Is it a resistor that will have variable amperage?

Razed11

Well I received a new panel from Renogy and it appears to work just fine. It's 11 AM and I'm pushing 2.48 A into the battery at just over 13.0 V. 

@Estragon
This article on electronic loads has a good explanation of a constant voltage feedback system. That piece of test equipment was quite handy as it displays the power, current, and voltage. When engaged in CV I can see these values in real-time as it emulates a battery. When disengaged it shows me Voc. It also has a short circuit option.

Here's a link to the model.

@solarvic
I was concerned about thermal effects initially but found similar behavior early in the morning when it was cool. My driveway is made of light tan colored pavers, the sidewalk is cement, and the cul de sac is asphalt. I can't stand on any of them too long with bare feet (which I find myself doing too often).

Estragon

I see what you mean now. Looks like a handy piece of kit. Sounds like maybe a defective connection in the panel maybe heating up over time and slowly pulling output down.

I bought a couple of Renogy panels recently that won't get installed for a month or so. Hopefully yours was an isolated problem, not a bad run!

Razed11

@Estragon
We have deployed hundreds of units with Renogy's 100 W panel and have only had one failure. The solder on one of the diodes gave way.

Today I received an email from Renogy saying that they'll send me a shipping label to inspect this funky 50 W unit.

Estragon

Good to know. Likely just a random problem.