Glitches and Problems with Midnite Solar Classic 150

[dharry]

Since energizing a new solar system on September 2nd, 2014, I’ve identified several issues with this controller (C150) and am curious if others have noticed these things. Two issues are mostly annoyances but one problem results in unnecessary battery drain and I’ll describe that first.
This is an off-grid, stand-alone system without generator or other back-up system. All power comes from 2.4kW of panels with a 235 Ah, 48V battery bank. The CC150 is paired with a Magnum 4448 PAE Inverter and RC-50 Remote display.

1) Sometimes, there is a significant time lag after a load is added to the system before the C150 adjusts and allows more power from the panels in to cover the load. During that lag, the batteries are carrying a large load even though there is an abundance of solar energy that is not being used. Then, suddenly, the C150 kicks in and the solar power soars and covers all the load and starts recharging the batteries (there is no change in sun availability during this time). I have seen this happen during Bulk, Absorb, and Float modes. Other times it works perfectly. It bothers me that my batteries are put under load and taken well down even though there is more than enough solar energy to handle the loads. Anybody else notice this happen?

2) The “Status” display on the C150 often (some days it’s perfect all day, other times it comes and goes throughout the day) shows incorrectly low values for Wattage and Amperes. For example, the C150 shows 500W at 50V, 10A as the total energy available in the system but the RC-50 monitor indicates the same voltage with batteries charging at 12 or 13A. I have verified with a meter that the RC-50 values are correct. It is especially obvious when the C150 shows 0 watts in the system, yet the batteries are charging and of course there are losses in the inverter as well. If a load kicks in, the C150 displayed values increase but still under report the amount of energy that is being used. This is annoying more than anything. The C150 appears to be diverting power correctly, just not displaying it. Midnite Solar has been responsive even though they say they have never seen this problem before. They sent a new display, which did not correct the problem and now an entire new charge controller they sent does the exact same thing. This indicates to me a widespread quality issue and I am amazed that I am the only one who is experiencing it.
3) A small glitch but I’ll mention it anyway: as you scroll across the maximum power data under the Daily Graph mode (in Logs), the displayed values truncate the thousands so all you can see is the last three digits. I made Midnite Solar aware of this and they said they would fix it in the next Firmware Update.


Any similar experiences with #1 or #2 above would be appreciated. There appears to be no fix for either that I am aware of.

[Mike90250]

I'm a Midnight solar user, have a Classic 200.

Midnight has had a string of minor issues, and general start-up bad luck. I keep finding out new stuff all the time, and let them know too.

But I've not had any issues like you describe. I've discovered that my Morningstar MPPT works better in cloudy weather than the classic, the classic has higher stand-by losses, both day and night, and the little battery in the graphic display panel, often is rendered non-functional (causing memory loss) from sticky goop from the pull-out shipping tab that insulates it. Works great in full sun, and has response in a couple seconds (remote wi-fi local app monitor) from loads, just like the morningstar does. I did hear of someone else with bad switches in the graphic panel.

But I would say you have a lot of panels for a small battery bank, and that may be causing some issues.


Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV || || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A | Classic 200 , 2kw 160Voc array

[dharry]

Thanks for the reply, Mike

I have another battery bank on order which will double the storage size.

I am curious why having a small battery bank relative to the panel output causes problems. I hadn't considered that.

I should clarify that the lag time I mentioned before the charge controller "adjusts" can be a minute or more. On a sunny day with plenty of sun to spare the batteries still can recharge no problem but on cloudy days where every bit matters, it's a real drag.

I am also curious why so many industry types say the Classic is the best controller out there. In what ways is it better than the Morningstar or the Outback? My, admittedly very limited, experience with it has been mixed for sure.

[Sunking]

I am curious why having a small battery bank relative to the panel output causes problems. I hadn't considered that.

Simple LEad Acid batteries are a chemical reaction and takes time to convert electrical energy into chemical energy. This limits the maximum charge and discharge rates to about C/8. Go above that and you spike the voltage up beyond gassing voltage and the result is heavy water use and heating as the extra energy is not converted and burned off as heat. What you describe sounds perfectly normal.

If you have say 100 AH batteries, you do not want to charge with more than 12.5 amps. To determine your charging amps = Panel Wattage / Battery Voltage.

[Mike90250]

I should clarify that the lag time I mentioned before the charge controller "adjusts" can be a minute or more. On a sunny day with plenty of sun to spare the batteries still can recharge no problem but on cloudy days where every bit matters, it's a real drag.

I've never seen this happen on my system. I see a 5-15 sec lag, which I chalk up to the wi-fi link update. My controller is 300' away, I don't camp out at it.

I am also curious why so many industry types say the Classic is the best controller out there. In what ways is it better than the Morningstar or the Outback? My, admittedly very limited, experience with it has been mixed for sure.

I guess midnight runs a great hospitality suite at the shows? I'm not impressed with it. My next controller for an array for an outbuilding, will be a morningstar. Silent, cool, thrifty, works in cloudy conditions.

I have a morningstar and a nearly identical array on the midnight, and in full sun, they are equal, in overcast/clouds/fog, the morningstar blows the classic away. Far away. In another galaxy. I have reverted to P/O mode (a very ancient MPPT algorithm) and it's behaving much better, but still not as good as the morningstar.

[dharry]

Mike - Thanks so much for that reply! I just switched over to P/O mode (even though I have not had a chance to find out what that really means) and the controller immediately started behaving better, at least in so far as the displayed values are reasonable, which they hadn't been all morning. I am cautiously optimistic that it will stay good and I very much appreciate your pointing me toward that.

I am also very interested in your ability to use WiFi to download or watch data remote from the controller. How are you doing that? Do you have a PC hooked to the charge controller and then use its WiFi to broadcast to another computer in your house? I was not aware of any Wifi routers that had a 300' range, which is almost exactly what I would need to monitor that controller from my house as well. Any info that you would be willing to provide on how I could rig this would be much appreciated!

Sunking - Thanks for providing the info about potential overcharge rates for a big array with a small bank. However, the problems I stated about my Classic 150 don't seem to be related to charge rates. As I said, the "display" problem is that the display shows less energy available in the system than is clearly being used (as verified by the RC-50 battery monitor and secondly by an amp meter on the battery cables to the inverter) . A charge controller that indicates no power coming into the system while a battery is charging at 2 amps makes for an interesting efficiency calculation, don't ya think?

I also don't understand why you would say that everything seems perfectly right, given the display issue and the fact that at times the controller doesn't allow solar energy into the system. Instead, batteries that are already floating are forced back into Bulk mode to cover a load even though there is an abundance of solar energy available that is not being used. That sure doesn't seem right to me, certainly not ideal to take the power out of batteries when it could be diverted directly from the panels. Plus, isn't it a bad idea (potential overcharging) to put the batteries through another Bulk/Absorb cycle when they have already completed it for the day?

[Mike90250]

I am also very interested in your ability to use WiFi to download or watch data remote from the controller. How are you doing that? Do you have a PC hooked to the charge controller and then use its WiFi to broadcast to another computer in your house?

I have a simple, 4 port linksys router in the battery shed. I've added a +9db directional patch antenna to it. I have a +26db flat panel antenna wired to a Alfa hi-power (from Amazon) USB-WiFi adapter. The Alfa has a better reciever circuit in it, which is why it can hear the linksys. I plug the Alfa USB into the laptop, and connected. Combox, Morningstar, Classic all plug into the ethernet of the router and I'm set. My Line-o-sight is through 3 walls, roofs and heavy brush. With the Alfa receiver and the 26bd gain antenna I can use hotspots 3/4 mile away.

[dharry]

Mike - Thanks much for that description, it sounds like exactly what I need to do.

I switched over to Legacy P/O mode and for the first time since I have had that controller, the displayed values made sense for the rest of that day and all day to day until I had to go back to Solar Mode. That was awesome and clearly indicates there is a bug in the Solar Mode algorithm. However, while on Legacy Mode, my well pump kicked in and the controller did not react. I was Equalizing at the time and there were no other loads on the system so it was using around 500 watts. The pump kicked on and the controller kept the same power level as the battery dropped from 62.4V to ~50V over 4 minutes. At the end of that, I kicked it back over to Solar Mode and then the controller kicked up to around 2300W to recharge the batteries. I cannot believe this is the way controllers are supposed to act! Taking batteries through a rapid discharge and then a rapid recharge seems like the worst way to treat them. There was plenty of power available for the battery to have stayed at full equalization voltage while the pump was running - it just wasn't utilized by the controller. How can this possibly be correct?

Do the Morningstar and Outback controllers behave this way also? How about PWM versus MPPT? By the way, while the batteries were getting drained, the controller displayed "Equalize MPPT" which to me says it should have been letting in the extra power that it needed from the panels rather than discharge the batteries. Am I missing something here?

One possibility I plan to test tomorrow is that the Soft Start on the well pump is such that the controller doesn't recognize a load has been added because it loads up over 2 seconds. Seems unlikely to me but I will test with other large, inductive loads tomorrow to see if I get the same results with them.

[Mike90250]

A charge controller is only interested in maintaining a preset voltage on the batteries. It can't tell if there is a load or not on the batteries, it should see the voltage drop 0.001 volt, and apply more power to keep it constant. If it runs out of sun, then the voltage sags. I've never seen the behavior you describe, with any controller I've worked with (trace, morningstar, classic). I'm thinking you may have a flaw in your wireing or connections, and the controller is not able to function properly.

And remember, even a tiny bit of haze can/will cut solar production drastically. Not just 10%, but more like 50%. When cloudy, I get less than 9% of PV nameplate power.

[dharry]

Mike - That's extremely helpful. If the charge controller should cut in with a 1 mV change in the batteries, that explains why when my works, it is seamless/instantaneous. There is some glitch, either in the external wiring to the controller or within the controller itself (my vote) that does not see the V drop sometimes. As I said, my batteries are allowed to drop several volts with no action from the controller - and this is with no shortage of sunlight. I live in sunny new Mexico and when it's clear, as it has been, my 2400 W panels can turn out over 3K and yet the battery is discharging and the controller is using allowing only 3-600 watts in.

I bought this as a pre-wired power center from Northern Arizona. The power panel came assembled, from Midnite Solar I believe. All the major wiring looks great and I have checked tightness on everything. Flaky grounds can sometimes cause weird intermittent problems but they all seem fine. It's brand new, no rust.

The intermittent charge controller issue combined with the intermittent display problem makes me think it is internal to the controller. I sure wish I had a non-MidNiteSolar controller that I could hook up just to see how it worked.

Thanks for your help Mike. Is that 1 mV battery drop action level shown somewhere so that I can wage the battle with MidNite solar and maybe get some action on this? I don't see it in the spec sheets anywhere. Or is that just something you have gleaned over the years?

[Sunking]

It was a figure of speech in an example. 1 mv is not enough to really be measured when operating at double digits voltages. Not even a volt meter can pick that up as it would be outside the accuracy range of 1%. 1 mv to 10 volts = .01%.

[dharry]

Sunking - Thanks for the reply. I'll buy that, but then what is a reasonable amount of voltage drop before the controller wakes up and accesses more power from the panels?

[Sunking]

Sunking - Thanks for the reply. I'll buy that, but then what is a reasonable amount of voltage drop before the controller wakes up and accesses more power from the panels?

First requirement would be the controller would have to be in FLOAT MODE because in Bulk or Absorb the controller is already pumping as much as it can so any additional load would just draw the voltage down further. Then even in Float Mode if the load is demanding more than the panels can supply at that time your voltage is going to sag because the batteries have to make up the shortage.

Just a hunch here but I believe we already discovered your batteries are too small for the panel wattage you are running and I suspect you are putting too much load on them and they just cannot support the load demand and thus you see voltage sag. Try this after sunset. Turn everything off, measure the voltage on the battery term post or Open Circuit Voltage. Then hit them with a load by turning stuff on, then measure the battery voltage again and see what you get. If the voltage drops more than 2% (.24 volts for each 12 volts or 1 volt on a 48 volt system) would indicate too small of a battery or a battery on its way to being a Boat Anchor.

[sunny79]

Out of curiosity do you have negative PV landed to the classic? My owners manual said it was optional which seemed just wrong so I landed negative PV directly to the controller.

[dharry]

Out of curiosity do you have negative PV landed to the classic? My owners manual said it was optional which seemed just wrong so I landed negative PV directly to the controller.

Hi Sunny, the Classic chasis is grounded directly (dedicated wire) to the ground bus bar in the power center. There is a separate dedicated line from the Classic to the DC negative in the power center, which is also where the negative PV connects in. This is what you were referring to, right?