Need help debugging solar installation

When my 4 solar panels are all full capacity (bright, sunny day), and the batteries are relatively full, the voltage across most of the batteries (reading either individually, or in groups/parallel) can read 15+ volts, and my inverter shuts down. If I turn off the solar panels (on the controller), the batteries will go down to reading about 12.8 - 13.4 volts. As soon as I turn on the solar again, the batteries go back up to 15+ This is both confusing, and also I wonder how can I ever truly charge (top) my batteries.

I'm attaching my wiring diagram (excuse the rough .pps format - I don't have any fancy wiring diagram tools), with the battery and solar panel specs on the second slide.

My question is - it seems I cannot get to the max charging capacity with this set up. Once the batteries get CLOSE to full charge (13.4, say), the voltage recorded on the batteries goes above 15.1 when charging, and I have to stop charging so I can use the inverter. Ironically, once the inverter shuts down, the battery voltage will go even higher (if the solar panels are connected).

I get that there is a difference in the voltage reading on the batteries when they're charging, but the difference in the charging voltage vs. resting (float) voltage seems unusually high.

Also, I thought my battery charge controller was supposed to take care of all of this. When I call Morningstar (I now own 3 Morningstar controllers), they are NO help - in fact, their tech guys have been pretty evasive (well, except for one, who cleared up a question I had about wiring the battery sensor, but did not end up providing a solution/answer to the battery over (or under?) charging issue). It's like they don't want the little guys (like me) getting into solar (?) go figure.

Can anyone help me understand what's going on, and how to fix it?

Comments

  • sueco
    sueco Registered Users Posts: 22
    Re: Need help debugging solar installation

    (ok, here are the attachments, in .jpg - since the forum doesn't take .ppt format)
  • waynefromnscanada
    waynefromnscanada Solar Expert Posts: 3,009 ✭✭✭✭
    Re: Need help debugging solar installation
    sueco wrote: »
    (ok, here are the attachments, in .jpg - since the forum doesn't take .ppt format)

    Looking at your photo, I don't see where the + is wired from terminal on controller marked "Batt +". The "sensor" connections look OK, but something's missing in the main current carrying circuit. Nor do I see the connection from PV Neg. to battery Neg.
    From your description, it sounds like either the controller is defective/blown, or the panels are in effect wired directly to the batteries, bypassing the controller. The controller should normally limit the voltage to more or less 14.5. That's a ball park voltage. BTW I have one Morningstar controller on my micro hydro, and two of their inverters in use 24/7 and have found them to be of top quality.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Need help debugging solar installation

    Everything looks OK to me on your drawings.

    The only things I can think of (guessing on my part):
    • Is Prostar set to "GEL" battery type?
    • Prostar will assume 24 volt battery bank if connected to >15.5 volt battery bus voltage. When you connect (first power up meter), do it in the "dark" (no solar panel power) and make sure that the battery is <15.0 volts when connecting controller. What does the controller indicate the battery state of charge is? Low, medium, or full when at 15+ volts?
    • Batteries are bad (batteries have little AH capacity left) and the controller is overshooting set-point (unlikely).
    • Controller is very cold (sub freezing) and is increasing battery charging voltage. What is the temperature of your controller/battery bank?
    • Controller(s) is bad.
    About all I can think of...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • sueco
    sueco Registered Users Posts: 22
    Re: Need help debugging solar installation

    First, thanks for your time/attention to look at this, and your willingness to lend guidance.

    ??? Is Prostar set to "GEL" battery type?

    SDC I meant to mention the controller is set to gel cell - as I knew you would ask :D

    ??? Prostar will assume 24 volt battery bank if connected to >15.5 volt battery bus voltage. When you connect (first power up meter), do it in the "dark" (no solar panel power) and make sure that the battery is <15.0 volts when connecting controller.

    SDC: I THINK this is fine. When I'm debugging this issue, I "reset" the controller, and it comes up as 12V (first reading). So, at least as far as the polling routine goes, it's aware that this is a 12V cell/battery bank. And I've done this quite a bit, so it's consistently coming up as 12V.

    ??? What does the controller indicate the battery state of charge is? Low, medium, or full when at 15+ volts?

    SDC: In all truthfulness, it's hard for me to remember, and I'll look at that today/tomorrow (next time the charge is near top/full). My routine is to turn off the charging coming in from the solar immediately, so the inverter alarm will stop. I'll make sure I look at that next time around BEFORE turning off the solar.

    ??? Batteries are bad (batteries have little AH capacity left) and the controller is overshooting set-point (unlikely).

    SDC: I thought of this, and wondering how to tell. I am currently using my ProStar15 as a battery meter, as I try and debug this. The voltage SEEMS consistent across all batteries individual, and all combinations of batteries (1, 2, 3, 4 in parallel). I THOUGHT that if there was a bad battery, it would not match the others, at least initially, as it would have more impedance than the others. However, the batteries all SEEM to read consistant (all high when above 15+ V, and all the same resting voltage immediately after I turn off solar charging).

    ??? Controller is very cold (sub freezing) and is increasing battery charging voltage. What is the temperature of your controller/battery bank?

    SDC: We're at about 70*F, a little colder at night, but not too much. And, yesterday was hot, but the batteries are in the garage, which was around 75*F. Also, controller sits on a panel mounted above the battery bank.

    ??? Controller(s) is bad.

    SDC: Again, how do I tell this? I have done a bit of research, and the industry seems to agree that Morningstar is high quality - I don't doubt that. But I must say, it can be the HIGHEST quality on the planet, and if they're not willing to troubleshoot with me, then all that quality does me no good.

    Again, thanks for your time - I'm open to suggestions of things to try - I've run out of things to try that I could think of.... that I was sure were safe.
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Need help debugging solar installation

    call it a hunch, but try disconnecting the battery voltage sense wires and see what happens.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Need help debugging solar installation

    I would suggest you get a "decent" DMM (digital multi-meter). If you have $60 to burn--I would suggest this DC/AC Clamp Current / Multi-Meter as a great entry level tool for your bench (i.e., good enough for the $$). There are better meters too (True RMS reading) for more money--Can be nice--but only really needed if you are working with MSW inverters or will be heavily into measuring "actual current" into large computer power supplies/battery chargers, etc... RMS (root mean square) is the "proper way" to measure non-DC / non-sine wave current & voltage. Nice to have--but not needed most of the time.

    Three points I would like to see voltage measured... At the Solar Panel input, the controller battery connection, and right at the battery bank (both daytime and night-time)--And what the controller "thinks" the battery charge is (there is an LED that indicates the charger's mode).

    Per Niel's suggestion, also check the voltage at the "battery sense" input (and that you have them wired correctly--heavy battery leads to controller output--and use lighter awg wiring for sense leads).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • sueco
    sueco Registered Users Posts: 22
    Re: Need help debugging solar installation

    Again, thanks for the suggestions/inputs.

    Here's my goal - to CORRECTLY wire/connect all components of my solar system, and have it work correctly. I'm way ok with any work around necessary, but I'd like to understand the workaround (why it works, and what it's fixing).

    That said, I have tried every combination of wiring the sensors you can imagine. What worked best (allowed the most charge without "sensing" too high a voltage) was the negative on an inside battery (not on a terminal wired for charging/load), and the positive not wired. However, WHY this worked best is a mystery to me.

    I have 4 guage wire on ALL battery connections (except sensing wires). I have 8 and 10 guage wire on all wires to solar from the terminal blocks (connected to the batteries, as shown on the diagram), depending on usage/load. I used 18 gauge wire (more than needed, because their only about 10 feet each, but it's what I had in the house) for all sensing. There's a 2 amp fuse close to the battery for the battery + connection to the battery meter. I have checked and double checked, and the wiring diagram - (which I made painfully from scratch) is exactly what's wired now.

    I'm starting to collect a log, taking current/voltage both while charging and while not charging. It is more difficult to take readings without the load, because I have to crawl under my desk (long story) twice to do so - bad design, I know - I'll fix that later.

    What it FEELS like has been happening: as the voltage on the batteries starts reading high (14+), the current on the solar is greatly reduced (unfortunately, reduced below my load). If I turn off my solar charging (on the controller), the resting voltage is actually around 12.6 (13.4 at the most). If I turn off the load, to try and top off my batteries, that's when the voltage goes about 15 V. I can no longer just turn on my inverter (accompanied by the most obnoxious alarm), so I have to turn off the solar charging. Once charging is off, I can turn my load back on, and then solar back on. But, now I'm back to where I was: solar charging is registering very low, the load is higher than the reduced solar charge, I'm now draining my batteries instead of topping them off. I can run like this forever, but I thought I'm supposed to top off the batteries for better battery life. And, when I say "top off", I'm trying to get the resting (float) charge to at least 13.8 - 14.0 would be ideal.

    I'm going to keep a log, so I can more accurately say what's going on. Again, I appreciate all of the advice/guidance. I'll also look into buying another meter - if it's required to get this resolved, then I'll do it. However, I bought the higher $ controller w/meter to see what was going on. Then I bought the battery meter, to see what the actual charge/flow through the battery bank. Now, another meter to??? get more readings???? again, if it's necessary, I'll do it.... but, I'm thinking the meters are not, in fact, fixing the problem, and I'm trying to move my garage to solar, and would like to spend the money on panels/batteries to make that happen instead.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Need help debugging solar installation

    If MorningStar has a problem (and it sounds like they may)... From what another poster said it appears the custom settings are causing the problem.

    Can you set to "Flooded Cell" and see what happens?

    I believe you are seeing real problems--and you are correct, a DMM is not going to fix the problem--just help you diagnose. But if it is a vendor problem--and they are not helping--not a good situation.

    I cannot find the post at the moment--but in the last day or so, a poster is saying that Morningstar has a long term design problem with their controllers (over voltage with non-standard flooded cell settings). And they are refusing to acknowledge the issue.

    Would a Xantrex or other PWM charge controller be acceptable?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • PhilS
    PhilS Solar Expert Posts: 370 ✭✭✭
    Re: Need help debugging solar installation

    http://forum.solar-electric.com/showthread.php?t=11205#post84111

    Here's the post you were referring to Bill... from Not So Bright post #16

    GOOD finding by him!

    Phil
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Need help debugging solar installation

    Thank you Phil:
    I have now tested four brand new sunsaver duo controllers and all of them go into over voltage. I have seen as high as 16.78 volts from them. the company is not admitting the problems but I have found others on the web that has experienced this. ms denies it is an issue.

    my 210ah lifeline AGM has popped the valve because of the ssd overvoltage.


    Ive been testing this unit extensively and it seems to only occur with custom settings (AGM specs) and on battery one only so far but oddly it will not go into over voltage while connected to pc with ms adapter while monitoring.

    I am able to recreate this overvoltage condition now that I have learned more and can direct others on how to also.

    their RM-1 meter is also not acurate and rarely shows the overvoltage but has on occasion.

    I have been recording voltages from two of them at the same time on the same batteries, one connected to pv and the other not. they do not agree with each other on voltages, not even close.


    I want ms to do the right thing but its doubtful at this point since they deny the ssd has any issues. Im gathering info from other ssd owners that have noticed the problem right now.

    when I am satisfied I have enough I will aproach the ms execs about refund and damage reimbursment.

    Im hopeful they will deal with this honerably.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • sueco
    sueco Registered Users Posts: 22
    Re: Need help debugging solar installation

    Thanks for all of your responses. I think I've learned a little from this solar thing, and have other questions. I'll start new posts for those.

    As far as the ms controller, I agree that it's not working as spec'd. If you need more ammo, I can give you my logs. At this point, I am stuck with them, and have created work arounds. Moving forward, I will not use ms, until this can be resolved.

    I am in the process of setting up a Kansas City "solar for the average joe" resource, which I hope to run after I retire as a service project. Thanks to all for your answers, and patience - really THIS is what's needed - being able to get good answers and guidance for simple/pratical questions.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Need help debugging solar installation

    You are very welcome Sueco.

    Please drop by when you can... This forum is the sum of its participants.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset