PV system check

zoneblue
zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
Hi,

Having some background in the electrical field i seem to have taken an overly complacent approach to our PV design. Having made the bigger purchases i now find myself embarking on crash course of reading including almost a full day on this forum. Good stuff, thanks!

Id be grateful for any feedback on things ive missed and a couple of remaining problems.

The draft schematic is
Attachment not found.
www.zoneblue.org/files/draft-pv-schem.png.

Ive also done some daily/insolation/battery SOC modelling here:
www.zoneblue.org/files/stand-alone-pv-perf.xls

In short its an off grid 24V system 1.8kWp PV, lots of DC loads, and a need to do full datalogging (compulsive analyst). Its 24V because i inherited near new 24V AGM, and we have a 24V fridge. The invertor will only be used for the washing machine and powertools, in an outbuilding. Low EMF life here we come.

So questions:

Charge controller. I'm not really happy with any of them with the possible exception of FX80. As its not quite right for us, im thinking about using a temporary controller (TS60) until something better comes along. In my ideal world a CC would do several things. Charge, LVD for my DC loads, monitor load on battery, and lastly an efficient and reliable load diversion to hot water heating. This system has stacks of surplus power, about 5kWh/day by my reckoning. That's a fair bit of hot water. Now the only solution that comes close to utelising all available power to my mind is the FX80 with its AUX SSR mode. But it still seems patched on to me, an afterthought. Data export is also important to me and FX80 requires more gear to achieve this out of the box.

My main question is how hard can i push the TS60? The datasheet says 1600W of panel at 24V nominal. Morningstars string calculator says 1800W is OK. My support email to morningstar advised limiting to 5 max 300W panels. This might be butt covering on their part, but my inexperience with PV in this particular geography leads me to some caution, and i certainly know from experience how these microprocessor powered devices can fail, either short or open. The latter isnt going to do the AGMs any good at all.

Anyway Isc STC for the array is about 52Amps. Without mppt and a fixed array angle, absence of snow etc, leads me to believe that the TS60 should work. Ive heard they can run at 60A happily ad infinitum. However here in NZ our PV does tend to be positive rated, (ozone hole or something).

Im aware that a mppt controller would allow me to run the panels at 2 strings of three reducing the 20m cable size. But at the cost of another $900 for a mppt controller. When i replace the tristar i can retire it as a LVD or 'large' SSR for the hot water.

Second question, reading the heated arguments about battery monitors, amp counters and their problems, im not really sure whether its even worth spending the 300 dollars on the recommended votronic. And its single shunt does little to reliably inform about in and out seperately. Instead id rather spend that money on a datalogger and set up a tablet or something slightly homegrown. Meanwhile a simple LED panel voltmeter, checked first thing in the morning.

Anyway i tend to write a lot of dribble so ill leave it there. Thanks for any 2c's.
1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
http://zoneblue.org/cms/page.php?view=off-grid-solar


«1

Comments

  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    Re: PV system check

    welcome, first thing I notice is that you need a fuse on each PV when you combine more than 2 in an array.
    add: you will want a PV disconnect before the CC
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: PV system check

    Also, you do not want to connect solar array frame ground to DC return line at the array--Solar +/- should be fully insulated wire from the array to the charge controller--No earth/safety grounding.

    And, as WestBranch says--Generally, you only need "combiner box" fuses/breakers if you have three or more parallel strings. One or two parallel arrays do not need fuses for safety.

    And you should run a "green wire/safety ground" from your common ground connection to each metal "box" (charge controller, inverter, loads, metal DC load center, etc.).

    Keep leads from battery bus to loads short/heavy copper wires. Fuses/breakers should be located "near" battery + bus connection (fuses/breakers protect "down stream" wiring from overheating if shorted/overloaded).

    Regarding your battery bank, review this website for how to parallel batteries (if you will be) in a bank for better current sharing.

    Also, your battery bank 24 volt and what total bank AH capacity (12x2??). A 400 AH x 12 volt batteries with 2 in series?

    A 4kW inverter is pretty big... Huge for a 12 volt battery bank. And while an AGM battery bank can supply scary amounts of current (they are used for UPS systems that can drain a battery bank in 15-30 minutes).

    A fully loaded 4 kWH inverter could drain that bank inside of 1 hour...

    And the fusing/branch circuit for a 4kW inverter on 24 volt bus should be:
    • 4,000 Watts * 1/21 volt batt cut off * 1/0.85 inverter eff * 1.25 NEC derating = 280 Amp branch circuit/Fuse/Breaker rating

    100 amp shunt is way too small for a 4kW inverter with 200 amp fusing (unless you are really planning on using less power than 4kW)... One poster here (John P) has good experiences using shunts as fuses too.

    Note that shunts add voltage drop too... Making battery charging voltage a bit less accurate (along with wiring drop).
    Charge controller. I'm not really happy with any of them with the possible exception of FX80. As its not quite right for us, im thinking about using a temporary controller (TS60) until something better comes along. In my ideal world a CC would do several things. Charge, LVD for my DC loads, monitor load on battery, and lastly an efficient and reliable load diversion to hot water heating. This system has stacks of surplus power, about 5kWh/day by my reckoning. That's a fair bit of hot water. Now the only solution that comes close to utelising all available power to my mind is the FX80 with its AUX SSR mode. But it still seems patched on to me, an afterthought. Data export is also important to me and FX80 requires more gear to achieve this out of the box.

    LVD is usually there to protect wiring--Most LVD's are set at 21 volts and don't really do a good job at "protecting" the battery bank. If you have a programmable LVD (usually a separate controller with LVD mode) which can be set to ~23.0 volts--That would be a better set point. In general, battery voltage varies so much with temperature, set of charge, and current (charging/discharging), it is difficult to protect the battery banks.

    If you are really into this--You could get a Battery Monitor (Xantrex/Victron are two I am aware of) with an alarm output... You could program it to "turn on" at 50% SOC and turn off at 80% SOC -- And connect to a low power high current magnetic latching relay.
    My main question is how hard can i push the TS60? The datasheet says 1600W of panel at 24V nominal. Morningstars string calculator says 1800W is OK. My support email to morningstar advised limiting to 5 max 300W panels. This might be butt covering on their part, but my inexperience with PV in this particular geography leads me to some caution, and i certainly know from experience how these microprocessor powered devices can fail, either short or open. The latter isnt going to do the AGMs any good at all.

    Are these MPPT or PWM type controllers (both are "TS" family units to add to the confusion)(never mind, PWM from next sentence). By the way, TS Morning Star charge controllers have remote battery voltage temperature sensors (and optional remote battery temperature sensors)... Both are a good idea to implement. Lead acid batteries are temperature sensitive during charging (cold batteries need higher charging voltage, hot batteries need less charging voltage). And AGM are more voltage sensitive/less forgiving vs flooded cell batteries.
    Anyway Isc STC for the array is about 52Amps. Without mppt and a fixed array angle, absence of snow etc, leads me to believe that the TS60 should work. Ive heard they can run at 60A happily ad infinitum. However here in NZ our PV does tend to be positive rated, (ozone hole or something).

    Heat is the main enemy (and thermal cycling)... If the area where the controllers are mounted will be "hot", you probably would want to derate the current a bit.... And/or add more cooling air flow to the area.
    Im aware that a mppt controller would allow me to run the panels at 2 strings of three reducing the 20m cable size. But at the cost of another $900 for a mppt controller. When i replace the tristar i can retire it as a LVD or 'large' SSR for the hot water.

    You might want to look at the Midnite Solar Classic charge controller (from the same guys that started Outback). Lots of development work still going on. And they have an active forum too.

    I believe the MorningStar TS can be configured as a LVD controller, if needed, for protecting your DC loads.
    Second question, reading the heated arguments about battery monitors, amp counters and their problems, im not really sure whether its even worth spending the 300 dollars on the recommended votronic. And its single shunt does little to reliably inform about in and out seperately. Instead id rather spend that money on a datalogger and set up a tablet or something slightly homegrown. Meanwhile a simple LED panel voltmeter, checked first thing in the morning.

    There is quite a bit of math/modeling behind Battery Meters (and their relatives). Reading voltage is not very accurate and you need temperature correction. With sealed/AGM batteries, you cannot measure specific gravity--So a battery monitor is the only good way of knowing what is going on from the battery's point of view.

    Smartguage (a voltage monitoring verson of battery monitoring) has a very interesting website if you want to read more about the issues. Trimetic is another popular BM system here.

    -Bill "I can run-on a lot too" B.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • petertearai
    petertearai Solar Expert Posts: 471 ✭✭✭✭
    Re: PV system check

    Hi The Victon batttery moniter is a great little unit. Not much over $300.00 nz. Esential in my opinion. All the beat with solar . The people here have a great deal of Knowledge.
    2225 wattts pv . Outback 2kw  fxr pure sine inverter . fm80 charge controller . Mate 3. victron battery monitor . 24 volts  in 2 volt Shoto lead carbon extreme batterys. off grid  holiday home 
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check
    BB. wrote: »
    Regarding your battery bank, review this website for how to parallel batteries (if you will be) in a bank for better current sharing. Also, your battery bank 24 volt and what total bank AH capacity (12x2??). A 400 AH x 12 volt batteries with 2 in series?

    Sorry, my diagram is not that flash. The batterys are 2V cells, in series. The PV, for the TS60 would be 6 in parallel, so string fuses, or else for mppt, two strings of 3, no fuses?.
    BB. wrote: »
    A 4kW inverter is pretty big... Huge for a 12 volt battery bank. And while an AGM battery bank can supply scary amounts of current (they are used for UPS systems that can drain a battery bank in 15-30 minutes). A fully loaded 4 kWH inverter could drain that bank inside of 1 hour...

    Im aware that oversized invertors are mentioned here a lot. My uneducated reasoning at the time was that if we cant afford a proper inverter at present then starting out with the better end of the chinese stuff. Hence i overrated it, so as to not push the thing so hard. All we need is a washing machine, an iron from time to time, you know those girl sort of 'necessary' things. I have a bunch of (lighter) woodworking gear which i'll content to run one at a time. Now the points you mention lead me to rethink this. They come in 2kW,3kW,4kW,5kW. Will 2kW run the gear listed above ok? Some of those motors have high start up draws. The invertor standby load will be less with the smaller model. Although it will be on a remote switch and used maybe only an hour every other day.
    BB. wrote: »
    And the fusing/branch circuit for a 4kW inverter on 24 volt bus should be:
    4,000 Watts * 1/21 volt batt cut off * 1/0.85 inverter eff * 1.25 NEC derating = 280 Amp branch circuit/Fuse/Breaker rating.
    100 amp shunt is way too small for a 4kW inverter with 200 amp fusing (unless you are really planning on using less power than 4kW)... One poster here (John P) has good experiences using shunts as fuses too.

    So 250 or 300A invertor fuse? The 100A shunt is a typo, it and the DC one are swapped. Yes, 50mV for the shunt, i wonder if i could use the entire length of the invertor cable as the shunt there. The battery computers use a shunt in the -Bat link. So the same probs there. Ill look into the shunts some more, thanks. I guess the alternative is some hall effect sensors. I also havent figured out yet how to rms correct for the pulsed waveforms that some of these power devices create.
    BB. wrote: »
    Note that shunts add voltage drop too... Making battery charging voltage a bit less accurate (along with wiring drop).

    I was thinking that the sense wires for the TS60 would be good for that.
    BB. wrote: »
    LVD is usually there to protect wiring--Most LVD's are set at 21 volts and don't really do a good job at "protecting" the battery bank. If you have a programmable LVD (usually a separate controller with LVD mode) which can be set to ~23.0 volts--That would be a better set point. In general, battery voltage varies so much with temperature, set of charge, and current (charging/discharging), it is difficult to protect the battery banks.

    The thing would need to be smart enough to recognise the short term volt drop from high loads, ie kicks in after some interval at the threshold value. I think /something/ is needed, for the worse case scenario, bat is low and we are out and some persistant DC load drains the battery. But in general vigilance and alarms will work better than LVD.
    BB. wrote: »
    If you are really into this--You could get a Battery Monitor (Xantrex/Victron are two I am aware of) with an alarm output... You could program it to "turn on" at 50% SOC and turn off at 80% SOC -- And connect to a low power high current magnetic latching relay.

    That seeds all sorts of fertile ideas, thanks.
    BB. wrote: »
    Are these MPPT or PWM type controllers (both are "TS" family units to add to the confusion)(never mind, PWM from next sentence). By the way, TS Morning Star charge controllers have remote battery voltage temperature sensors (and optional remote battery temperature sensors)... Both are a good idea to implement. Lead acid batteries are temperature sensitive during charging (cold batteries need higher charging voltage, hot batteries need less charging voltage). And AGM are more voltage sensitive/less forgiving vs flooded cell batteries.

    Agreed, both sensors are obviously well worth the little money they cost.
    BB. wrote: »
    Heat is the main enemy (and thermal cycling)... If the area where the controllers are mounted will be "hot", you probably would want to derate the current a bit.... And/or add more cooling air flow to the area.

    With the TS60 the PV will need to all be in parallel, ie 6 strings of one, ie 28V. I don't really want to stump a $1000 for the TS-mmptp-60 not knowing how hot it will run. If i did go mppt it would be FM80, or 2 mmpt 45A models. I guess i can always connect 5 of the panels to the TS60, at the hottest time of the year which is soon, keep an eye on the real world current and review. My plan to summer bias the array by 6 degrees will go on hold until the water diversion gets installed and the controller settled.
    BB. wrote: »
    You might want to look at the Midnite Solar Classic charge controller (from the same guys that started Outback). Lots of development work still going on. And they have an active forum too.

    Ive heard good things about those, but they dont seem to be available here? Even though most of the development seems to be grid tie focused, Im picking that we will see some pretty dramatic changes in all this gear in the next 2 years, and stumping $1500 (FM80+mate+sensors) now, seems just wrong to me. And Mppt seems to have a lot more to go wrong, if it is to be worked hard. As PV comes down and down, arrays get bigger, higher amp controllers will be needed. Hot water diversion will stop being seen as an optional extra.
    BB. wrote: »
    I believe the MorningStar TS can be configured as a LVD controller, if needed, for protecting your DC loads.

    Id spotted that!
    BB. wrote: »
    There is quite a bit of math/modeling behind Battery Meters (and their relatives). Reading voltage is not very accurate and you need temperature correction. With sealed/AGM batteries, you cannot measure specific gravity--So a battery monitor is the only good way of knowing what is going on from the battery's point of view.

    Smartguage (a voltage monitoring verson of battery monitoring) has a very interesting website if you want to read more about the issues. Trimetic is another popular BM system here.

    Yes. Amp counting will work well enough for the first few years, but the danger as i read it, is that as the battery capacity degrades the monitor reads more and more optimistically, allowing deeper discharges leading to faster degradation. Nasty spiral.

    Thanks Bill.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: PV system check
    zoneblue wrote: »
    Sorry, my diagram is not that flash. The batterys are 2V cells, in series. The PV, for the TS60 would be 6 in parallel, so string fuses, or else for mppt, two strings of 3, no fuses?.

    Correct. You look at the "series protection fuse" rating for the panel and the Isc (short circuit output current) too... Typically, Ifuse is less than 2x Isc--So you need a fuse to protect a shorted panel from being fed by two or more other panels in parallel.
    I'm aware that over-sized inverters are mentioned here a lot. My uneducated reasoning at the time was that if we cant afford a proper inverter at present then starting out with the better end of the Chinese stuff. Hence i overrated it, so as to not push the thing so hard. All we need is a washing machine, an iron from time to time, you know those girl sort of 'necessary' things. I have a bunch of (lighter) woodworking gear which i'll content to run one at a time. Now the points you mention lead me to rethink this. They come in 2kW,3kW,4kW,5kW. Will 2kW run the gear listed above OK? Some of those motors have high start up draws. The inverter standby load will be less with the smaller model. Although it will be on a remote switch and used maybe only an hour every other day.

    It is hard to know... Some inverters have very good surge (2x continuous rated watts/VA)... Others will not even output rated power. "Cheap" MSW inverters are probably cheap enough to take a risk and see what happens.

    Expensive TSW inverters--You would be better off getting ideas from people that use that same brand/model and see if would meet your needs (here or elsewhere).

    In some cases, you get a small TSW inverter to run your "critical loads" (electronics, fridge, lighting--that can be damaged by MSW inverters). And get a second big old MSW inverter to run your shop tools (and larger than that, a 5+kW generator to fire up when needed).

    In any case, short/heavy cables with a capable battery bank is a requirement to ensure there is enough current to run the inverter when supply surge current/heavy AC loads.
    So 250 or 300A inverter fuse? The 100A shunt is a typo, it and the DC one are swapped.

    Yep--that is what you are looking at if you want 4kW max power (and, in theory 8kW surge support).
    Yes, 50mV for the shunt, i wonder if i could use the entire length of the inverter cable as the shunt there. The battery computers use a shunt in the -Bat link. So the same probs there. Ill look into the shunts some more, thanks.

    Copper wire is not "temperature stable"--If you are OK with roughly +/- 50% accuracy--then you can do that...
    I guess the alternative is some hall effect sensors. I also haven't figured out yet how to RMS correct for the pulsed waveforms that some of these power devices create.

    Good enough vs exact is a real problem with measuring power... Inverters have a 120 Hz sine wave squared current wave form... How does any specific monitor read DC and 120 Hz sine squared current--Have not a clue... In the end, if you are within 10%--that is probably good enough for monitoring a system.
    The thing would need to be smart enough to recognize the short term volt drop from high loads, ie kicks in after some interval at the threshold value. I think /something/ is needed, for the worse case scenario, bat is low and we are out and some persistent DC load drains the battery. But in general vigilance and alarms will work better than LVD.

    One way is to set your voltage point to 11.5 volts / 23.0 volts (12 volt / 24 volt numbers) minimum under load set point voltage (or higher). The 10.5/21.0 volt inverter input, from my point of view supports a 11.5/23.0 volt battery output and 1/2 volt drop for wiring/fuses/shunts.

    If the battery is not under heavy load (i.e., somebody left a few lights and fan on)--Now you don't have the 1/2 volt wiring drop and the inverter shuts off at "dead battery" range of 10.5/21 volts.
    With the TS60 the PV will need to all be in parallel, ie 6 strings of one, ie 28V. I don't really want to stump a $1000 for the TS-mmptp-60 not knowing how hot it will run. If i did go mppt it would be FX80, or 2 mmpt 45A models. I guess i can always connect 5 of the panels to the TS60, at the hottest time of the year which is soon, keep an eye on the real world current and review. My plan to summer bias the array by 6 degrees will go on hold until the water diversion gets installed and the controller settled.

    MPPT charge controller is great for large arrays with longer wire runs (use much less copper in your wiring). And for matching non-standard Vmp-panels to the voltage required to properly recharge your battery bank (a 24 volt batter bank needs around 35<Vmp-array<40 volt for efficient use with a PWM controller... And both MPPT and PWM need Vmp-array>35 volts for full power operation in hot weather.
    Ive heard good things about those, but they dont seem to be available here? Even though most of the development seems to be grid tie focused, Im picking that we will see some pretty dramatic changes in all this gear in the next 2 years, and stumping $1500 (FX80+mate+sensors) now, seems just wrong to me. And Mppt seems to have a lot more to go wrong, if it is to be worked hard. As PV comes down and down, arrays get bigger, higher amp controllers will be needed. Hot water diversion will stop being seen as an optional extra.

    It is very difficult to "grow" a solar PV system over time--As well as migrate between pure GT system (panels+GT Inverter) vs full off grid power (+battery bank+charge controller+off grid inverter+....).
    Yes. Amp counting will work well enough for the first few years, but the danger as i read it, is that as the battery capacity degrades the monitor reads more and more optimistically, allowing deeper discharges leading to faster degradation. Nasty spiral.

    Some folks use 50% of battery capacity as their "bank AH set-point". Then just run 100% to 0% SOC without having to explain to guests/spouse/etc. that things are "going south" if the battery is below 50% SOC.

    Batteries are not perfect... The vendors usually define an end of life battery at 80% of capacity... And if you only use 100% to 50% SOC--You certainly can simply use the battery as it gets below 80% capacity...

    The numbers--I try not to get too warped up in them... Basically, if you can measure within 10% of the "real" or "predicted" numbers--You are doing well... And batteries that are within + 20% of rated capacity--they are OK too.

    There are trade-offs to be made here... Cheap batteries will give you the best bang for your buck--And if you murder your first set or two (not uncommon)--much less out of pocket risk. But you will get around 5-8 years from them (and may get as little as 3 years or less from cheap/automotive/marine batteries).

    Better batteries, you can 10+ years... And fork lift batteries will get 15-20+ years. However you will need a larger array for forklift batteries (more self discharge, especially as they age. And they tend to use more distilled water over time (cost/hassles of distilled water). There are folks that have gotten used fork lift batteries (may have 80% capacity guarantee) and been very happy (a few have even got "scrapped" Forklift batteries and changed/bypassed cells with good results).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check
    BB. wrote: »
    Expensive TSW inverters--You would be better off getting ideas from people that use that same brand/model and see if would meet your needs (here or elsewhere).

    These are sine wave invertors, Kosun is the brand.
    BB. wrote: »
    In some cases, you get a small TSW inverter to run your "critical loads" (electronics, fridge, lighting--that can be damaged by MSW inverters). And get a second big old MSW inverter to run your shop tools (and larger than that, a 5+kW generator to fire up when needed).

    This crossed my mind too. Use the 4kW in the workshop. The problem is her iron, it draws 1800W!
    BB. wrote: »
    Copper wire is not "temperature stable"--If you are OK with roughly +/- 50% accuracy--then you can do that...

    Duh, true. Silly me. So bigger shunts = less drop. 500A maybe if im happy with less resolution.
    BB. wrote: »
    Good enough vs exact is a real problem with measuring power... Inverters have a 120 Hz sine wave squared current wave form... How does any specific monitor read DC and 120 Hz sine squared current--Have not a clue... In the end, if you are within 10%--that is probably good enough for monitoring a system.

    Then youve got the variable square wave from the controller in absorb and float. I wonder how the battery monitors do it?
    BB. wrote: »
    MPPT charge controller is great for large arrays with longer wire runs (use much less copper in your wiring). And for matching non-standard Vmp-panels to the voltage required to properly recharge your battery bank (a 24 volt batter bank needs around 35<Vmp-array<40 volt for efficient use with a PWM controller... And both MPPT and PWM need Vmp-array>35 volts for full power operation in hot weather.

    Id be interested to try to measure the difference with mppt in our climate. Its kind of rainy and 400m asl, but mild temps. Obviously for the right controller its a no brainer, but as i said im not there yet.
    BB. wrote: »
    It is very difficult to "grow" a solar PV system over time--As well as migrate between pure GT system (panels+GT Inverter) vs full off grid power (+battery bank+charge controller+off grid inverter+....).

    Thats easy, ill retire the system to the granny flat we want to build. Or worse case break it into bits for the sleepout and/or motorhome. Then start all over with cheaper, better, bigger parts. If the Kurzwiels of the world are to be believed and the planet makes it to 2017 then the whole world will be 100% solar.

    Meantime i just need to get the tristar 60 through 18 months. What's my odds?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check

    Ok i can understand any ambivalence about judging whether a ts60 will blow up for my use case. Fair enough!

    On the hot water diversion thing, the best info i can find is this thread on the OB forum. 4 years ago.

    Ive been giving some thought to a couple of things: keeping a reign on the invertor, and data logging.

    How to protect ourselves from ourselves. Im the kind of person that burns pots on the cooker. How can i minimize the damage the invertor can do to our batterys, from something inadvertantly left on? My original idea was to divide the cabin into 2 or 3 small ac local circuits each with a 10 or 15 amp breaker. 10 amp breakers have a habit of popping on motor starts, so this isnt ideal. Many folk use an ac watt hour meter displayed prominantly, but my view on that is out of sight out of mind. Another idea is a kind of one shot timer, press button at powerpoint, point stays live for 30 mins. Another idea is a kind of tapered alarm system, that gets noisier and noisier relative to the current being drawn and length of time its drawn. The things that come out of ones mind while one is sleeping.

    On the other matter, my problem with your average battery monitor is its only measuring a single current value. It undertands the bettary by understanding the differential of inputs to outputs. But knows nothing about either. It seems to me that if i am going to spend several hundred dollars on monitoring, *and* lose 50mV of peak energy i may as well at least capture all the relevent data while I am messing with it. Heres some options:

    Build something from scratch using a an old android smart phone or cheap tablet, and a usb interface. Maybe $400 plus shunts. eg
    MC miniLAB-1008 or Labjack u12 .

    Run my old 8 channel serial DAQ with an old low wattage VIA motherboard or hacked wifi routerand run it as a server accessible on the network.
    Cost $0 + shunts.

    Then i just found these, the pentametrics. Combine PM-5000-U with PM-101-CE US$320 say landed here $450 + shunts. Has three 3 current inputs, 2 voltage inputs. Looks quite promising. Anyone had experience with them?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • ggunn
    ggunn Solar Expert Posts: 1,973 ✭✭✭
    Re: PV system check
    BB. wrote: »
    Also, you do not want to connect solar array frame ground to DC return line at the array--Solar +/- should be fully insulated wire from the array to the charge controller--No earth/safety grounding.

    Repeat for emphasis. Grounding the negative DC conductor will disable ground fault detection.
  • petertearai
    petertearai Solar Expert Posts: 471 ✭✭✭✭
    Re: PV system check

    Independent power in Auckland stock midnite cabinets . They may also stock the new controllers.
    2225 wattts pv . Outback 2kw  fxr pure sine inverter . fm80 charge controller . Mate 3. victron battery monitor . 24 volts  in 2 volt Shoto lead carbon extreme batterys. off grid  holiday home 
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check
    Independent power in Auckland stock midnite cabinets . They may also stock the new controllers.

    They do! And they are cheaper than the FM80.

    Thanks!!!

    To keep you up to speed we have decided to go with Classic 150 lite. I had to actually do all the numbers to see how poorly the TS60 would work in this situation. For a start our PV has quite a high Vmp at 36.2v. The TS60 would have run way down the PV's power curve. Secondly when you go to 2 strings of three panels, when compared with 6 strings of one its not 3 times more cable its 9 times more cable. Ouch.

    The midnite seems to run more efficiently with 3 strings of two panels, than what i was planning on using, 2 strings of 3. I dont know what will be in it, but the string fuses clinch the deal. However if i decide to add two more panels the only configuration that the classic will handle at 24v, is 4 strings of 2. That will actually push 92 amps through the thing . Amazing.

    As for the opportunity hot water, ill figure that out later!

    Ill post my final draft schematic tomorrow.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • petertearai
    petertearai Solar Expert Posts: 471 ✭✭✭✭
    Re: PV system check

    The warehouse has a 1200 watt iron.
    2225 wattts pv . Outback 2kw  fxr pure sine inverter . fm80 charge controller . Mate 3. victron battery monitor . 24 volts  in 2 volt Shoto lead carbon extreme batterys. off grid  holiday home 
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: PV system check
    zoneblue wrote: »
    How to protect ourselves from ourselves. I'm the kind of person that burns pots on the cooker. How can i minimize the damage the invertor can do to our batterys, from something inadvertantly left on?

    Does your inverter have an adjustable low-voltage-disconect? If not, you can buy one or build one from a voltage controlled relay. Perhaps you can use the aux on your Classic to trigger a disconnect of your inverter at a particular voltage.
    zoneblue wrote: »
    Another idea is a kind of one shot timer, press button at powerpoint, point stays live for 30 mins.

    Excellent idea. I use spring-wound timers, such as these: http://www.outdoorlightingtimer.com/InWall-Timers.html

    please note: I just found that link on a google search. I can't recommend them because I have never dealt with them (but since I do need another timer, I might give them a try).

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check

    Thanks guys, you are full of bright ideas.

    Heres the latest drawing, hopefully down to fine details.

    Attachment not found.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: PV system check
    zoneblue wrote: »
    Thanks guys, you are full of bright ideas.

    Heres the latest drawing, hopefully down to fine details.

    I do not see any indication in the drawing of whether the system DC negative is grounded. If it is not, then you should have a ground detector on the panel side and probably should have a fuse in the negative lead from the panels as well as the positive lead.
    If it will be grounded, then you need to show that explicitly in the drawing rather than having it hidden inside a fuse box or other component.
    SMA SB 3000, old BP panels.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check
    inetdog wrote: »
    I do not see any indication in the drawing of whether the system DC negative is grounded. If it is not, then you should have a ground detector on the panel side and probably should have a fuse in the negative lead from the panels as well as the positive lead.
    If it will be grounded, then you need to show that explicitly in the drawing rather than having it hidden inside a fuse box or other component.

    Should the neg bus still be grounded with the ground fault unit present in the classic. Or should i just disable it?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: PV system check
    zoneblue wrote: »
    Should the neg bus still be grounded with the ground fault unit present in the classic. Or should i just disable it?

    Your best bet, I think, would be to indicate in a note in the diagram that the Classic contains a ground detector, which references the negative panel wire to ground but will disconnect the panel from the CC on detecting fault current to ground. Different people have different ideas about the wisdom of their setup, but it seems to be accepted by AHJs and has not caused a problem with their UL listing.
    SMA SB 3000, old BP panels.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check- UPDATE

    Hi guys,

    I have finished our ground mount, borrowing a few ideas from some of you, thanks.

    Attachment not found.

    The panels will go up later today, and the wiring tomorrow.

    Theres a few changes of note. Ive decided to go 3 strings of 2 panels for 2 reasons. a) to stay under the 120V ELV regs. b) going from 90v to 120V, for the same PV to controller cable (6 gauge) you gain 1% of cable losses, but lose 2% in controller losses. So nothing really gained to my eye.

    A few related questions have arisin after talking to our local solar expert:

    1. he has made a reasonably good case for going unearthed on the DC side. Here if the PV voltage is over 120V, then the DC side must be earthed, but under that its optional. He said that from his experience there is more lightning damage in earthed systems.

    To change my circuit to unearthed, i basically:?
    - remove all the earths on the dc side
    - put pairs of breakers for neg and pos side of each section, ie controller in, controller out, and DC loads

    2. Breaker woes. When i picked up the classic lite 150 the midnite dealer gave me a DIN DC only breaker for the controller input, and a panel mount AC/DC breaker for the controller out. Trying to mount this eclectic lot, led me to try to go all DIN rail, but theres some issues there too. The DIN ones only go as high as 60amps. My system will run at around 60A, and that feels too tight to me? If the 60 is ok, then i will get a twin for the neg side,. In that case im confused about which way to mount them. TO my mind when the battery is charging, the controller side is more positive, but in a fault situation the battery side will be more positive. Midnite say to mount the pos side to the battery. But it also says to not reverse polarity!?

    Alternatively if i go all panel mount. But then the midnite guy said that you cant use the AC panel mount breakers for the PV side because "the controller pulses them" and they are called pv breakers for a reason.

    3. Lastly the local expert said that regs here require string fuses regardless of the no. of strings. In my case i was following the standard logic around here that you only need string fuses when the the number of strings less one current exceeds the string fuse size, usually 3+ strings. In my case Isc is 8.8amps, but the sticker string fuse is 20 amps. That means i can do 3 strings without string fuses, and ive got combiner bits appropriately. His preference is to run PV cable to each string and mount the string breakers inside. That also doesnt match the cable i have. If i go that route ill need more cable (which in itself is ok as it was cheap) but im back to more DC breakers. Is there really no way around using both DIN and panel mount? I suppose if i do all the above then ill have 6 string breakers, 2 controller in, all DIN, and the pair of panel mounts for the controller out plus a pair for the DC load board. Thats more doable i guess, as i have a section of din rail that will hold the 6, and the other four can go in the side of the case.

    Reality bites now. Any thoughts appreciated. Cheers.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: PV system check- UPDATE

    My suggestion--You may need some "X" (cross) bracing under the arrays and at the front (possibly more at the rea too)....

    Before you install the solar panels--Push on the corners (left to right from the camera view) and see if the system is rigid or flexes--You don't want any flexing from forces applied from any angle.
    zoneblue wrote: »
    Hi guys,

    I have finished our ground mount, borrowing a few ideas from some of you, thanks.

    Attachment not found.

    1. he has made a reasonably good case for going unearthed on the DC side. Here if the PV voltage is over 120V, then the DC side must be earthed, but under that its optional. He said that from his experience there is more lightning damage in earthed systems.

    Yes--There is the "issue" of conversion from common mode voltage (black and red go up and down together) vs grounding the black (black stays at ground, red goes up and down with strike--hitting the DC devices with >>12-48 volts of differential energy).

    However, a few years ago, I asked our host Windsun where he saw the most lighting induced failures--And he said it was on the AC output of the inverters....

    My theory is that the DC Battery Bank is both chemically and capacitively holding the black/rad wires at 12-49 VDC--And the (relatively) few amps from a lighting strike is not enough to overwhelm the battery's ability to keep the differential voltages at normal levels. (again, just my theory).

    So, I am agnostic as to "floating" DC ground vs tying to a ground rod in terms of lighting. Plus, the lightning voltages are so high--the 1/8 ince to 1/4 inch or so gaps that are usually used in DC and AC power systems are "nothing" to lightning---So why not earth ground the negative the battery bus and get rid of the static build up issues (which is a safety issue itself).
    To change my circuit to unearthed, i basically:?
    - remove all the earths on the dc side
    - put pairs of breakers for neg and pos side of each section, ie controller in, controller out, and DC loads

    I would ground the DC Bus at the battery bank to earth ground rod with 6 AWG wire. And I would use Midnite's new lightning arrestors at the building entrance for the solar/wind power lines (idea is to prevent lightning energy from entering the home/battery shed).

    It may be overkill--But if you have lightning strikes in your area--Why not try it.

    Breakers will not really do anything at the voltages lightning operates at--I would not count on them doing anything useful.

    In theory, if you have a floating DC power system (no DC to earth ground), then you need ganged breakers (two pole breakers) just like we use on 120/240 VAC circuits. You need to assume a fault to either leg to ground, and want to kill both power leads.

    There are failures that can cause the "return" wires to carry too much current and you need fuse/breaker the return leads too--If you do not earth reference the DC negative bus.

    I have to go now--I will look at the rest of the post later.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check- UPDATE
    BB. wrote: »
    My suggestion--You may need some "X" (cross) bracing under the arrays and at the front (possibly more at the rea too)....

    Before you install the solar panels--Push on the corners (left to right from the camera view) and see if the system is rigid or flexes--You don't want any flexing from forces applied from any angle.


    The front legs are short, less than 2 feet, and feel pretty solid. But yes youre right, i agree some strapping under the panels or under the 2x4 rails would be good. That will produce 5 solid planes, leaving the back as minimally braced as it is.
    However, a few years ago, I asked our host Windsun where he saw the most lighting induced failures--And he said it was on the AC output of the inverters....

    His words were similar, a ground strike nearby comes in via the earth cable, and kills the inverter via its neg lead.
    "in every case where ive seen a dead inverter it was in an earthed system".
    So why not earth ground the negative the battery bus and get rid of the static build up issues (which is a safety issue itself).

    Can you say more...
    It may be overkill--But if you have lightning strikes in your area--Why not try it.

    We are not a high lightning area compared to many parts of the world, but are elevated, and locals regularly report that their telephones die periodically.
    In theory, if you have a floating DC power system (no DC to earth ground), then you need ganged breakers (two pole breakers) just like we use on 120/240 VAC circuits.

    Thanks, you reminded me i forgot to ask whether the midnite breakers can be ganged. They have a pin hole that looks promising, but i know some breakers dont have enough trip force to trip multiple poles.


    Cheers.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: PV system check- UPDATE
    zoneblue wrote: »
    So why not earth ground the negative the battery bus and get rid of the static build up issues (which is a safety issue itself).
    Can you say more...

    The question is can I say less :p (and we all know that answer).

    My belief is that with the DC Battery Bank and the heavy copper cables (and relatively high ratings of fuses/breakers and the short hit from lightning), that the conversion of lightning from common mode to differential mode is not a common failure (at least on the DC side).

    It appears that the AC output stage of the inverters is the "weak link" (with a data point of 1 from our host). Which is a little odd in the fact that most modern AC devices are tested with ~1,800 VAC (highpot) from the factory to pass UL/NRTL certs.

    I don't know if it is the ground converting the common mode lightning energy to differential--getting into the Inverter, or AC wiring typically runs "everywhere" while DC wiring is relatively contained to around the battery bank (remember that lighting is RF energy and does not like to follow long runs of cables very well)... Who knows, even the solar panels are big diodes/flat plate capacitors and may keep the differential conversion volt low too (by shorting any higher voltages).

    This is getting into guess work on my side--I don't have enough experience or information to go really deep into this stuff.

    But--In the end, I don't like floating circuits because A) you can get static build up (100-300 volts per meter is the earth's E-Field, and under thunder storms it can quickly climb to thousands of volts)--So grounding gets rid of all the static charge issues. Earthing of metal towers/antenna/solar panel frames/etc. is standard practice.

    Plus B), the whole issue that a "truly" floating DC power system should have Ganged breakers on each +/- lead that leaves the battery bank (real quick, worst case--Battery + gets grounded to metal conduit/box/battery frame/etc..--No fuses blow. Now, DC power is no longer floating but + referenced at Battery voltage/current availability. Now get a short from a 12 volt negative wire to one of your DC LED fixtures, and you have Battery Voltage/Current of 100's of amps going through that little wire with no fuse/breaker in path to limit to 5 amps). Note--if you have DC signaling voltages (RS 232, 485, etc. and signal grounds, the floating/positive ground can easily fry the signal wiring/comm. cards or ports in a computer).

    The "modern" NEC codes (being rewritten now?) does include a "DC GFI" detection fuse/breaker--I have many reasons why I think it is unsafe to ever use this setup--and has only limited improvement with Arc Fault safety (why it was implemented)--Anyway a whole 'nother thread.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: PV system check- UPDATE

    Assuming those are 4x4's... They do not have a lot of strength when subjected to bending forces (something like a 10' pressure treated 4x4 is good for sustaining a 125 lb load--or something like that).

    I would highly suggest diagonals or even 3/4" exterior (painted and sealed) plywood with lots of nails (I use screws--but for earthquake codes they actually require nailing) to make diaphragm (if you have parallel legs) or gussets if you do not (a "triangle" to strenthen the "T" section down near the ground--to reduce the "unsupported" length).

    I have a friend, back when we were "kids", put a 4x6 on his van for a a bumper--A simple parking lot tap was enough to snap it like a twig...
    zoneblue wrote: »
    2. Breaker woes. When i picked up the classic lite 150 the midnite dealer gave me a DIN DC only breaker for the controller input, and a panel mount AC/DC breaker for the controller out. Trying to mount this eclectic lot, led me to try to go all DIN rail, but theres some issues there too. The DIN ones only go as high as 60amps. My system will run at around 60A, and that feels too tight to me? If the 60 is ok, then i will get a twin for the neg side,. In that case im confused about which way to mount them. TO my mind when the battery is charging, the controller side is more positive, but in a fault situation the battery side will be more positive. Midnite say to mount the pos side to the battery. But it also says to not reverse polarity!?

    You may need to talk with Midnite (or try their forum)... Some breaker are marked Line and Load and others are marked +/- ... It depends on what is being protected... Sometimes it is the direction of the arc control (heat rises) and sometimes it is the actual direction of current flow (they use small rare earth magnets to "blow out" the Arc when the contacts open--wrong polarity and the Arc can "blow" the wrong direction.

    For breakers marked "line/load"--I put the Line towards the source of the most current in a single fault condition... With a solar array to controller wiring, that would be (my guess) the solar array (solar PV panels do not supply more than Isc--so breakers on the main solar wiring don't make sense--For combiner boxes, the one protected string is the "load" and the source of current is the rest of the array feeding a shorted string).

    For Controller to Battery--The Battery is "Line" where the current is, and the controller is the "Load" even though it is the source of the current in normal operation.
    Alternatively if i go all panel mount. But then the midnite guy said that you cant use the AC panel mount breakers for the PV side because "the controller pulses them" and they are called pv breakers for a reason.

    Not sure I understand that as an issue--But many AC breakers have very low DC interrupt ratings (may work on 240-600 VAC, but have 48 VDC DC rating, and DC amp interrupt rating can be reduced too).... That is why usually you need "special" DC breakers that are designed for >>48 VDC and high current interrupt capabilities.
    3. Lastly the local expert said that regs here require string fuses regardless of the no. of strings. In my case i was following the standard logic around here that you only need string fuses when the the number of strings less one current exceeds the string fuse size, usually 3+ strings. In my case Isc is 8.8amps, but the sticker string fuse is 20 amps. That means i can do 3 strings without string fuses, and ive got combiner bits appropriately.

    2x 8.8 = 17.6 amps--So, yes, in this case you could have used 3 parallel connected strings without breakers/fuses. But the inspectors are "god" here--So what they say goes--And usually it is not a problem unless A) it costs a lot of extra money or B) it is unsafe (see NEC DC GFI rant of mine).
    His preference is to run PV cable to each string and mount the string breakers inside. That also doesnt match the cable i have. If i go that route ill need more cable (which in itself is ok as it was cheap) but im back to more DC breakers. Is there really no way around using both DIN and panel mount? I suppose if i do all the above then ill have 6 string breakers, 2 controller in, all DIN, and the pair of panel mounts for the controller out plus a pair for the DC load board. Thats more doable i guess, as i have a section of din rail that will hold the 6, and the other four can go in the side of the case.

    Are you saying he did not want a weather proof DC Combiner box (with single pole fuses/breakers out at the array)? But, instead, he wanted the combiner box mounted in the home/battery shed?

    I can see wanting the Combiner Box in shade (or even a weather resistant enclosure--maybe). But forcing small wire runs from array seems a bit petty... Perhaps it is their way of making it "painful" to add more solar strings later and possibly overloading the combined current wiring....

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Re: PV system check- UPDATE

    The framing looks pretty good to me! I hope your mounts are slotted, you may find it dificult to be 'perfect' with drilled mounts. I mounted the bottom of the rails and attached the single hole angle mounts to the panels and then to the rails and attached the top of the rail last. You will understand if you have issues.
    Looks very stout for a 6 panel setup, but then again your more exposed on the back side. Here's my cross bracing on the back and additional brace along the side (20 panel setup);
    Attachment not found.
    BB. wrote: »
    For breakers marked "line/load"--I put the Line towards the source of the most current in a single fault condition... With a solar array to controller wiring, that would be (my guess) the solar array (solar PV panels do not supply more than Isc--so breakers on the main solar wiring don't make sense--For combiner boxes, the one protected string is the "load" and the source of current is the rest of the array feeding a shorted string).

    For Controller to Battery--The Battery is "Line" where the current is, and the controller is the "Load" even though it is the source of the current in normal operation.

    The Midnite breakers are marked "+" and "-", and the rule of thumb (as I understand it) is to between the panels and the controller keep the "+" toward the panels and after the charge controller keep the "+" toward the batteries.

    Midnite now makes din rail breakers in 80 and 100Amps(they do use 2 slots), perhaps if you think your safe with 60 amp use them for now until you can get a din rail when they are available there.
    12.jpg 139.5K
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: PV system check- UPDATE

    If the breakers are marked + and - it could be for two reasons:
    1. The electrode construction is such that the interrupting capacity with no electrode erosion during arcing is only available when interrupting current flow from + to - , or
    2. The + and - are just poor ways for labeling the line and load sides, and reflect the way the breaker gets triggered.

    The current to be interrupted by opening a breaker on the panel side of the CC will be flowing from the panels as energy source toward the CC as load. But that current is limited and there should never be a fault condition that causes a trip on current in that direction, only manual opening as a disconnect.
    On the other hand, if the CC fails and there is a short in the panels, a much much larger fault current will flow from the batteries back into the damaged panels. This is when the overcurrent will trip the breaker, and it should be a very very rare occurrence. Having to replace the breaker in that case is not a big deal as long as it does its job once.

    So the right way to face the breaker will be determined by which of the two reasons applies to the breaker in question. For breakers labelled line and load, the line side should always face the batteries, regardless of which side of the CC the breakers are on.
    IHMO anyway....
    SMA SB 3000, old BP panels.
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Re: PV system check- UPDATE

    Well I understood what "Fizzy" said to be true, until I read this thread at Midnite. Which includes the statement "...In a PV combiner the + sign marked on the breaker connects to the PV positive output." As they are the manufacturers, unless I'm reading this wrong...

    ...or perhaps as there is now a breaker required(?) ahead of the charge controller in the powercenter/breaker box maybe they're redundent and the breaker in the powercenter should be "+" toward the charge controller and battery...

    Now I'm confused, also in the thread there is reference to the 2014 NEC code requiring bi-polar(I'm getting there) breakers, which don't exists for din rail breakers, yet...

    I truely am not overly concerned, and mostly consider these switches, I'd bet 12 years ago most small solar was setup without breakers or fuses between the panels and the charge controller. When I fisrt saw the Pulse Energy Power center with breakers (big honking things) between the panels and the charge controller I thought it was neat! at $1500 I couldn't afford the fancy box. This was state of the art in 1999, was that really all that long ago?
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: PV system check

    Depends on which side of the charge controller you're on. The (+) goes to the positive of the power source. On the PV side that's the solar panels. On the battery side it's the battery, even though we always think of power flowing from the controller to the battery. If something goes wrong with the controller, the power will flow from the battery to the controller so you want the breaker the other way 'round.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: PV system check- UPDATE
    Photowhit wrote: »
    Well I understood what "Fizzy" said to be true, until I read this thread at Midnite. Which includes the statement "...In a PV combiner the + sign marked on the breaker connects to the PV positive output." As they are the manufacturers, unless I'm reading this wrong...

    I find the thread at Midnite more confusing than enlightening, not the least because it is not clear which of the participants are talking about breakers on the panel side of the point where they are combined and which are talking about the single wire connection from the combiner to the CC.

    Since a breaker at any point in the circuit from the panel to the battery, and on either side of the combiner for that matter, could theoretically be subject to fault current in either direction, the comment that the breakers will only trip properly on current flowing in one direction has me very concerned!
    And when used as switches, the breakers will be carrying current in one direction at the time they are switched off, while for breakers on the panel side of the combining point the only likely direction of fault current is the exact opposite (multiple strings ganging up to force high current through a broken string.)

    It really sounds like the only way to meet all requirements would be use a polarity-insensitive (non-polarized) breaker in the first place.

    Finally, just to muddy the waters farther, if you have a positive ground panel system, the breakers will have to be in the negative line from the panels, and so the polarity of the breaker will have to be sort of reversed, with the - on the breaker toward the panel's - lead.
    SMA SB 3000, old BP panels.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: PV system check- UPDATE
    inetdog wrote: »
    Finally, just to muddy the waters farther, if you have a positive ground panel system, the breakers will have to be in the negative line from the panels, and so the polarity of the breaker will have to be sort of reversed, with the - on the breaker toward the panel's - lead.

    Which might lead to reversing it twice because being on the negative side the current is already flowing 'the other way' so turning the breaker 'round means it's now backward for that side.

    How's that for confusing? :p
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Re: PV system check- UPDATE

    I guess all the confusion has lead to the NEC Code change mentioned in the thread, all breakers will have to be non polarized after 2014...

    If I'd read carefully I would have just purchased 3 panel mount breakers rather than the din rail ones, but I'd also have to order another plate for my E-Panel, I wonder if they are stocking up on plates for panel mount breakers over there? perhaps discontinuing the din rail breakers...

    I'm actually putting in my din rail breakers in the morning, and getting my system a bit closer to 'code' at least for a year or 2 (maybe 10 since it takes a while for Missouri to adopt code) I'll be asking a question over in 'wiring' if and about these breakers, wire input and spacing...
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: PV system check- UPDATE

    By way of update, a few days ago we put the panels up and the battery bank and controller were wired up. A pair of 6 gauge cables about 16m long each in its own conduit, as its a cheaper automotive grade of cable. Ground stake whacked in and connected using 6mm, which the midnite manual said was ok for 60amps.

    Testing comprised one last check for correct wiring, quick test for shorts and battery and array polarity with a multimeter, pv breaker off, each string terminated to the combiner but leaving an mc4 in each string open. Closing each string one at a time, to double check polarity of each string. Classics ground fault set to on, dip switches set to AGM, 24V, static IP, manual EQ. After studying the battery literature i eventually concluded that the classic default voltages were ok.
    Turn on breaker between battery to classic . Boots briefly, then into sleep, not much to see on the lite. Turn on pv breaker, goes into bulk, and takes less than a half hour at 8am to get bank back to float, after 2 months left sitting. No fans.

    Loads are still light for now, (fridge, lights and computers), and despite the most atrocious weather the controller is always in float by 8 or 9am. So havent had a chance to test it at full load. Max ive seen is about 240W. Still keen to test the max pv output, cable and controller temps, string balance, cable losses, etc.

    Here's a couple more photos.

    Attachment not found. Attachment not found. Attachment not found.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: PV system check- UPDATE
    zoneblue wrote: »
    Still keen to test the max pv output, cable and controller temps, string balance, cable losses, etc.

    Congratulations! Don't forget to check that your temp compensation is working correctly. --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i