c60 for 24V parallel array

dellsworthdellsworth Registered Users Posts: 20
Does anyone see any reason I can't parallel 6-250 watt panels to an old xantrex c60 charge controller? If you don't know the controller will run in 12 or 24 volt mode. The V in specs on the c60 is 55 open current and, of course, the load amps are 60 with a peak of 80+. The panel amps are in the range of 8.1 to 8.6. Of course I'd rather not but I have one and it's a Guatemala remote indigenous installation that has no money. I'd like to unload what I have around rather than charge them more for a new mppt. Thanks, Dan

Comments

  • BB.BB. Super Moderators, Administrators Posts: 29,189 admin
    Re: c60 for 24V parallel array

    The big issue is these panels have Vmp~30 volts Vmp (I guess). The voltage is almost 2x higher than a PWM controller should use for charging a 12 volt battery bank (Vmp~17.5-18.6 volt or so ideally or Vmp~35-38 volts for a 24 volt battery bank).

    So, on a 12 volt bank, the panels would output approximately:

    250 watts * 17.5v/30v = 146 watt panels

    And on a 24 volt bank, the battery bank would not fully charge (Vmp-array on a hot day would be around 24 VDC).

    So, a 12 volt system would function, it really depends on the cost of the 250 panels--Are the "cheap enough" to be thought as/useful as ~146 watt panels?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • dellsworthdellsworth Registered Users Posts: 20
    Re: c60 for 24V parallel array

    Hi Bill,
    Thanks for your thoughtful response. It'd be for a 24V bank.

    It's easy enough for me to get 145's instead of 250's from the same company (ERDM solar). Those have a vmp of 17.86 and an imp of 8.15. If I got 12 for 6 parallel strings it should work fine. Even at the max isc of 8.67 it should be within the limits of a C60. Price per watt is the same. Cost goes up a bit because of wiring and mounting but not much.

    So to clarify the calculations and so I can learn something....

    On the 12 volt you're saying I'd be throwing away a fair amount because of the difference between what the c60 puts out to the bank.compared to what it is taking in. Or to put it another way... Even though it can take in 55V it really only processes 17.5-18.6? I saw it taking in up to 55V (31 according to ERDM specs) and didn't think about where the excess voltage was going. I can understand that.

    The 24V is a bit harder to get. The panel is rated at 31 imp. If I pair a couple of 145's (17.86 imp) I get it that they're going to be 35+ compared to 31. Does a hot day derate down to 24V? What's the derating on a pair of 145's?

    This brings up a very interesting question for me. I've a customer that's been having problems getting their new batteries to charge. The panels are some cheap 80's they got years ago and it's an undersized array for their consumption. They've had problems for years during the rainy season and were simply too cheap to upgrade the array and ran a generator instead. But it seems to have gotten worse and I even recently put in a smaller bank for them since the previous installer had a bank way too big for the array size. I just thought it was mostly rainy season problems with insufficient consecutive hours of sun. Would the imp have dropped on those panels through aging enough so that they might not have the force to charge up the batteries? It's also a 24V bank. This is a pretty relevant question for all of us. The boom in the industry has been relatively recent and off-grid systems panels are starting to age. Are we going to see more problems with charging banks that previously were sized okay?

    If so I suppose we need to start re-configuring the arrays and putting in mppt's to extend their life.

    Dan
  • BB.BB. Super Moderators, Administrators Posts: 29,189 admin
    Re: c60 for 24V parallel array
    dellsworth wrote: »
    On the 12 volt you're saying I'd be throwing away a fair amount because of the difference between what the c60 puts out to the bank.compared to what it is taking in. Or to put it another way... Even though it can take in 55V it really only processes 17.5-18.6? I saw it taking in up to 55V (31 according to ERDM specs) and didn't think about where the excess voltage was going. I can understand that.

    More or less... Remember that solar panels are (close enough for our usage) constant current devices... If you load the panel in full sun at ~18 volts, you will get ~8 amps out. If you load the array to 14.5 volts (battery charging with PWM controller), you still will get 8 amps out.
    • 18 volts * 8 amps = 144 watts out of panel into "ideal load"
    • 14.5 volts * 8 amps = 116 watts into "ideal battery"

    If you had a 30 volt panel into a 12 volt battery bank with PWM controller--Similar but worse:
    • 30 volts * 8 amps = 240 volt STC rated panel
    • 14.5 volts * 8 amps = 116 watts into battery

    There is no "other power" being lost in the system... More or less, each photon of light create a micro pulse of current. That current can either be used at 18 volts (for Vmp~18 volt panel) or it can be used at even 0 volts (shorted output and power = zero watts).

    We are so used to how batteries (and utility grids work), that current mode sources/sinks are just a whole different animal that really twists how we think of electricity (and from an electrical/electronic engineer point of view, the differences between current and voltage sources carry through to many other design issues--it is really mind blowing in some cases).

    So, why rate a 12/24 volt PWM controller to 55 Volts? Basically, what is being rated is the "switch" (FET -- power transistor). A very cold panel (sub freezing) will go over >50 volts when there is no load -- I.e., Voc-cold -- Voltage open circuit cold). The switch should not fail in worst case conditions (in this case, when the solar panels are wired to support a 24 volt battery bank on a very cold day).
    The 24V is a bit harder to get. The panel is rated at 31 imp. If I pair a couple of 145's (17.86 imp) I get it that they're going to be 35+ compared to 31. Does a hot day derate down to 24V? What's the derating on a pair of 145's?

    I assume you intended to type Vmp=31 volts, not Imp=31... The problem is on normal to hot days, you can loose a significant amount of output voltage (hot solar cells output lower voltage).
    So, for a 24 volt battery bank, you really need to charge at ~29-31 volts (depending on battery, temperature, etc.). And you need a volt or so voltage drop for wiring/charge controller. And account for the hot solar panel dropping Vmp.

    For example, if a solar panel drops 0.4% Vmp per degree C. Panel is rated at 25C, ambient is 40C, and 30C rise.
    • 38 volts Vmp * (0.004 vmp per C)(40C+30C-25C) = 6.84 volt drop
    • 38 Vmp - 6.84 volt drop = 31.16 volt Vmp-array-40C (hot array)

    Note that Vmp is not a "sharp" point, but a soft "knee"--So, you still have a +/- volt or so where the panels+batteries will work OK (i.e., a battery trying to equalize at 32 volts will not see "zero" current).
    This brings up a very interesting question for me. I've a customer that's been having problems getting their new batteries to charge. The panels are some cheap 80's they got years ago and it's an undersized array for their consumption. They've had problems for years during the rainy season and were simply too cheap to upgrade the array and ran a generator instead. But it seems to have gotten worse and I even recently put in a smaller bank for them since the previous installer had a bank way too big for the array size. I just thought it was mostly rainy season problems with insufficient consecutive hours of sun. Would the imp have dropped on those panels through aging enough so that they might not have the force to charge up the batteries? It's also a 24V bank. This is a pretty relevant question for all of us. The boom in the industry has been relatively recent and off-grid systems panels are starting to age. Are we going to see more problems with charging banks that previously were sized okay?

    There are lots of reasons for systems to perform poorly. In general, a good glass+crystalline solar panel (from what I have read here) should be pretty close to factory specs even 20 years down the road (or assume ~80% of factory specs if you go by panel warranty).

    And batteries become less efficient as they age. Many times batteries are replaced not because they had a "hard failure" but because they no longer store enough energy for the application (i.e,. house goes dark before sun rises the next day).

    Looking for obvious problems (delaminating panels, brown/black spots in panels), melted/charred wires/connectors, corrosion of wiring/terminals, hot connections or battery/battery cells, etc... can point to specific problems that need fixing rather than to a generic "panels must be going bad".

    I really push getting a DC Current Clamp DMM (Digital Multi Meter) (such as this very cost effective unit "good enough" meter from Sears).

    With most solar power systems, there are lots of parallel components. I.e., parallel solar panels. You clip the meter on a wire from each array string and measure the current... If you see more than a 10% difference in current flow between strings, the low current string may have a weak panel or bad wiring connection somewhere. Similar with battery banks--If you have parallel battery strings, measure them under load (and under charging) and make sure they are sharing current reasonably (batteries don't "share current" well by nature--So you are looking for larger differences and issues with wiring, cells, etc.).

    So you can compare and see if you have a failing component or not (measuring absolute current is less easy--Sun, clouds, temperature, battery state of charge, etc.). For my GT System, it was not "obvious" that I had a failing solar array until the output dropped by amost 50% day over day (sort of "mixed weather conditions" made it difficult to "see" 25% drop in output for my GT system).
    If so I suppose we need to start re-configuring the arrays and putting in mppt's to extend their life.

    If you have a "weak" panel (or battery), reconfiguring them is not going change much. And putting an expensive MPPT controller in there is not going to help unless the array was miss configured in the first place (say using Vmp~30 volt panels on a 24 volt battery bank).

    And, with older panels -- There can be "non-standard" Vmp ratings (some older panels are Vmp~15 volts and design to charge a system without using a charge controller--i.e., "self regulating" panels--not really good for batteries, but people did that to save money).

    And when all is said and done, is it worth spending a lot of time trying to mix old and new panels (a bunch of 80 watt panels vs strings of 240 watt panels with different Vmp/Imp ratings, etc.)--Not always... Some folks have done well by selling of the old (but still good) smaller panels and using the money to build out a new large panel array and new MPPT charge controller (MPPT controllers are almost always required for use with >140 watt solar panels these days--You have to check the Vmp ratings of the panels).

    There are some other nice reasons to use MPPT over PWM controllers (>800 watt arrays, longer wiring runs from array to charge controller/battery shed, more logging functions on new MPPT controllers, more networking options, more adjustments in software, etc.).

    In the end, I suggest people do several paper designs (old, new, mix) and see what makes most sense to them.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • dellsworthdellsworth Registered Users Posts: 20
    Re: c60 for 24V parallel array

    Hi Bill,

    I get all of the above and I really appreciate your responses. It's like getting a short course in off-grid designing. Did you happen to mean VOC instead of Vmp when referring to 38 volts? You were right that I'd meant Vmp but I'd think you'd use the 31 Vmp instead of 38 Voc in your calculations. That would put it at your original 24V figure which would be safely conservative for design purposes.

    That's an excellent point about measuring with the clamp-on between strings of batteries. I've two good (expensive) clamp-ons and do measure output between panel strings frequently (and lots of other things in the old house wiring down here. My Comitán house was built 150 years ago and the wiring is inventive, scary, and fascinating.) But I've not measured between battery strings during load or charging. Just the overall in and out. I'll have to start doing that.

    I've replaced plenty of battery banks. I usually size them around the 5% rate of charge figure for the array. What has worked very well is to yank out the poorly performing batteries as they near the end of their lifespan. The rate of charge goes up and (if you designed well originally) you've still plenty for the house consumption - just less leeway. With the brand of cheaper but solid batteries I generally use down here it works out to about 4 years (it's the one component you can get in Mexico that has better value than in the states - Mexican batteries). Isolating the weakest ones at that point and re-configuring the bank usually results in another 6-8 months of life for only about 30 minutes of work. I'd call that a nice percentage of extended life on 4 years. I've not found that setting up a smaller (but still adequate) bank results in any life span differences. Of course putting in an inadequate bank blows them pretty quickly. Fortunately I've only seen those on other guys "designs". As you've pointed out there are differences even in the same components. Banks show they are aging and nearing the end of their lifespan not only because they don't hold the charge as long as they used to but also they charge up very quickly. Under conditions that used to take say 3 hours to fully charge; a used up bank can charge up in an hour. At that point I usually start looking for bad batteries in the bank by simply looking at resting voltage when the system is shut down a few hours (not BAD but more significantly used up). But I like the idea of checking between strings. It'll probably give better indications earlier and also helps isolate problems with bad connections. Also as you pointed out rarely do we swap out a bank because they are used up. But you can extend the life on the bank by refinement.

    Interestingly enough down here there are a lot of small mom and pop places that will "rebuild" the batteries. Usually they're just for car batteries but I've checked for doing deep cycle ones. Getting the plates on those are really the issue and they usually DO try to swap them out on the car batteries. I've yet to do it but have considered simply cleaning and swapping out fluid. Usually the plates are more or less acceptable when we swap out the banks. As you pointed out they are not BAD. I'm not convinced it's totally worth it but for some client with no initial capital to swap out an entire bank maybe I'd help.

    I consider my work down here almost voluntary considering some of the clients I work with! I've taken up the philosophy of rob from the rich and give to the poor from the point of few that I work less using better, costlier, and prefabricated components for rich folks and use some of their swap outs and do more inventive rebuilding and homemade stuff for the poor ones.

    Also down here there's a small but growing industry in simply buying and selling used up batteries. The shell replacement credit has gotten to be 20% of the cost of a new one. For people trying to scratch out a living it's decent money.

    It's good to know that panels stay close to specs. Actually I'd forgotten I'd checked on the previously mentioned array of 80's - probably because I hadn't done it directly. The client had called me. Since I was about 13 hours away I had a friend take over to him one of my clamp-ons. I talked him through measuring the array current when it was about middle of the day but the batteries still hadn't gotten to bulk charge (I'd already changed out the bank). Considering the bit of loss for various things the array output was almost right on spec. So you're absolutely right in that regard. In this case I suspect it's excess consumption the last couple of years with a new renter in the house. He can't be bothered to change his habits like the original owner did to compensate. Also they use a pressure tank for water and the tank diaphragm (or bladder) is shot (and has been for 2 years) - resulting in more frequent pump start-ups. They already contracted with me to change out the tank when I get back up there. With more huge start-up draws on the pump I'm sure it's the main culprit.

    I'm a new convert to MPPT's. I've installed just 4 in the last couple of years and the last one was to my own system. I had some left over panels that I couldn't sell (having bought them at higher prices). So I reconfigured and remounted everything I had and it works wonderfully. I was pleasantly shocked at how much additional output I gained simply adding the MPPT. When I did the conversion I first reconfigured my original 12 panels from 6 pairs for my 24V system to 2 6's and put the mppt in place. I got about 15-18% extra current charging my batteries. Then I put in the other 6 that were new and I feel like I could sell electricity to my neighbors (if I had any)! The panels aren't all the same but they are close (120 and 115). I, of course, put like panels in series so the differences were isolated between parallel strings.

    Thanks again. Every time I post to this group I learn so much! I really appreciate your well thought out and completely detailed answers. Obviously it'll make a client happier also that I get it right for them (even if they don't realize it).

    Dan
  • BB.BB. Super Moderators, Administrators Posts: 29,189 admin
    Re: c60 for 24V parallel array

    Hi Dan,
    dellsworth wrote: »
    Hi Bill,

    I get all of the above and I really appreciate your responses. It's like getting a short course in off-grid designing. Did you happen to mean VOC instead of Vmp when referring to 38 volts? You were right that I'd meant Vmp but I'd think you'd use the 31 Vmp instead of 38 Voc in your calculations. That would put it at your original 24V figure which would be safely conservative for design purposes.

    No--I really intended for ~35 to 38 Volts Vmp for PWM (and minimum voltage for MPPT) controllers. If you have two Vmp~18.6 volt panels in series = ~37.2 volts (near my 38 volts). And you get ~18% voltage drop on a hot day for Vmp.

    An 18% drop on Vmp=31 volts means: 0.82*31 volts Vmp = 25.42 Volts "hot panel" Vmp

    Or, not enough to properly recharge a 24 volt lead acid battery to ~29 volts (AGM vs Flooded Cell/hot vs cold batteries, voltage drop for wiring+controllers+fuses, etc.).
    That's an excellent point about measuring with the clamp-on between strings of batteries. I've two good (expensive) clamp-ons and do measure output between panel strings frequently (and lots of other things in the old house wiring down here. My Comitán house was built 150 years ago and the wiring is inventive, scary, and fascinating.) But I've not measured between battery strings during load or charging. Just the overall in and out. I'll have to start doing that.

    We always suggest folks start with understanding the Smart Gauge battery wiring web page... Basically, batteries have very low internal resistance and any variation in wiring resistance can direct current to/away from a parallel string of batteries and cause uneven charging/discharging current among the strings.
    I've replaced plenty of battery banks. I usually size them around the 5% rate of charge figure for the array. What has worked very well is to yank out the poorly performing batteries as they near the end of their lifespan. The rate of charge goes up and (if you designed well originally) you've still plenty for the house consumption - just less leeway. With the brand of cheaper but solid batteries I generally use down here it works out to about 4 years (it's the one component you can get in Mexico that has better value than in the states - Mexican batteries). Isolating the weakest ones at that point and re-configuring the bank usually results in another 6-8 months of life for only about 30 minutes of work. I'd call that a nice percentage of extended life on 4 years. I've not found that setting up a smaller (but still adequate) bank results in any life span differences. Of course putting in an inadequate bank blows them pretty quickly. Fortunately I've only seen those on other guys "designs". As you've pointed out there are differences even in the same components. Banks show they are aging and nearing the end of their lifespan not only because they don't hold the charge as long as they used to but also they charge up very quickly. Under conditions that used to take say 3 hours to fully charge; a used up bank can charge up in an hour. At that point I usually start looking for bad batteries in the bank by simply looking at resting voltage when the system is shut down a few hours (not BAD but more significantly used up). But I like the idea of checking between strings. It'll probably give better indications earlier and also helps isolate problems with bad connections. Also as you pointed out rarely do we swap out a bank because they are used up. But you can extend the life on the bank by refinement.

    We have been using 5% to 13% as a good "rule of thumb" for charging current (xx% of 20 Hour Battery Capacity--I.e., 200 AH battery would be 10 to 26 amps charging current).

    As batteries get more expensive and solar panels get less expensive, we have been tending towards suggesting folks tend to the 10-13%+ rate of charge.

    One of the reasons is that many people these days are using power during the day (washing, computers, A/C, etc.)... So, a 5% solar array with loads during the day, starts pulling the charging current below that "magic" 5% minimum rate. Many battery mfg. suggest around 5% or even 10%+ rates of charge for "long" battery life (over ~13% rate of charge can be a bit "wasteful" of panel $$$--i.e., batteries get to float before noon, also deeply cycling banks can get hot from >C/8 rates of charge, so ideally, you should have a remote battery temperature sensor on the charge controller to help reduce the chance of thermal run-away).
    Interestingly enough down here there are a lot of small mom and pop places that will "rebuild" the batteries. Usually they're just for car batteries but I've checked for doing deep cycle ones. Getting the plates on those are really the issue and they usually DO try to swap them out on the car batteries. I've yet to do it but have considered simply cleaning and swapping out fluid. Usually the plates are more or less acceptable when we swap out the banks. As you pointed out they are not BAD. I'm not convinced it's totally worth it but for some client with no initial capital to swap out an entire bank maybe I'd help.

    If you can get old fork lift batteries--That may be a hit. One poster here would get (for example) 36 volt forklift batteries and bypass the bad cells (and keep the extra "good cells" as spares). He found only one or two bad cells, and the rest of the cells had years left of life.
    I consider my work down here almost voluntary considering some of the clients I work with! I've taken up the philosophy of rob from the rich and give to the poor from the point of few that I work less using better, costlier, and prefabricated components for rich folks and use some of their swap outs and do more inventive rebuilding and homemade stuff for the poor ones.

    We all do the best we can for those around us. Such as this forum with all the wonderful people here that volunteer their time to answer these types of questions.

    Also down here there's a small but growing industry in simply buying and selling used up batteries. The shell replacement credit has gotten to be 20% of the cost of a new one. For people trying to scratch out a living it's decent money.
    It's good to know that panels stay close to specs. Actually I'd forgotten I'd checked on the previously mentioned array of 80's - probably because I hadn't done it directly. The client had called me. Since I was about 13 hours away I had a friend take over to him one of my clamp-ons. I talked him through measuring the array current when it was about middle of the day but the batteries still hadn't gotten to bulk charge (I'd already changed out the bank). Considering the bit of loss for various things the array output was almost right on spec. So you're absolutely right in that regard. In this case I suspect it's excess consumption the last couple of years with a new renter in the house. He can't be bothered to change his habits like the original owner did to compensate. Also they use a pressure tank for water and the tank diaphragm (or bladder) is shot (and has been for 2 years) - resulting in more frequent pump start-ups. They already contracted with me to change out the tank when I get back up there. With more huge start-up draws on the pump I'm sure it's the main culprit.

    Big difference between the owner and the renter/vacation user... You will have to figure out exactly what is causing issues. Usually starting current surge is "short" so that if you have a good bank, "short pumping" because of a failed pressure tank diaphragm should not dramatically affect the battery bank/inverter performance (may kill pump/electrical contacts on pressure switch).
    I'm a new convert to MPPT's. I've installed just 4 in the last couple of years and the last one was to my own system. I had some left over panels that I couldn't sell (having bought them at higher prices). So I reconfigured and remounted everything I had and it works wonderfully. I was pleasantly shocked at how much additional output I gained simply adding the MPPT. When I did the conversion I first reconfigured my original 12 panels from 6 pairs for my 24V system to 2 6's and put the mppt in place. I got about 15-18% extra current charging my batteries. Then I put in the other 6 that were new and I feel like I could sell electricity to my neighbors (if I had any)! The panels aren't all the same but they are close (120 and 115). I, of course, put like panels in series so the differences were isolated between parallel strings.

    In regions with "cold" weather, yes you can get somewhat better performance with MPPT controllers... But we try not to "sell" readers here that is why to purchase MPPT. Normally MPPT makes sense because you can buy "24 volt panels" (which are really Vmp~30 volts) and rewire them to ~60-90 volt Vmp-array setups. That allows less copper and/or longer runs from panels to battery shed, and allows you to use "non-standard" Vmp panel on battery bank (again, issue of 31 Vmp hot panel only outputting <26 volts Vmp on hot days)

    Thank you Dan for the kind words.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
Sign In or Register to comment.