The drawing board has drawn me back with more questions

elesaver · October 2014

Here's the situation: my system is up and running. I am tweaking it at the moment and seem to be unable to unravel what is the best configuration of things. Components: 6 ea 300W panels (36.73W nominal, 8.18 a nominal), Midnite Solar combiner box outside, Midnite Solar mini dc breaker box inside, Midnite solar Classic 200 charge controller, 1500W Cotek 24V inverter, 8 ea Trojan 125 batteries @ 240 aH each in series/parallel for 480 aH.

The panels have been wired as 2 panels in series with 3 strings to combiner box. If I understand, that only gives me 24 amps for battery charging. I believe I need more amps than that...10% of 480 = 48 amps. The batteries do go into float during the day but the aH total is never back to the 480 even though the Classic is saying the batteries are 100%.

I have run some different things on the system to see what happens. I ran a 135W, 5 amp freezer for 12 hours overnight with a battery aH "cost" of 95 aH. During the day, even though the batteries were showing 100% charge, the aH total available did not return to the full 480. Using a voltmeter on the batteries, they were above the daily reading but not to the float value but the charger was in float.

I may not have the correct settings entered into the controller and that is limiting the process. I just can't figure it out.

So, my question is twofold: should I put the panels into 2 parallel strings of 3 panels each and then run them to the combiner box? That would output the 48 amps I think I need. The Voc would be approx 110. OR should I change the settings in the controller? The re-bulk value is a bit of a mystery to me. I put that value just below the float value so if the charge falls below float, the batteries would go back into the bulk MPPT.

If I need to give more info to help define the problem, I'll be happy to do so. Meanwhile, I will keep scratching my head to figure this out.

Thanks to anyone who will give this a shot for me.

Is there a way to know what the panel output is?

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

You have six 300 Watt panels on an MPPT controller? That's 1800 Watts. You have a 24 Volt system so the maximum output current will be about (1800 * 0.77 / 24) 58 Amps. No matter how you configure the panels, providing the Voltage minimum is met for a 24 Volt system (i.e. array Vmp >34).

Your panel specs are "wrong": I think you mean 36.73 Volts, not Watts, and I guess that is Vmp and the 8.18 is Imp. The word "nominal" does not apply here. If my guess is correct then that does come out to 300 Watt panels (36.73 * 8.18 = 300.45). At that Vmp they could all be used in parallel and even on a PWM controller with a resulting current maximum of (6 * 8.18 ) 49 Amps. I don't know how you came up with a 24 limit.

Your existing configuration of three parallel strings of two in series is probably optimum, depending on wire size & length from the combiner to the controller. If you put them as two strings of three you will probably lose power as the controller's efficiency will decrease slightly having to down-convert the higher array Vmp.

Do not expect to see this maximum current just because it is possible. The batteries will only take what they will, regardless of PV potential. This can sometimes be more than what is good for them, but usually isn't (especially at low DOD). The 58 Amps potential on 480 Amp hours is a 12% max charge rate and that should not be a problem.

How are you measuring this 95 Amp hours used by the freezer? And if it is not being measured there's no reason to believe it is accurate: calculations do not work for this sort of thing.

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

Yes, my error in 36.73 Volts per panel, not watts. Could this be the problem? I have had 3 strings of 2 panels wired in series. These, then, are wired into the combiner box to their own 20 amp breaker. I have 3 breakers total with one string on each. I thought that the amps per 2 panels would be 8.18 for a total of 24.54 amps for the entire array.

For some reason, this is just not registering with me. I do not understand how to wire the suggested 3 parallel strings of 2 in series.

Today, I have placed 3 panels in parallel with those + and - going to one breaker in the combiner box and the other 3 panels in parallel going to another breaker. So, I have 2 parallel strings of 3 panels. This, in my mind, would yield the 48+ amps...24 + 24 from each string.

I think I'm making this harder than it is.

The reason I thought the freezer used 95 aH was due to the reduction of battery aH remaining the following morning. Yes, the inverter would've used some but this is how I came up with that number.

inetdog · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

Yes, my error in 36.73 Volts per panel, not watts. Could this be the problem? I have had 3 strings of 2 panels wired in series. These, then, are wired into the combiner box to their own 20 amp breaker. I have 3 breakers total with one string on each. I thought that the amps per 2 panels would be 8.18 for a total of 24.54 amps for the entire array.

For some reason, this is just not registering with me. I do not understand how to wire the suggested 3 parallel strings of 2 in series.

Today, I have placed 3 panels in parallel with those + and - going to one breaker in the combiner box and the other 3 panels in parallel going to another breaker. So, I have 2 parallel strings of 3 panels. This, in my mind, would yield the 48+ amps...24 + 24 from each string.

I think I'm making this harder than it is.

The reason I thought the freezer used 95 aH was due to the reduction of battery aH remaining the following morning. Yes, the inverter would've used some but this is how I came up with that number.

The word string implies that the panels are connected in series.
So you do not have any strings of panels, you just have two groups, each of which is three panels in parallel.

If you wire the + of one panel to the - of the next, and the same from the second to the third panel, then there is one leftover - and one leftover +. On those two leads (the string output) you will see the single panel current, but three times the voltage.
If you take two such strings and put the in parallel, you have two strings of three panels.
If you make your strings of only two panels each, then you would have three two-panel strings and you could connect those in parallel to get three times the single panel current at twice the single panel voltage.

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

I think I better look and see if I've got some ready-made pictures for panel wiring as the words are getting confused.

With panel Vmp at 36 you could have all the panels in parallel on a 24 Volt system, no problem. The current from the array would be 6 * 8.18 or 49 Amps. When using an MPPT type controller the array specs are not actually relevant to the output specs except for Wattage, as it will adjust V and I to what it thinks is best for maximum output. The array Voltage needs only to be higher than the battery (charging) Voltage.

You will not get maximum power from an array unless conditions are met: there must be full, direct sun on the PV so that it can produce as much as it is capable of (which will still average less than nameplate rating) and there must be enough load (including battery charging) on the output of the controller to demand this power. So you can have 1200 Watts of PV potential but unless there is a 1200 Watt (or more) demand on the output the controller will not pass all that power.

Image shows two parallel strings of three in series (inverse of what you have).
Attachment not found.

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

inetdog,

Thank you for your assistance. I still have a question, however. You are saying, "If you make your strings of only two panels each, then you would have three two-panel strings and you could connect those in parallel to get three times the single panel current at twice the single panel voltage." This would result, then, in a 24+ amp current and 72+ voltage unless I am still calculating incorrectly.

24 amps for a 480 aH battery bank doesn't seem like enough. But, again, it may all be due to my miscalculations. If I have 2 "groups" of 3 panels in parallel with those 3 panels into 2 separate breakers, then would I not have 48 amps and 110 volts?

Bear with me...I'll get it. :-)

inetdog · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

inetdog,

Thank you for your assistance. I still have a question, however. You are saying, "If you make your strings of only two panels each, then you would have three two-panel strings and you could connect those in parallel to get three times the single panel current at twice the single panel voltage." This would result, then, in a 24+ amp current and 72+ voltage unless I am still calculating incorrectly.

24 amps for a 480 aH battery bank doesn't seem like enough. But, again, it may all be due to my miscalculations. If I have 2 "groups" of 3 panels in parallel with those 3 panels into 2 separate breakers, then would I not have 48 amps and 110 volts?

Bear with me...I'll get it. :-)

Let me make up some numbers instead of trying to figure out yours:

One panel, 300W, in the form of 10A at 30V.
Three of those panels in parallel = 30A at 30V.
Three of those panels in series = 10A at 90V.
Both ways the total power from the panels will be 900W, and your MPPT CC can accept the input power in either form.

Or, using six panels instead of only three you can put two panels in series to get 10A at 60V, then put three of those strings in parallel to get 30A at 60V, for a total of 1800W.
With an MPPT controller, that full 1800W will be converted to, say, 24V at 75A (Power IN equals Power OUT with MPPT)
Assuming of course that your MPPT CC can handle a 75A output in the first place.

Rule 1: When you put two things in parallel, voltage stays the same and amps add up.
When you put two things in series, amps stay the same and voltage adds up.
Rule 2: When you use an MPPT controller, both volts and amps at the output can be totally different from the volts and amps at the input. What stays the same from input to output if the power (Volts times amps.)

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

Cariboocoot,

"A picture is worth a thousand words." I tried to google to find a picture of what we are discussing but I was unsuccessful. The calculations seem to be throwing me off. Is this correct? The 3 panels in each series is going to total 8 amps and with 2 strings, it becomes 16 amps. By then putting those in parallel, the amp total increases to 32 but the voltage does not change from the 110 series voltage. The volts would be about 110 at the final parallel connection.

Am I getting close?

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

inetdog,

Thank you for your assistance. I still have a question, however. You are saying, "If you make your strings of only two panels each, then you would have three two-panel strings and you could connect those in parallel to get three times the single panel current at twice the single panel voltage." This would result, then, in a 24+ amp current and 72+ voltage unless I am still calculating incorrectly.

For input to the charge controller that is correct:
Two panels is series brings the Voltage to 72, three 8 Amp strings in parallel brings the current to 24.

This is completely irrelevant to the output from the controller to the batteries. Because it is an MPPT type controller the output to the batteries will be based on input Watts, no matter how those Watts are configured (providing minimum Voltage is met).

Six 300 Watt panels, however arranged, is 1800 Watts.
1800 Watts * 77% derating / system nominal Voltage = output current.
Or as I said before 58 Amps (57.75).
That is more than enough for 480 Amp hours.

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

Cariboocoot,

"A picture is worth a thousand words." I tried to google to find a picture of what we are discussing but I was unsuccessful. The calculations seem to be throwing me off. Is this correct? The 3 panels in each series is going to total 8 amps and with 2 strings, it becomes 16 amps. By then putting those in parallel, the amp total increases to 32 but the voltage does not change from the 110 series voltage. The volts would be about 110 at the final parallel connection.

Am I getting close?

Almost there. Leaving off the decimal places to ease the math:
If you put three 36 Volt 8 Amp panels in series you get 108 Volts at 8 Amps. Double them and you get 108 Volts @ 16 Amps.
If you put two 36 Volt 8 Amp panels in series you get 72 Volts at 8 Amps. Triple them and you get 72 Volts @ 24 Amps.
If you put three 36 Volt 8 Amp panels in parallel you get 36 Volts at 24 Amps. All six in parallel you get 36 Volts @ 48 Amps.

Again this does not (significantly) effect the output of the controller. You only need that 36 Volts for charging a 24 Volt system and any of these configurations meets that minimum Voltage requirement. All should be able to produce the 57-58 Amps peak current (if demand requires and insolation is sufficient).

There is no reason to go with 108 Volt series strings unless you have very high wiring losses between the array and controller (long distance, small gauge wire). In fact the higher the array Voltage is in respect to the system Voltage the less efficiently the controller functions so you may encounter an actual decrease in available power.

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

"There is no reason to go with 108 Volt series strings unless you have very high wiring losses between the array and controller (long distance, small gauge wire)."

So, this is to say NOT to wire these as 3 panels in a series and then parallel? That's what I thought the illustration was showing was correct.

The amps is what I have been focusing on, but perhaps it's the volts that is the most important? If true, since the panels are 36V, it wouldn't really matter how they are wired? I thought the combiner box was a way to "combine" various strings of panels in series and that the output wire from the combiner box to the controller was doing just that...combining the values of the panel strings. I had had 3 strings of 2 panels in series so that resulted in 72 volts at 24 amps into the combiner box. Those 24 amps did not seem enough (to me) to charge my 24V bank but apparently it is the volts that counts. SO, in the tweaks menu, I need to set the max voltage to be more than 36?

(I am going to be bald shortly...maybe you are, too...from pulling my hair out. :-) )

The perceived problem began when I haven't thought the batteries were returning to their full charge. Yes, the charger went from bulk to float. I have used a voltmeter on the batteries (I'm getting an hydrometer to get an accurate SG which will probably answer a lot of questions.) and the voltmeter shows the batteries to be slightly less than the daily recommend from Trojan even though the controller says 100% SOC. The aH total is not back to the full 480 which also throws me. I thought it must be the way I have the panels wired so that adequate current is being supplied to charge these. The problem may exist in the way I have the controller programmed. I love Midnite Solar but the manual is less than helpful for those who are new to all of this. There are no explanations given as to what changes (up or down) in the values will mean. When I don't understand the whys of something, I can't figured out what to do. Is there information somewhere that helps with understanding what controller setting changes do? For most of the people who take on this challenge of a solar array, it is "easy peasy." BUT, there are a few of us (or I may be the only one) who is slogging through learning. Fortunately, this board is a HUGE help but sometimes I feel ridiculous in my lack of grasp on things. Thanks for the gentle and kind spirit that is here.

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

I've been bald for years.

The illustration I put up was just an explanation of what two parallel strings of three looks like, not a recommendation.

Really the array configuration makes no difference to the output of an MPPT controller.

Unless you have a shunt-based battery monitor you do not have any meter which can accurately show SOC. Voltage does not count, especially "on the fly". Charge controllers (other than the Classic or Kid with a WhizBang jr. attached) do not display this info. The Outback FNDC monitor would. The TriMetric would. These last three are shunt-based. It is the only electronic way to measure SOC. And if they are programmed wrong they read wrong. Only a hydrometer can tell you for sure what the SOC is, and you have to read that right.

You have more than enough PV to charge the amount of battery you have. If they really aren't being charged fully, then the parameters are probably set wrong. Trojan wants fairly high Absorb Voltage (29.6 on a 24 Volt system) and the other major culprit is likely insufficient Absorb time. Keep in mind a Classic can be programmed for maximum time, End Amps, and/or SOC based ending of Absorb. Get the numbers wrong and it goes to Float before it should.

Sorry the MidNite manual is not clear. I didn't write it*.

And anyone who thinks this stuff is easy has not been paying attention. It's easy to get it wrong, and there the "easy" ends.

*I don't work for them or any other company, including the site host.

vtmaps · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

I had had 3 strings of 2 panels in series so that resulted in 72 volts at 24 amps into the combiner box. Those 24 amps did not seem enough (to me) to charge my 24V bank but apparently it is the volts that counts. SO, in the tweaks menu, I need to set the max voltage to be more than 36?

It's been explained, but I will try with a more concrete example:

You have 3 strings of 2 panels in series so that resulted in 72 volts at 24 amps into the combiner box.

72 volts X 24 amps = 1728 watts at the combiner.

Let's say all of that power gets to the controller (no resistance loss in the cable).

So 1728 watts into the MPPT controller.

Let's pretend that your controller is 100% efficient so that means 1728 watts is going into your battery.

At the moment your battery happens to be at 28.5 volts.

1728 watts ÷ 28.5 volts = 60.63 amps going into the battery.

Now, your panels will seldom produce 1728 watts
There will be some power lost in the cable.
Your charge controller is not 100% efficient.

But as Cariboocoot explained, you have enough panel for those batteries.

--vtMaps

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

I see what you're doing there, Cariboocoot. :-) The drawing was to teach me something, not instruct me as to what to do. I've think I'm getting it. What I've also got is more of the concept, I think. (By the way, I do have a Whizbang, Jr. That's where I am getting my readings about SOC% and aH remaining in the batteries.)

So, to "know" what's going on with this whole system, one needs to know how many watts are going to the controller. This is a fixed number, based on the panels nameplate minus a small amount because they never "do as advertised," so to speak. Then, with that number, you can divide the panel watts by the battery voltage to see how many amps are charging at the moment. For the sake of any others who may be finding this thread informational, I will admit to my former ways. I would see on the Classic panel a listing for amps. That number would change frequently. Then, I would see the accumulated KWh which would change but not as frequently. Then, I would go to the Whizbang, Jr screen and see the SOC% and the aH of the batteries. That would frequently say 100% as SOC but the battery amps were far lower than full charge. In one of the menus, there is a selection for maximum volts. That was set at a default of 31.1 I didn't know (maybe still don't) what that should be. That is not explained in the MS Classic manual. Now I think that it must be how many volts you want to come in to the controller. Since I need at least 36 volts for a 24 system, I've upped this value. In fact, I put it at 48 Volts as the maximum.

I'll continue to slug along here, gaining valuable experience and information. I do so appreciate the time and effort of all of you to help. I'm sure it's frustrating to try to help someone who doesn't understand when it is so obvious to you. Your kindness is not overlooked. Thank you.

As Columbo would so..."Oh, and another thing..." My panel configuration is not as critical as I had originally thought. I'm going to return to the 3 strings of 2 panels each into the combiner. That way, each string can be isolated in the event there is a problem in the future.

By George, I think I've got it. (It's a mistake to say that because an error is on the way!) Thanks again.

Cariboocoot · October 2014

Re: The drawing board has drawn me back with more questions

I believe the maximum Voltage on the Classic is for the output, not the input. People who have MidNight controllers can explain exactly what the settings are for and where to find them.

The reason why you need a Vmp around 35 is because once the panels heat up their Voltage goes down and it goes down further due to wiring losses. So if you want to have enough Voltage to bring your batteries up to 28.8 or 29.6 or EQ at 31 ... you've got to start out with more Voltage than you need.

I suspect you have a problem along the lines of incorrect Amp hour capacity or efficiency entered, or perhaps just wrong Voltage/time/End Amps for SOC.

Photowhit · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

By George, I think I've got it. (It's a mistake to say that because an error is on the way!)

Loaded up my dehydrator Sunday and went to town, dang thing doesn't work very well when not turned on!

While boB, and Robin stop by on this forum, There is an active Midnite Forum, where you might have questions answered by the engineers who designed the Charge Controllers. Wish I had my Whiz Bang set up, I might be able to help, but the last part of my system goes in this fall and then I'll have the Whiz Bang installed and do some of the more fancy things with my extra energy...

Rybren · October 2014

Re: The drawing board has drawn me back with more questions

I have a Classic 150. I don't recall a Maximum Volts setting (but there may be one)

You need to set the Absorb Voltage, Float Voltage, and Equalize Voltage. In addition, you can set the Maximum Absorb time and End Amps.

As for the AH capacity in the WBJr menu, I believe that the capacity is temperature and efficiency adjusted. So if you told the controller that your battery capacity is 450AH at 25 degrees C, and the actual temp is 15 degrees, then the displayed AH will be lower than 450AH, even though the batteries may be at 100% SOC.

The ONLY way to be certain that your batteries are at 100% SOC is to measure the SG (and then adjust the readings for temp)

Alaska Man · October 2014

Re: The drawing board has drawn me back with more questions

Hard to tell from the terms you use to desrcibe what's going on with your Classic, but You may be misinterpreting "Amps in" as State of Charge? The Classic will reduce amperage going into your batteries as the bank gets closer to 100% SOC.

Change your "end amps" to .5 and increase your "Absorb Time" by an hour and see if that helps.

For what it is worth, the "Volts" reading really isn't a good way of determining anything. You can have 1,000V showing on the Classic and have zero power going into your battery bank.

inetdog · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

Cariboocoot,

"A picture is worth a thousand words." I tried to google to find a picture of what we are discussing but I was unsuccessful. The calculations seem to be throwing me off. Is this correct? The 3 panels in each series is going to total 8 amps and with 2 strings, it becomes 16 amps. By then putting those in parallel, the amp total increases to 32 but the voltage does not change from the 110 series voltage. The volts would be about 110 at the final parallel connection.

Am I getting close?

Not sure about that....

If you take three panels and put them in series, yes you get 8A, and if you put two of those strings in parallel you would get the same three-panel voltage but a total of 16A.
So far so good. Now you say you are going to put "those" in parallel to get 32A. Does that mean that you would have a total of 12 panels? Or have you just lost track of how many you have used already?
And yes, regardless of how may thee panel strings you parallel you will still have close to 110 volts.
But that would be the Vmp of the array. The Voc would be closer to 140V, and even higher on a cold day. That voltage is what the CC would have to be able to safely accept. If it cannot do that, then you are back to using two-panel series strings, each string giving you 8A at 72V. Three such strings in parallel would give you 24A at 72V.
If you add six more panels, for a total of 12, you would have 48A at 72V.

elesaver · October 2014

Re: The drawing board has drawn me back with more questions

Thanks for the suggestions. I have set the minimum absorb, maximum absorb, all the voltages per Trojan for a 125 battery, end amps, etc. I have set the end amps at .5 but didn't notice any change. Perhaps it would take more than a few hours. In another thread somewhere, it suggested setting end amps at 10% of bank aH, so that's why I set it at 4.8. I'll keep "tinkering." Right now there is no load at all so tomorrow I'll begin anew. Thanks. If you think of anything, please bring it to my attention.

I did go back to the menu to find where I saw the voltage limit. That is under Limits and has to do with the temperature compensation.

I may not be doing it right but I'm sure learning a lot! Life is good.

vtmaps · October 2014

Re: The drawing board has drawn me back with more questions

elesaver wrote: »

I have set the end amps at .5 but didn't notice any change. Perhaps it would take more than a few hours. In another thread somewhere, it suggested setting end amps at 10% of bank aH, so that's why I set it at 4.8.

When you set end amps very low (like to zero) you effectively disable it. You disable it because you can never achieve it (end amps usually doesn't go much below 1 or 2 percent of the battery ah capacity).

When you set the maximum absorb time, you are setting the time that absorb will end if end amps has NOT been achieved.

If you want to use end amps charging, you must set a maximum absorb time that is long enough to let end amps be achieved.

Your controller will switch to float when either condition is met: you have achieved end amps or you have run up against the time limit.

elesaver wrote: »

I did go back to the menu to find where I saw the voltage limit. That is under Limits and has to do with the temperature compensation.

When your batteries are cold the actual charging voltage is higher than what you set. Some inverters shut down if the voltage gets too high, so the max voltage setting lets you put a cap on how high temperature compensated voltage can get.

--vtMaps

The drawing board has drawn me back with more questions

Comments

Categories