310w panel Vmp issue

BB. · October 2016

On my phone, so cannot look up big specifications/user manuals... But the Victron, while a good product/company, the Vmp-array-low-voltage limitation is a Physical design limitation of buck mode switching (most mppt use some version) and will not fix anything.

If you pay for extra panel and material costs, and he gives free labor, would seen to be an ok solution (you are getting more panel/power, if you want that).

2s is just on edge. Solar systems are exposed to wide ranges of temperature and weather. 2s with a pwm controller would be a better fit... But you have an expensive mppt controller and 50 feet to the array (mppt is great for use with long distances from array to controller and use of non-standard Vmp voltage panels (gt panels are larger and cheaper than most battery+pwm controller compatible panels in Vmp~18 volt increment panels), to answer why mppt in the tropics).

I try not to argue with installers on site. I am not there and do not know the details. Instead I try to educate the owner so they can talk details with the installer, and sometimes even catch problems with the install. And give the owner enough information to do the install, if they choose to.

No worry about you "saving" money. We all started at the same point and received help from others/teachers.

-Bill

Photowhit · October 2016

I think the Victron has the same input specs with a lower output capability...

I'll check when I get back.

littleharbor2 · October 2016

Trading out a top of the line 93 amp controller for a middle of the road 60 amp controller doesn't sound like a solution to me. You have a couple choices.
1, lose two panels and go 3Sx2P.
2, Get one more panel, 3Sx3P.
3, Get a higher Voc controller, ie; Midnite 200, Schneider 80/600, or Magnum PT-100.

Photowhit · October 2016

Looks like they need to reconsider the option of using a Victron 150/60 bluesolar, the standard spec's show;

"Maximum PV open circuit voltage of 150V absolute maximum coldest conditions! This would be the VOC of your panels x4 or 45.9 volts x 4 = 183.6!

Data sheet for charge controller;
https://www.victronenergy.com/upload/documents/Datasheet-BlueSolar-charge-controller-MPPT-150-45-up-to-150-100-EN.pdf

I think you should talk with them to be sure they understand the nature of your problem There is some problem with their understanding of the problem or communications.

dapago · October 2016

Photowhit
When designing a system and selecting the CC according to the wiring of the paneles, do you take the Vmp figure in consideración or the VOC?
With 2s4p (at least on that aspect of my issue) the 150V CC should work or am I wrong here?
I am trying to tell them to add one more panel and get 3s3p the 150V CC should work as well if I follow your maths.

BB. · October 2016

2s is on the low Vmp-hot side for any mppt type charge controller (there is one or buck-boost charge controller, but it is very small). Warm too hot climate, cool battery bank, mppt controller all is on the edge.

3s system is perfect for warm climate with 48 volt battery bank.

If you lived in an area with frosts and snowy/cold weather, then Vmp-cold would be over 150 vdc. Vmp-array over roughly 100 vdc can cause Voc-cold in sub freezing climates.

You do have to design systems for where they are installed, as well as all the other stuff.

-Bill

Dave Angelini · October 2016

" Midnite appears to one of the few companies to use the remote battery shunt to measure charging current to the battery bank.."

Outback does this BTW Bill and Schneider does it with an internal shunt on the MPPT's as well as external shunts with the battery Monitor and back all the way to Trace. Nice tutorial thanks for the coffee....

This is another reason for a large installation to use a networked solution from one company. The inverter knows its load current and the system subtracts this value from the charge sources (CC's gen, grid) and tells you the battery charge current and uses this value for end amps in battery charging. No external shunts are required with Outback and Schneider, maybe magnum is doing this also?

westbranch · October 2016

read the specs for that CC!

Right now you have a MN Classic it can receive (input of) 150 Volts PLUS Hyper voltage safety net, the battery bank voltage, eg 24V ,so that equals 194Volts where it goes into self preservation mode and shuts down. it also throttles down below the 150V level but does not shut off at that point..

The Victron unit has the same 150V ABSOLUTE MAXIMUM COLDEST CONDITIONS https://www.victronenergy.com/upload/documents/Datasheet-BlueSolar-charge-controller-MPPT-150-45-up-to-150-100-EN.pdf
AND a lower Amperage that it can send to the batteries

That makes the Victron a lower performing CC than the Classic 150.... your choice

dapago · October 2016

Bill. Could you explain why PWM would be a better fit? I thought PWM controller were less efficient than MPPT.

BB. · October 2016

More or less, a properly designed pwm controller is just about the same efficiency as a mppt controller.

For paper designs, I treat then the same.

In general, for less than 400 Watt systems with the array close to the battery shed, PWM will end up a bit cheaper.

For systems over 800 Watts, they can be build less expensive with larger format "cheaper Grid Tied" solar panels vs the more expensive MPPT controllers.

But there are several good reason for MPPT...

1. You can run higher voltage arrays so you can run longer (smaller diameter) cable runs (save lots of money on copper).
2. You can run higher Vmp-arrays so in Hot Climates, you can have higher charging voltages during hot weather (60-63 volts or more sometimes).
3. You can run cheaper solar panels (140 watt panel with Vmp~18 volts is ~$2 per watt, vs >200 Watt panels with Vmp~30 volts and $1 per watt).

Until you do the paper designs, you will not see the advantages and disadvantages of the different designs... And which will work best for you.

Your system at 2,400+ watts, MPPT charge controller makes sense. You just have to use 3 in series to do that. You could (technically) put 2 in series and a PWM charge controller (less than 5 volt drop), and the system would work well--But the cable diameter might be too much... For example, say you have 50 foot cable run, want 1% drop (~0.78 volt drop) on 78 Vmp array, and 8.5 amps per string--The copper wire would look like:

http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=2.061&voltage=78&phase=dc&noofconductor=1&distance=50&distanceunit=feet&amperes=8.5&x=43&y=12

8 AWG, 1 pair for each series string back to the controller (4p would be 8x 8awg cable):
Voltage drop: 0.53
Voltage drop percentage: 0.68%
Voltage at the end: 77.47

Or, if you had all 4p into one cable (4*8.5=) 34 amps

http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=0.5127&voltage=78&phase=dc&noofconductor=1&distance=50&distanceunit=feet&amperes=34&x=37&y=13

2 AWG, all current in one pair of cables for 50':
Voltage drop: 0.53
Voltage drop percentage: 0.68%
Voltage at the end: 77.47

So--you can see that very low voltage drops (needed to run PWM controller efficiently) require large diameter of very expensive copper cable.

There is a "hole" in system designs (especially a few years ago before there were "high voltage" MPPT charge controllers) were certain voltage panels can be difficult to "hit" the Vmp~panel and Vmp~array "optimal" input voltage range. For example, we we used to get ~60 volt Vmp panels (early GT type thin film panels)---60 volts was to low for charging a 48 volt battery bank, but two in series (Vmp-array~120 volts) was too high for colder climates...

But you have to make do with what you have... I.e., with the panels you have, you are limited to 3 series (3 panels, 6 panels, 9 panels, etc.). It is not a bad limitation, but if you had your heart set on 8 panels--Then that is an issue with a 150 volt Vmp controller. Go with a 200 volt controller--You can go 4/8/12 panels... But the 200 volt controller has reduced output rating (not ~90 amps, but 60-70--So it is limited to a smaller array). May not matter to you at all.

Not bad or good--Just limitations that we have to live with/work around.

-Bill

dapago · October 2016

Thank you for your answer. It is not that I have my heart on 8 panels but just the space available on the roof. If I want one more panel The installer has to redesign all the panels structure and installation and still don't know where he will find the extra space for it. Not impossible of course but not easy, more old tiles will be broken (difficult to replace), etc, and the result will definitively not be aesthetic.
Time now to argument with the installer. Or replace the 150 CC with a 200, or add 1 panel and redesign the roof structure and installation. Will keep you updated.

Photowhit · October 2016

dapago said:
When designing a system and selecting the CC according to the wiring of the paneles, do you take the Vmp figure in consideración or the VOC?
With 2s4p (at least on that aspect of my issue) the 150V CC should work or am I wrong here?

You must take both VOC and VMP into consideration, at least with MPPT type Charge Controllers (CC).

These CC's can be fouled up with a VOC that is too high, and that must be considered for the absolute maximum. I don't understand the idea, but the higher VOC can create problems for the controller, this is often before the CC can begin producing any output and is something I don't understand. (Bill might, having stronger electronic background)

The VMP is during production and is the maximum the panel will produce in warm climates (panels can actually produce higher Wattage in cold climates!) Typically your panel will produce less volt when warm/hot, there is a NOCT value that some manufacturers provide, others just provide a coefficient to calculate the values.

Whether a 2s4p is the point we have been discussing. As I read Midnite's information, it would not work well.

As Bill has said, a PWM would likely work fine with a 2s4p, if there wasn't a particularly long run from the array to the CC/Batteries. There are other reasons to have a MPPT type CC though. Ones you may well enjoy. Midnight has a shunt that sits in front of your battery bank and monitors the actual current running into your battery bank. This could be a very nice feature with someone who is running a lot of daytime loads (as you have described your system). Schnieder has a similar setup and Outback, uses a second controller that does some math to figure out how much is going into the battery bank (I have never installed an Outback system, but Dave may explain it better, I think it requires use of an Outback inverter, though this may be erroneous information.

Clear as mud?

dapago · October 2016

Photowhit said:

dapago said:
When designing a system and selecting the CC according to the wiring of the paneles, do you take the Vmp figure in consideración or the VOC?
With 2s4p (at least on that aspect of my issue) the 150V CC should work or am I wrong here?

You must take both VOC and VMP into consideration, at least with MPPT type Charge Controllers (CC).

These CC's can be fouled up with a VOC that is too high, and that must be considered for the absolute maximum. I don't understand the idea, but the higher VOC can create problems for the controller, this is often before the CC can begin producing any output and is something I don't understand. (Bill might, having stronger electronic background)

The VMP is during production and is the maximum the panel will produce in warm climates (panels can actually produce higher Wattage in cold climates!) Typically your panel will produce less volt when warm/hot, there is a NOCT value that some manufacturers provide, others just provide a coefficient to calculate the values.

Whether a 2s4p is the point we have been discussing. As I read Midnite's information, it would not work well.

As Bill has said, a PWM would likely work fine with a 2s4p, if there wasn't a particularly long run from the array to the CC/Batteries. There are other reasons to have a MPPT type CC though. Ones you may well enjoy. Midnight has a shunt that sits in front of your battery bank and monitors the actual current running into your battery bank. This could be a very nice feature with someone who is running a lot of daytime loads (as you have described your system). Schnieder has a similar setup and Outback, uses a second controller that does some math to figure out how much is going into the battery bank (I have never installed an Outback system, but Dave may explain it better, I think it requires use of an Outback inverter, though this may be erroneous information.

Clear as mud?

My site is on a lake (island) so clear as water there can be :-)
Never thought I would have to get my head so much on a technology (solar) I am totally ignorant to. The good thing in a way is that I have now to learn fast and I did not have any other alternative anyway.I suppose this is what you have to cope with when you live in a third world country and want any technology to work trouble free. I was expected at least some time with the system running to monitor it and learn how it works with help from this forum. The good thing is I haven't paid the full amount yet of the solar bill to the installer/dealer as I suspected something might go wrong and asked for the system to run test at least for a week until giving the 50% left to pay. Did not last a minute...
Not even talking about the legal system here protecting the consumer this is why I was so cautious. As a matter of fact, the installer/dealer, since I proposed the 2 solutions given here (apart rewirering the panels), bigger CC or 1 more panel, does not answer my calls or emails anymore.
Photowhit:
what is a shunt?

Bill.

Just had a check on the copper cable installed and it says 6AWG. Until a solution is found, I am now working with 3s 2 p (2 panels are disconnected).

Sorry guys (and Bill in particular) more questions (hope you all don't get bored by my requests):
- with the 6AWG installed, shouldn't it be 4 cables (2 per string)?
- if I run 3s3p (with luck), what cabling should be used?
- Will the 6AWG be enough to run a PWM with the 50 feet distance and run 4s 2 p?
- And last (for the moment lol) considering a PWM, what model would you suggest?

Dave Angelini · October 2016

A shunt is how the current is measured in specific places in a system. In a netorked system such as Outback or Schneider the inverter knows how much current it is using and the system subtracts this value from the charge sources (CC,s Gen, Grid) and gives you the current into your battery without external shunts. This battery current is then used to send the charging to float once the end amps spec is met. Much simpler to install and easier for people to understand. This has been around for nearly 12 years BTW.
Magnum is new at this but I would think their CC should communicate with the inverter and eliminate the external shunt also.
Anyone with a magnum system know?

BB. · October 2016

A "shunt" in this case, is a precision high power/current resistor, usually with attachment screws that connect to a low voltage (milli-volt) meter, like this:

The big bolts are for your heavy cable (typically from battery negative post to the negative bus bar), and the small screws are for accurate voltage measurement by a meter or battery monitor (computerized volt meter). The volt meter is measuring (full scale) around 0.05 to 0.10 volts (50 to 100 mVolt at rated current). The meter uses V=IR; I=V/R to give you current.

The numbers I gave you as an example in the previous post were for 1% drop--And pretty heavy cables... If you have 6 AWG from the 2s x 4p array with 50' run and 6 AWG cable (you do not want to run 4s x 2p with a PWM controller, even if you had one with high enough input voltage rating, which you may not find, you would lose about 1/2 of the panel wattage--But this may be just a typo on your part 4s x 2p in your question).

8.3 amps * 4 parallel strings = 33.2 Amps total (Imp-array)
37 volts * 2 series panels = 74 VDC (Vmp-array)
50' run (one way), 6 AWG, 74 volts nominal array voltage

Voltage drop: 1.31
Voltage drop percentage: 1.77%
Voltage at the end: 72.69

http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=1.296&voltage=74&phase=dc&noofconductor=1&distance=50&distanceunit=feet&amperes=33.2&x=63&y=8

So--That 6 awg cable will give you around 1.31 volt drop (less than 2%)--With a PWM charge controller, that is OK. And 6 awg is good for 55-75 amps (conservatively speaking) * 0.8 NEC derating = 44 to 60 amps continuous

https://lugsdirect.com/WireCurrentAmpacitiesNEC-Table-301-16.htm

-Bill

dapago · October 2016

Yes sorry that was a typo error.
Good news in a way that the cabling is ok for mppt or PWM with a 2s4p wiring and 50 feet distance. IF you think the loss 1.77 is acceptable (I find It a lot but anyway) than I will not ask the installer to change the cables for a 2s4p config. I suppose that the same 6AWG cable will do the trick with a 3s3p config, well I hope because now I do not trust "nada" coming from those guys here.

The shunt looks a pretty simple handy tool. Really don't know the protocol to install that tool so maybe I leave it for the moment. But maybe the magnum inverter has an internal one.

Dave Angelini · October 2016

The magnum should know its output but it won't know the solar battery charge current without using an external shunt for the Midnite or using a magnum charge controller that is on a network and in fact does communicate as a newtork.

BB. · October 2016

A 60 amp pwm controller is much cheaper so not much money lost to try.

You can try connecting array directly to the battery bank (through fuse/breaker, watch polarity of array and battery bank, connecting backwards will ruin the array). And see if you get the expected charging current middle of hot day.

4x18=72 volts Vmp was the standard before mppt controllers.

-Bill

dapago · October 2016

Thanks Bill. Will ask the installer to do that then if I don't get the expected voltaje I'll make a point.

Generally how long takes the equalizing mode to complete and can it be done manually?
If the equalizing process is stopped because of low voltaje (noon time and hot panels) but has completed half of the necessary time, can the other half be completed the next day or does it has to start all over again until the voltaje needed is fulfilled during the necessary time at once?
i am asking that because during the morning sun start kicking the panels around 7am/8am. Obviously the panels are cooler and equalizing could be programmed or manually activated from 7am/8am until 9am/10ish when then the sun is high and hot. That way maybe I can get enough voltaje to complete equalizing witch seems to be the main problem here for me.

Photowhit · October 2016

dapago said:
- with the 6AWG installed, shouldn't it be 4 cables (2 per string)?
- if I run 3s3p (with luck), what cabling should be used?
- Will the 6AWG be enough to run a PWM with the 50 feet distance and run 4s 2 p?

This actually presents another issue; A combiner box!

The original installation of 2 strings did not need a combiner box. The reason there were only 2 wires coming down is that the strings were combined at the array.

With 3, or more, strings, there really should be a combiner box so that a short in one string isn't "over powered" by the energy from other strings. This is US NEC code. The combiner box provides for a breaker of fuse for each string.

Photowhit · October 2016

dapago said:

Generally how long takes the equalizing mode to complete and can it be done manually?
If the equalizing process is stopped because of low voltage (noon time and hot panels) but has completed half of the necessary time, can the other half be completed the next day or does it has to start all over again until the voltage needed is fulfilled during the necessary time at once?
i am asking that because during the morning sun start kicking the panels around 7am/8am. Obviously the panels are cooler and equalizing could be programmed or manually activated from 7am/8am until 9am/10ish when then the sun is high and hot. That way maybe I can get enough voltage to complete equalizing witch seems to be the main problem here for me.

I'll give you my answer for this, as I have a different view from others (though at least one manufacturer states something very close!)

First, equalizing is a controller over charging of the battery bank. The battery must be fully charged before equalizing. Having a BTS (Battery Temperature Sensor) is recommended, particularly since you live in a warm/hot climate.

There are 2 types of equalizing, a maintenance equalizing, often required for tall case batteries, that largely mixes the electrolyte and prevent stratification. This is usually done monthly and can be handled by the charge controller. FWIW - There has been some talk as to the functioning of the Midnite Classic in reference to this function not working. I use Midnite Classic's and just use the front button to equalize for this. I don't do this in reference to my checking SG, which I do monthly.

Then there is corrective equalizing, which is done when one cell's (or more) Specific Gravity(SG), is a good bit out of sync with the rest of the battery bank. Most consider more than .02 out of sync requiring this process. This require interaction of you. When equalizing check the specific Gravity each hour to see if the cell is coming up and when it reaches the other cells, or 2 hours have gone by with out change, you would stop. This can take a little or a lot of time. I think we have had a guy run2+ days getting the SG back up on some Rolls Surette batteries that had been chronically undercharged.

BB. · October 2016

As Photowhit says--Equalization happens after the battery bank is fully charged.

And equalization is performed with ~5% to 2.5% charging current (20 Hour battery rating --Say 100 AH , then 5 amp to 2.5 amp equalization current).

So even though your solar array's Vmp-hot may be below the equalization voltage, you are not expecting 100% of Imp at that point either (unless you have day time loads too... I.e., 2.5% to battery charging and 7.5% to run your day time loads).

Would 3s x 2-3p with MPPT be ideal (and what you paid for)--yes. But could you get by with 2s x 4p with PWM and possibly limiting your loads while equalizing once per month--You should be able to "live with it".

-Bill

dapago · October 2016

If I well understand you (or not my head starts fusing), maintenance equalization (less recovery one) can take several hours and will no be fully performed during the morning (8am/11am) when the batts gets the most voltaje in a 2s4p config with a PWM controller. I am very upset because even with my very limited knowledge on solar, I pointed that out with the installer/dealer when at first I was considering a PWM controller with my climate. The installers gave me his words that there will be no problem on that matter. In a hot climate, cels can reach 20 Celsius above ambient temp. There is always some winds in my location (island) but during more than 7 months a year in Granada Nicaragua (I am an European expat here), ambiente temp get over 33 Celsius during most of the day and some days reaches 38 Celsius at noon time.
I am very very far to become an expert (and will certainly never be one) but from what I understood in this forum (maybe wrongly), wIth a voltaje derating due to heat @ -0.41%/C and a 33C ambient temp + 20C, I am loosing 11.48V+ 2V in cables=13.48V. WIth 2s and a Vmp @37v per panel @ STC, that means the voltaje reaching the batts for equalization is (37v*2)-13.48V=60,52V. Trojan says equalization performe @ 2,58v per cel. I have 24 cels so that makes 61,92V. 1.40V under the required voltaje for equalization. If you guys thinks that can work then I have no issue.

Right now weather is clear since early morning, noon here, slightly windy (as often on the island) and ambient temp around 30C. With the 3s2p config they left me with, I am getting those values (see pic). And this is where I realize I don't understand a €&$@ thing about all this maths. If I divide the 115v getting in the system at noon by the 3 panels in series, I am getting 38.33V per panel!!! WTF? (sorry for my French).

Raj174 · October 2016

Hi Dapago,

The PV voltage in, varies a good bit through out the day. Sometimes higher, but more often lower than STD VMP. It is especially high when the batteries are in float. I have 300 watt panels similar to yours arranged 2 in series in 6 strings and the average PV in is between 63 and 64 volts. This is a screen shot of PV output over the course of a day ambient temp about 85 F at midday. The lower points near the center are about 61 volts, the higher 68 volts. See the spike up when float is reached.

Image: https://us.v-cdn.net/6024911/uploads/editor/t3/5r24i3op8lax.png

Hope this helps,

Rick

dapago · October 2016

Hello Rick. Yes it does help thanks a lot. Why the watt value at noon on my screen shot is so low? Today never went over 100W and it was a sunny day all day, not a cloud, float mode all day (no load today apart the fridge) and at 3.30 pm the system registered only 1.7kwh harvest.

Raj174 · October 2016

Yes, you can only harvest what you use, either charging the battery or used by the loads.

westbranch · October 2016

dapago said:

float mode all day (no load today apart the fridge) and at 3.30 pm the system registered only 1.7kwh harvest.

my system, with a fridge only harvest about the same amount in a day if the battery is full (>95%) in the morning...

If yours is in Float all day, you will have achieved the ABSORB stage, ie you have met the criteria you have set... then the CC goes into FLOAT..... Please recheck the settings for ABSORB...

ADD: oh yes, the CC will throttle back on the reported Watts coming in as that is what the CC does when in FLOAT..

dapago · October 2016

Ok. Now I understand. if when in float mode at noon the CC only shows 100 watt IN maximum when the panels are harvesting a lot more, where then goes the extra energy produced by the panels?
tomorrow will try to check the Absorb settings and see it is correspond to the Trojan manual.

Photowhit · October 2016

Solar, unlike wind or hydroelectric, does not HAVE to produce electric. So if not needed the charge controller just shuts down the energy coming from the array. One of the reasons you may like having the MPPT type charge controller is it can sense this lack of need and start loads for you, How are you heating water? likely solar thermal, but if you are using gas, you can have a tank ahead of the gas water heater and ore-heat the water during the afternoon, rather than shutting down the solar array.

Wind and Hydroelectric sources must produce energy and a diversion load would be needed. Something not needed with solar.

Understand that Off Grid Solar must be over sized, for the days when you don't have sun. Roughly an off grid solar system needs to be 3 times as large as a grid tied system to produce similar used energy!

It seems odd to me as well, that the voltage is higher than the VMP, but there are several things than could be in effect here. First panels are sold by wattage. Today, typically, cells are pre-tested and matched to be similar. So your panel may be made up of cells that are higher voltage. Panels will often have this information on the back showing higher VMP/ lower amperage.

It may also be that the panels represent a near short circuit since there is a very minimal load while in float. Even at high temperatures the open circuit voltage will be higher than the VMP.

dapago · October 2016

Thanks photowhit.
I looked behind the panels and it says Vmp 37V, VOC 45.9V. I remembered when they first connected the 4s2p the reading was 159V, same weather condition, just a bit less sunny. SO it seems those panels produce more Vmp even if the information on the back says 37V. In my case this is good news in a way as I can run 2s and no trouble me with low voltaje issue for getting 100%SOC in 1 day and equalizing so much. Yesterday the CC did not totally shut down the array as a small amount of watt was introduced into the batts all day (<100W). Maybe those watts correspond to my load at the time.

My place does not have hot water. The ambient temp of year long is hot and I dont need it. The place is meant to be a small eco lodge (4tree house)/bar/restaurant and the idea here is NEVER use the gen (I don't have one anyway) to complete batts charge as it make no sens to me to advertise as ecolodge and run a gen. All was calculated for that and of course my budget. My only concern are the bedrooms fans during the night (4 in total) that I will add in the future specially when guests forget to turn them off. I just need them on 5 hours a day (from 9pm to 2am) and will rise my load needs to 3.5kwh a day to around 4.5.
As suggested Westbranch, I will get into the CC settings (at least try and no make mistake here) and check the absorb settings and come back here later on.

good day to all of you.

310w panel Vmp issue

Comments

Categories