Configuration recommendation

[rmak]

Would like some input on a couple of configurations. Right now I am running 2 strings of 3 for total 6 panels into a Classic 150 to charge 740Ah 24v bank. Been reading past post and see that having over 130% of bank voltage from the PV is good for the charger efficiency. My Classic shows the PV input voltage >105v with my current setup.

I am off grid in Colorado so I do have a good amount of sunny days and would like to add some sort of dump load possibly for a water heater in the future. So I am thinking of adding to my current system and can go two ways. Add 2 panels and run 4 strings of 2 for 8 total. Add 3 panels and run 3 strings of 3 for 9 total panels.

Both configurations check out OK by the string tool but which would be more efficient?

Does running the higher voltage and wattage cause higher heat in the controller reducing its life?

Thanks
Ray

[tecnodave]

Ray,

I have done just what you are describing. I have 2 Classic 150's and two banks of panels. At one time I had the array further away from the controllers wired as 4 strings of three panels each for a VMP of 105 volts and the other controllers array was wired 2serial, 6 parallel, for a VMP of 70 volts. All panels were Sharp NE-165's. Everything else was identical. System is 24 volts. The Classic that had the higher voltage ran noticeably hotter and the fans run constantly. I changed the first array to the 2S 6P wiring configuration and the temperature dropped and fans did not run constantly.

So absolutely too high a input voltage absolutely makes the Classic work harder and get hotter.

After several years of the 2 Classic / 2Kid setup I verify that too much difference between battery voltage and PV input voltage will make the Classic /Kid run hotter, much more than having more power.

Right now I am running a test setup with two very early beta Kid controllers and am getting the same result:     Two Kid's , two array's , each 4 ea. Sanyo 96 cell,200 watt monocrystalline/amorphous polycrystalline multilayer panels, one system all parallel at 56.5 VMP, the other 2S 2P wiring 113 VMP.

Same result, the Kid with the excessive input voltage runs quite a bit hotter, especially when I switch in more panels to compensate for cloudy weather. I run 6 200 watt panels into each Kid , this is a serious overloading of the kid as it is rated just over 800 watts at 24 volts , I am feeding it with 1200 watts. More extreme heating resulted especially when going to absorb as the input voltage will rise when the controller does not need all the power.

I will be uploading data including temps, power levels etc very soon now.

My advise on this is to use the 2 panels in series as your Classic will run cooler and probably last a lot longer.

I will be buying more breakers soon so I will have no input voltages over 70 volts for my 24 volt system. 105 volts was too high and my Kid test is proving what I allready knew.   

Hope that you will find this info useful.

David

[rmak]

Thanks for the real world experience David.

Probably a dumb question. Is the excess power the Classic is not using turned into heat and dissipated via heat sink and cooling fan? Can that heat energy be diverted instead of going through the battery and/or through the inverter? Going directly to the load instead of through the battery. I've read about the direct from panel to load but that has to be a stand alone system if I understand it correctly.

Ray

[tecnodave]

Ray,

Yes you can use diversion from the solar input to use some of that power. There are several threads that describe just that. That diversion power will be varying voltage and modulated by the diversion device. The Kid has a PWM diversion controller built in but the Classic  needs a few external components to do this, including a SSR, solid state relay. There has been a lot of information on this posted here but the consensus is to use a Crydom SSR controlled by aux 1 using "waste not hi"

Best use is resistive loads, heating water, etc.

I am using diversion using a separate controller controlled by Classic Aux 1 port,  Trace C-40 controller set to diversion mode (modified)


David

[rmak]

David
Sorry for the noob questions but I'm trying to get a better handle of all this.

The systems I read about all seem to power the heating element from the battery. Is there a way to power the load directly from the array? Like a DC SSR to divert array output to either the charge controller or the heating element? Size the heating element to the highest voltage the array can produce and the SSR would be triggered by the CC and thermostat.

I guess that is what your doing with the 2nd separate CC. Do you just set up the 2nd CC at a higher voltage output to the heating element? Charge 24v bank with one and power 48v water heating element with the 2nd. Can 2 CC be run parallel from one array?

I'm probably over complicating this. It would be great if the Classic had a 2nd high output for a heating element that can be switched to maximum voltage after bank is charged.

Ray

[tecnodave]

Ray,

Both the Classic and the Kid can do that. With the Kid  menu system is a choice to divert from battery or input.  The Classic has a PWM output mode that can drive an external SSR , I think that with the correct wiring the Classic can divert from PV input.

I have several Trace C-40 charge controllers that can be a diversion controller, also one Morningstar PWM 45 amp charge controller which can also be used for diversion. I use those for diversion from the battery, leaves me free to use any of my charge controllers without changing the diversion setup. I have two Classics, two Kids, one Morningstar MPPT 60 amp charge controller, so I keep spares on hand just in case.

Use the search function and you will find quite a few threads on the subject. Quite a few users of the Classic use a SSR controlled by aux ports on the Classis.

David

[boB]

You can certainly use excess power to heat something but it will typically have to be run at the battery voltage and triggered on and off by the Aux 1 or 2 output control using some kind of SSR.  But that's from the battery side, not the PV side.

You ~could~  somehow have the PV input go to a separate load but you won't be able to do that AND run the PV into the charge controller at the same time.  I don't know anybody that does this exact thing though.  Would require another PV switch to to the CC OR the load.  Even then, the voltage from the PV array would not necessarily be constant enough to use very well.   And then your batteries might not get charged.

boB

[rmak]

You ~could~  somehow have the PV input go to a separate load but you won't be able to do that AND run the PV into the charge controller at the same time. 

boB

Thanks, you answered my question.

[Zardiw]

  The Classic has a PWM output mode

How do you set that?

z

[Vic]

Hi z,

In the Classic manual Rev I,   page 36  begins the description of the Classic Aux functions.   Aux 1  has a PWM output that can be used run some Opportunity loads,   such as heating water during Absorb & Float,   etc:
http://www.midnitesolar.com/pdfs/Classic_manual_REV_I.pdf

FWIW,  Vic

[tecnodave]

Zardi,

I guess I said that, I had to go look for sure. To be clear the Classic is a MPPT controller. It cannot PWM output to the battery.  What I was referring to it has the ability to output a PWM signal to drive an external SSR which can be wired to divert excess battery power to external loads. That is fairly complicated to set up and get running. There are several threads on "waste not" . Go to the main menu and search "waste not"

Dvd

[Zardiw]

Ok.....Thank you.......Still struggling to get my Classic to put out the full potential of my array.......and looking back, it seemed the TriStar 45  PWM that this Classic replaced, was putting out twice what this Classic is.....fwiw.

I'm going to redo the experiments I did in the 'Why Low Power' thread and try to nail this thing down.

I talked to Todd at Midnite, but he was unable to resolve it.....seemed to be stuck on shading, angle, the fact that I built these panels myself........etc etc.

His suggestion was to make 3 panel series strings....which would drive the voltage to over 100v on each string.........HOWEVER, I read another thread, where the guy did that and caused heating issues in the Classic......so he went with 2 panel strings.....which should put PLENTY of volts into a 24 v system and the Classic SHOULD work with that....I'm not going to go 3 panel strings.......I'll go 2 panel......which I already did, and noticed about a 5% improvement in watt output over just having them in parallel.

If I can't resolve this.......you can start looking for a Classic 150 on EBay........even though I absolutely love the design, the software monitoring/controlling, the whole concept.......but if I'm only getting 50% of the power that my array is potentially putting out.......well.......those features and my feelings have to get kicked to the side of the road...........

z

[boB]

Zardi, higher input voltage or 2 panels in series for your Classic should work better for your 24V battery system.

One question you will need to answer is, at MPPT maximum power point voltage, what is the PV Vmp operating voltage and what is the Voc open circuit voltage ?  These two voltages will tell you if you are getting the correct tracking or not.

IF the PV input voltage is around 80% of Voc  (+2/-4% or so), then there is something else amiss.  You cannot know if another controller will be any better or not without knowing this without just trying it.  I would like to understand what is going on in your system too.

Also, if you are expecting 2000 watts and are only seeing 1000 watts output and the Vmp is correct, then that extra 1000 watts can NOT be wasted in the Classic.  That amount of heat would be at least 10 to 20 times the amount of heat that any controller would dissipate.

But for sure, putting two panels in series and then paralleling strings of two will get you over the low Vmp that you have been seeing.

It may very well be that a PWM controller might be better for these modules and just running in parallel but measurements need to be made to know for sure what is going on.

boB

[Zardiw]

Zardi, higher input voltage or 2 panels in series for your Classic should work better for your 24V battery system.

One question you will need to answer is, at MPPT maximum power point voltage, what is the PV Vmp operating voltage and what is the Voc open circuit voltage ?  These two voltages will tell you if you are getting the correct tracking or not.

IF the PV input voltage is around 80% of Voc  (+2/-4% or so), then there is something else amiss.  You cannot know if another controller will be any better or not without knowing this without just trying it.  I would like to understand what is going on in your system too.

Also, if you are expecting 2000 watts and are only seeing 1000 watts output and the Vmp is correct, then that extra 1000 watts can NOT be wasted in the Classic.  That amount of heat would be at least 10 to 20 times the amount of heat that any controller would dissipate.

But for sure, putting two panels in series and then paralleling strings of two will get you over the low Vmp that you have been seeing.

It may very well be that a PWM controller might be better for these modules and just running in parallel but measurements need to be made to know for sure what is going on.

boB

Thank you Bob......

I'm going to start over and do some more testing.

Currently have the mode at Solar......will see how many watts I'm getting.

At High Noon, I'm going to measure the voltage at the array......that should be when Vmp is at it's highest.

Currently Voc of the array is around 37 volts with them all in parallel.

I'm going to be doing some experiments with 2 panels only hooked up. Both in parallel, and then in series.

Need to nail this thing down.

But I have a suspicion that the Emperor has no Clothes...........at least from all the testing and experimenting I've done so far.

I may run a jumper from PV + and hook it directly to the battery bank Pos. Then measure the DC Amps coming from the array with a DC Amp meter, and compare that to the Amps while the Controller is 'controlling'......lol

Bottom line: This is not rocket science. Although people make it out to be.

Question: Has anybody actually added up the watts listed on their panels, and compared that to the watts being shown on the controller?......Be interesting to get some feedback on that. One would think it would be APPROXIMATELY close to the same...........i.e. I have 9 panels putting out approximately 200 watts each in high sun.....SO, I would think that the controller should also be putting out ABOUT 1800 watts.

z

[boB]

I may run a jumper from PV + and hook it directly to the battery bank Pos. Then measure the DC Amps coming from the array with a DC Amp meter, and compare that to the Amps while the Controller is 'controlling'......lol

Bottom line: This is not rocket science. Although people make it out to be.

Question: Has anybody actually added up the watts listed on their panels, and compared that to the watts being shown on the controller?......Be interesting to get some feedback on that. One would think it would be APPROXIMATELY close to the same...........i.e. I have 9 panels putting out approximately 200 watts each in high sun.....SO, I would think that the controller should also be putting out ABOUT 1800 watts.

z

Not rocket science BUT there are scientific measurements to make like the Vmp/Voc calculation and sanity check that I mentioned before.

Good deal on connecting directly to battery.  I have a feeling that you may be close to the Vmp by doing that but that will depend on how full or empty your batteries are because that voltage is going to determine if you are near the Vmp or not.

As for what people typically get out of their systems...  It varies.  Sometimes you can get the power out that the nameplate rating is and once in a great while, during cloud edge effects, you can get more power than nameplate for maybe a minute until the cloud moves.

More typically though, I see/hear about 80% of the PV nameplate power rating being harvested at full sun.

BTW, the Vmp/Voc ratio of around 80%  is only if there is NO partial array shading.  Not shadows on the array.

Does your PV modules have bypass diodes interspersed in a few places ?  Those will make some difference in partial shading harvest but with such a low Voc, that probably won't help unless the PVs are wired in series.  Partial shading and bypass diodes are one good reason for wiring PV modules in series.

boB