Hello,
I am dimensioning a 2 kW system with the following specs:
8 panels in 2 strings of 4.
Stringvoltage is 173 Volt @ 5,4A, I calculate with safety factor 1,25 = 216,5 VDC
I need to cover 250 ft before reaching the Charge controller.
My batterybank is 24V
I have a 3500VA Outback FX Inverter
My question is the following: the classic 250 has a Max Current Out limit of 62A@24V. Does this mean that I will never be able to utilize the full Wattage of my array? (2kW) Since 62A x 24V = 1488 Watt ?
Why does the classic 150 not have this limit but a much higher 94A@24V?
Am I making a mistake here? If somebody could help me, that would be greatly appreciated.
Brecht
Brecht,
You are correct. The reason the higher voltage CC has a lower output rating is that the controller is
less efficient. The components have higher resistance because of their ratings, unfortunately.
However, the system efficiency that one might gain by running the input at a higher voltage
can overcome the cost you would need to expend for the same efficiency running at a lower
input voltage by using larger wire.... OR, the system efficiency will be higher because the losses
of the long PV wire will be less because of the lower input current.
Brecht,
With a string voltage of 173v and string of 4 panels that means each panel has a voltage of 43.25v
Is this the Open circuit voltage? and if so what is the Maximum power voltage? about 35v to 36v?
dgd
Brecht,
This has to do with the voltage conversion ratio. You are proposing about conversion ratio of about
6 to 1. The Classic 150 is operating at about 2.5 to 1 to 3.5 to 1input volts to output volts. The MPPT converter is more efficient at the lower conversion ratios so the classic 250 are rated at a lower current because of the extra heat generated by the lower efficiency.
On the power handling of the classics you are using the wrong voltage, a 24 volt battery on charge is 28.8 to 29.6 on charge. This is the correct number to use for that calculation. 28.8 X 62 = 1785.6. Right where Classic charts are 1786. No worries your C250 will handle that array (1488 watt) with ease.
td
Quote from: elbrechtel on May 29, 2014, 09:43:41 PM
My question is the following: the classic 250 has a Max Current Out limit of 62A@24V. Does this mean that I will never be able to utilize the full Wattage of my array? (2kW) Since 62A x 24V = 1488 Watt ?
Why does the classic 150 not have this limit but a much higher 94A@24V?
Welcome to the forum.
The reason is efficiency and heat. The more down conversion of voltage the controller has to perform, the less efficient it is. When power is lost (due to inefficiency) it creates heat, and heat is what sets the limits.
We are not talking about major inefficiencies... if 1500 watts is passing through the controller, even a 1% loss of efficiency is 15 watts. 15 watts is a lot... put a 15 watt bulb in a metal box the size of the classic and feel the heat.
The classic 150 cannot always handle 94 amps... that assumes an input voltage of 90 volts. With a 120 volt input the classic 150 can only handle 82 amps.
The classic 250 limit drops down from 62 to 60 amps when the input voltage rises from 180 volts to 200 volts.
What I will now tell you about is the wonders of higher voltage...
If you have a 48 volt system your controller will have less down conversion of voltage to perform. Your controller can handle 55 amps with an input voltage of 180 volts. 55 amps X 48 volts = 2640 watts.
The classic 150 can handle over 5000 watts if its input is 70 volts on a 48 volt battery.
I presume that your panels are 250 watt with Imp = 5.4. That means Vmp= 46.3.
If you get another panel and put your panels in three strings of three panels, a Classic 200 could handle over 2000 watts. Ideally you would want #6 copper cable to go the 250 ft, but even #8 would be OK.
Finally, to wrap up... We have been discussing the LIMITS. In my opinion it is not a good idea to be pushing the limits continuously. Good conservative design does not do that.
--vtMaps
edit: wow... three other posts came in while I wrote this :)
thanks for your quick reply`s.
The guy i am helping just gave me the exact specs for the panels, they are slightly different from the above.
Pmax<W> 280.0
Vmp<V> 35.6
Imp<A> 7.76
Voc<V> 44.1
Isc<A> 8.31
Max Series Fuse Rating<A> 15.0
If I make 4 strings of 2 panels and do the 250 ft run like that (88.2 V), connect the panels to a classic 150, what size cable do you recommend?
Is it more economical to buy two Classic 250? And connect 1 string of 4 panels to each? Or to do the run on lower (88 V) voltage?
Making a 48V battery bank is not an option, I am stuck with the 24V.
thanks,
Brecht
Quote from: elbrechtel on May 30, 2014, 09:12:57 AM
Pmax<W> 280.0
Vmp<V> 35.6
Imp<A> 7.76
Voc<V> 44.1
Isc<A> 8.31
Max Series Fuse Rating<A> 15.0
If I make 4 strings of 2 panels and do the 250 ft run like that (88.2 V), connect the panels to a classic 150, what size cable do you recommend?
First of all, when you are trying to figure out the maximum voltage of your string you use Voc and temp coefficient. When you are trying to figure cable size you use Vmp. Thus I see a string voltage of 71.2, not 88.2.
Are you sitting down? You will need to use #0 awg copper cable to have a 2% voltage drop.
If you use #4 awg you will have 5.5% voltage drop, representing an actual voltage drop of 3.9 volts. That voltage drop is only at full nameplate power of 2210 watts... rarely achieved, and then only briefly. The power lost in the cable would be 122 watts at full nameplate power. NOTE: the power lost in the cable goes as the square of the power produced. Thus, at half power the loss would be only 30.5 watts.
In my opinion, a little loss of efficiency in the cable is not such a bad thing... If you're so worried about it, it might be more cost effective to just add another panel than to upgrade the cable from #4 to #0. Many folks get into the trap of raising their string voltage to reduce cable loss. What happens is that they shift the power loss from their cable to their controller. Better to heat up the cable than the controller.
--vtMaps
Good point of keeping the losses more limited to your cable and reduce the stress on the controller.
I will go for a #4AWG cable and think about a #2AWG perhaps. Is read the classic goes up to 4AWG for the connection. How would you reduce the size to 4AWG? Is there a clean looking method?
Brecht
And thank you for pointing out that cable size is calculated with Vmp. it makes much sense, since this is when current flows.
Brecht
Quote from: elbrechtel on May 30, 2014, 10:25:38 AM
Good point of keeping the losses more limited to your cable and reduce the stress on the controller.
I will go for a #4AWG cable and think about a #2AWG perhaps. Is read the classic goes up to 4AWG for the connection. How would you reduce the size to 4AWG? Is there a clean looking method?
Your combiner to controller cable does not connect directly to the Classic. It has to go through a circuit breaker. You can use #4 from the breaker to the controller. A few inches of #4 won't add any significant resistance.
On the PV negative side, a split bolt connector can get you from #2 to #4. You could also land the PV cable on the negative battery bus and run #4 from that bus to the classic. (batt neg and PV neg are interconnected in the classic).
--vtMaps
Thank you so much! Really appreciate your help. Couple more questions:
If i would use 2 x 250 Classic with 4 panels in series on each (1120 Watt). Can I run just one cable (8 / 6 AWG) and split it at the end? How do I connect 2 classics in parallel? Is it possible for them to communicate to each other?
Still making the trade off..
Brecht
Quote from: elbrechtel on May 30, 2014, 12:32:23 PM
If i would use 2 x 250 Classic with 4 panels in series on each (1120 Watt). Can I run just one cable (8 / 6 AWG) and split it at the end? How do I connect 2 classics in parallel?
No, each classic must have direct access to its own array. No trouble sharing one battery, though.
Quote from: elbrechtel on May 30, 2014, 12:32:23 PM
Is it possible for them to communicate to each other?
Classics have industry leading communications. Read about "follow me".
--vtMaps
I read in another thread that full parallel operation is coming to the Classic's soon.
td
Quote from: tecnodave on May 30, 2014, 02:24:55 PM
I read in another thread that full parallel operation is coming to the Classic's soon.
td
I seem to remember that myself but for The Kid. It would be a cool feature for either.
Tom
that will be cool but would it also work for classics of different voltages and well as different pv voltages?
Hi all,
I struggle to find a good calculator to determine voltage drops with, what is your favorite one?
B
Quote from: elbrechtel on May 30, 2014, 10:10:56 PM
Hi all,
I struggle to find a good calculator to determine voltage drops with, what is your favorite one?
B
I really like this one: http://www.calculator.net/voltage-drop-calculator.html (http://www.calculator.net/voltage-drop-calculator.html)
Thank you! that helps a lot!
Hi,
Can I hang 2 midnite controllers in parallel on 1 string? Or do i need to run two cables to each array separately? ( 4 in total)
thanks!
Brecht
Quote from: elbrechtel on June 02, 2014, 12:51:29 PM
Can I hang 2 midnite controllers in parallel on 1 string? Or do i need to run two cables to each array separately? ( 4 in total)
No. see reply #12 in this thread... I think you're asking the same question... if not please clarify. --vtMaps
Excuse me, i thought i had asked this question on another forum.
My apologies, haven't slept a lot!
B
Mahendra,
Quote from: mahendra on May 30, 2014, 05:16:58 PM
that will be cool but would it also work for classics of different voltages and well as different pv voltages?
My understanding of the full parallel operation as implemented in the kid is inputs and outputs are tied in parallel and that the MPPT functions would have to be sync'ed as well so, no this kind of paralleling will not work with Classic's with different inputs. Full paralleling of the Classic's would be inputs and outputs connected together to handle larger loads. Paralleling only the outputs with separate inputs as implemented in follow me mode only shares charge state and set point, not the MPPT loading of each input.
td
Right now, we cannot parallel inputs of more than one Classic.
Only the outputs of different Classics can connect to the same battery.
However, in some wind-turbine installations, there have been successful input paralleling
of multiple Classics when each Classic has their own 3-phase rectifier coming from
the same turbine. Different thing than solar though but thought I would mention it.
Quote from: boB on June 03, 2014, 02:58:12 PM
Right now, we cannot parallel inputs of more than one Classic.
Only the outputs of different Classics can connect to the same battery.
However, in some wind-turbine installations, there have been successful input paralleling
of multiple Classics when each Classic has their own 3-phase rectifier coming from
the same turbine. Different thing than solar though but thought I would mention it.
Now thats an interesting idea.. ;)
So the only separation between the two Classics in a wind application is the two three phase rectifiers which are no more than diodes. So I presume one function these diodes serve is to isolate the inputs of the two Classics from each other?
So would the same not work with two Classics connected to the same PV array? simpoly by using a decent size diode to isolate the +ve input to each Classic?
dgd
Quote from: dgd on June 03, 2014, 05:27:04 PM
Quote from: boB on June 03, 2014, 02:58:12 PM
Right now, we cannot parallel inputs of more than one Classic.
Only the outputs of different Classics can connect to the same battery.
However, in some wind-turbine installations, there have been successful input paralleling
of multiple Classics when each Classic has their own 3-phase rectifier coming from
the same turbine. Different thing than solar though but thought I would mention it.
Now thats an interesting idea.. ;)
So the only separation between the two Classics in a wind application is the two three phase rectifiers which are no more than diodes. So I presume one function these diodes serve is to isolate the inputs of the two Classics from each other?
So would the same not work with two Classics connected to the same PV array? simpoly by using a decent size diode to isolate the +ve input to each Classic?
dgd
Sounds like an experiment for our resident Classic destroyer, Ryan! See if he can make one or both release the mystical smoke.
Seems a single bridge be plenty of isolation on a DC source?
Tom
Quote from: TomW on June 03, 2014, 05:48:09 PM
Quote from: dgd on June 03, 2014, 05:27:04 PM
So would the same not work with two Classics connected to the same PV array? simpoly by using a decent size diode to isolate the +ve input to each Classic?
Sounds like an experiment for our resident Classic destroyer, Ryan! See if he can make one or both release the mystical smoke.
Seems a single bridge be plenty of isolation on a DC source?
Just thinking about this and does the same apply with DC turbine/Clipper that two Classics can be connected to the Clipper to divide the load between Classics?
The input to the second Classic would need to have another diode (or second rectifier same as Clipper currently uses)
connected to inputs to Clipper's rectifier?
I suspect it would be necessary to use an identical 3ph bridge rectifier where the +ve from turbine goes to commoned 3ph inputs and only +ve output from retifier is used and connects to Classic, hence using 3 diodes of the bridge.
This would mean balanced DC outputs to Classics. IMHO.
The part I don't get is how to ensure the Classics get a near equal proportion of the input power. Perhaps this is just relying on the input impedence of similar Classics being quite closely the same.
Maybe it would not work with different models of Classic or a Classic/KID combo (but would be real nice if it did) 8)
I like the Ryan idea ;D Will he try it?
dgd
Quote from: dgd on June 03, 2014, 05:27:04 PM
Quote from: boB on June 03, 2014, 02:58:12 PM
Right now, we cannot parallel inputs of more than one Classic.
Only the outputs of different Classics can connect to the same battery.
However, in some wind-turbine installations, there have been successful input paralleling
of multiple Classics when each Classic has their own 3-phase rectifier coming from
the same turbine. Different thing than solar though but thought I would mention it.
Now thats an interesting idea.. ;)
So the only separation between the two Classics in a wind application is the two three phase rectifiers which are no more than diodes. So I presume one function these diodes serve is to isolate the inputs of the two Classics from each other?
So would the same not work with two Classics connected to the same PV array? simpoly by using a decent size diode to isolate the +ve input to each Classic?
dgd
I think that the difference here is that in wind mode the Classic is working with values programmed into the wind curve for that wind generator, it knows the rpm,s from the pulsing on the generator signal ans sets up the correct MPPT from that data where in solar mode the controller sweeps the voltage from Vmax to some value maybe 50 % of Vmax to find the Maximum power point. The Classic logic would need to be linked somehow to cause the sweeps to be in tandem or better yet only 1 Classic does a sweep then tells the other what set points to use. In any case it is not trivial. A huge amount of programming will be needed to accomplish this.
td
Quote from: tecnodave on June 04, 2014, 01:20:57 AM
I think that the difference here is that in wind mode the Classic is working with values programmed into the wind curve for that wind generator, it knows the rpm,s from the pulsing on the generator signal ans sets up the correct MPPT from that data where in solar mode the controller sweeps the voltage from Vmax to some value maybe 50 % of Vmax to find the Maximum power point. The Classic logic would need to be linked somehow to cause the sweeps to be in tandem or better yet only 1 Classic does a sweep then tells the other what set points to use. In any case it is not trivial. A huge amount of programming will be needed to accomplish this.
I am not actually sure this would be case or a requirement to have the MPPT scanning somehow synchronised.
If, for example, a 6Kw PV array was connected to two Classics via protection/isolation diodes then ideally you would want 3Kw power through each as a maximum. It would not really matter if below a full power input the division of power between the Classics was different. You just do not want all 6Kw trying to squeeze through one Classic.
There is already a Classic setup parameter that specifies the maximum input and output currents that the Classic can process. So just specify 3kw worth of amps in each Classic and let them sort out individually what they want to do.
I can't see any reason this would not work. The Classics cannot 'see' each other so should just see a varying PV input as if just connected to its own array.
dgd
Quote from: dgd on June 04, 2014, 03:37:48 AM
I am not actually sure this would be case or a requirement to have the MPPT scanning somehow
I can't see any reason this would not work. The Classics cannot 'see' each other so should just see a varying PV input as if just connected to its own array.
dgd
I think that you are not looking at the issue, When one Classic does an MPPT sweep it will pull down the PV input voltage at least to some degree , at that point it will cause the other one to sweep and I would think that you would have an unstable loop in which the classics would fight for control and result in an unstable oscillation. In any case there are only two companies who are doing true paralleling of inputs and outputs. It is difficult to achieve.
Quote from: tecnodave on June 04, 2014, 12:01:24 PM
I think that you are not looking at the issue, When one Classic does an MPPT sweep it will pull down the PV input voltage at least to some degree , at that point it will cause the other one to sweep and I would think that you would have an unstable loop in which the classics would fight for control and result in an unstable oscillation. In any case there are only two companies who are doing true paralleling of inputs and outputs. It is difficult to achieve.
Ok,maybe you are correct. We don't really know what would happen until it's tried and with configurable current limits, mppt sweep depths and intervals there may be possibility they could work together..
Dgd
dgd,
Ok I did mess around in trying to parallel several Chinese Tracer MPPT controllers, it did not work!, they could not set a stable input, even with dropping resistors. I tied my oscilloscope across the drop resistors and when one would sweep the others would be affected. Instability resulted. Gave up on that approach and was using 3 Tracers on one battery bank with separate arrays until I could budget the Classic. Lessons learnt! Classic paid for itself by being able to top the batteries without running the Onan on dinosaurs.
td
td,
I like experimentation and its good to read you tried parallel Tracers together. My limited knowledge of these would lead to doubts that Tracer experiments would be a good yardstick for Classic behaviour (unless boB unwittingly wrote the mppt scanning code for the Tracers).
However, the principle may be valid that having more than one PV mppting device on one PV array just leads to confusion unless they can somehow co-operate (KID bully mode etc..)
If I had a large enough PV array and a couple of Classics I would certainly give it a go.
dgd
dgd,
I am always experimenting in one form or another. A whole lot of reading tells me that the kid does have PWM control of the load output. So when I get him back from babysitting I am going to try about 900 watts PV in and battery diverted PWM out to a 24 volt 600 watt water heater element. I was trying to do that with the Classic but you can't have PWM in and out at the same time. (WBjr in and opportunity out) but it seems that the kid does have that ability with its load port and aux port together.
td
I am not surprised that 2 or more paralleled charge controllers would not play well together without some sort
of communications. Either some separate comm cable or by power line carrier on the DC input or battery side.
I was not really expecting it to work but I tried. I had a rather long run and only one set of wires 500kcmil aluminum utility direct burial 8 kv. Wire I was using. I ended up with separate positives and common negatives on the 4 conductor 3 phase feeder cable, kinda overkill but it was free.
dgd,
I think boB would not design the tracer on his worst day. I have never seen any MPPT controller that sweeps as slow as the tracer. Even the Well See does better! Looking at epSolar's lineup it is obvious that they are coping Morningstar. epSolar's EDPIPDB.com is a dead copy of the Morningstar SunSaver Duo , their remote readout is so close to the Morningstar they are maybe interchangeable? They look nearly identical. and the eTracer is so similar in design, layout, and features you cannot help to think it is a clone of the TS- MPPT line, reading the manuals will have you convinced of this. Even the charts are copied!
Lesson learnt, cheep junk will do the basics but that is where you will stay until you find something that is user programmable.
td