Wire recommendation from panel

[Mike90250]

Just gotta say that this is so far fetched, that looses in the DC wires cause inverter failure.
These string inverters are designed to cover a range of DC inputs, using strings of 8 to 15 panels of various voltages. Panels vary voltage and wattage from summer to winter too, i just don't see how an additional 1% loss in a wire is going to cause an issue.
But it's not my flaky design, I built solid stuff that can take hard rads, +-20% voltage variation and last 20+ years

[ncs55]

Inverters adjust to the max power point - they do not know the difference between resistance within the modules and resistance within the wiring between the inverter and the modules. They can see the voltage/current curve of the sytem and pick the best operating point - and while the wiring affects that point the inverter must be able to handle different panels and different conditions for the panels (ex. bypass diodes being used) which also affect that curve.


There might be some truth to that - I can see some of the heat of the wire being dissipated into the inverter - so more heat in the wire would mean more heat being brought into the inverter. How much would be very very dependent on a lot of things - like wire dimensions, length, how the wire is thermally insulated and the environment it's exposed to (ex. wires in a conduit on a roof would probably gain heat from the sun, while wires in a buried conduit would dissipate e heat to the ground)[/B]

I think most of what you've shared about inverter manuf. concerns about voltage drop is on the AC side though.
And there they are concerned about getting out of range of the expected grid voltage. And since they will automatically turn off if/when the voltage is too high, that means when they're doing max generation they are more at risk of hitting that point and turning off - which the customer will be unhappy with.

I see that you obviously know more about the algorithms than the manufacturer and what they instruct me on. I am only imparting my experiences and what I am taught. I will not argue this with you. And the manufacturers are concerned about the DC side voltage drop and the parameters in the software are not as broad range as you are speculating. I am not saying that heat is traveling through the wire to the inverter because it would dissipate in the DC disconnect and not make it to the internal components of the inverter. you need to understand exactly what is going on inside of the components, boards, power supplies etc. The internal heat generated is not an instant high heat, it is a constant higher than component tested value and that is where the boards fail. It is the same principal with computers and why some of them are liquid cooled. Long term higher heat exposure will kill circuit boards and or the components on those boards. Pay the price to get trained by the manufacturer to repair these inverters and you will be taught the same thing as I have been taught.

#31.4

foo1bar commented
Today, 11:33 AM
Editing a comment

Ok - 50kW system producing 50kW most of the day because it's using trackers (not what you would actually see for someone's offgrid system or even any residential system - but sure let's use the numbers you supplied)

0.56kW extra lost to voltage drop (comparing the two possible configurations.)

Let's say a day is 8 hours and 365 days in the year - so 2920 hours per year

50kW * 2920 = 146000kwh
0.56kW * 2920 = 1635kwh

In 25 years the total losses are 0.56kW * 2920 * 25 = 40880 kwh

And 40880kwh is 28% of one year's production (146MWh)

BTW you can use whatever number of hours per year you want - it cancels out and comes out that still only 28% of a year's worth of production is lost over that 25 year span.

So in short - 1% of production loss is *AT MOST* ~25% of a year's production when looking at a 25 year span.
2% of production lost is *AT MOST* 50% of a year's production when looking at a 25 year span.
And most likely it will be significantly less because most people don't use trackers and most of their power generation is not at that max current condition. (Even with trackers, you're not going to be at that spot most of the year - but you will certainly be a lot closer to it for a much larger amount of time)

Thank you for doing the math proving the production value and showing why the production losses are cheaper to address in the beginning. 28% is more than enough to pay for one wire size larger. I was not referencing off grid and I see plenty of residential trackers. That was an example used from an article I found on voltage drop as a reference not specific to what I initially said. i am just trying to show the value of a 1% drop in planning. By the way with trackers you are at that spot most of the day because the modules always run cooler and are always facing optimal direction all day long therefore optimizing and extending the solar window. Agreed about roof and fixed mount systems.
Another thing I consider is that the modules do degrade with time and will harvest less as they get older. In that scenario where you came up with 28% of a years production gained by changing the voltage drop to a lesser value. The extra production gained by less voltage drop also helps to offset the degradation of the modules in the long run.

[DaveDE2]

There might be some truth to that - I can see some of the heat of the wire being dissipated into the inverter - so more heat in the wire would mean more heat being brought into the inverter. How much would be very very dependent on a lot of things - like wire dimensions, length, how the wire is thermally insulated and the environment it's exposed to (ex. wires in a conduit on a roof would probably gain heat from the sun, while wires in a buried conduit would dissipate e heat to the ground)[/B]

You have to admit though that wires conducting any signifcant amount of heat into an inverter is pretty far fetched. And if the heat were caused by the sun, bigger wires would conduct more heat in.

I think most of what you've shared about inverter manuf. concerns about voltage drop is on the AC side though.
And there they are concerned about getting out of range of the expected grid voltage. And since they will automatically turn off if/when the voltage is too high, that means when they're doing max generation they are more at risk of hitting that point and turning off - which the customer will be unhappy with.

Agree. My SunnyBoy has output voltage range of 211v to 264v which is about 240v +/-10%, well above NEC voltage drop (3%) and expected line voltage variations. It will happily operate anywhere within that range, and it will operate at a voltage just high enough to overcome the voltage drop between it and the grid. In short, the inverter doesn't "know or care" if there is any voltage drop or whether its operating voltage is a result of higher wire resistance or the grid voltage itself. No extra heat is generated in the inverter if it has to operate at a slightly higher output voltage to overcome wire resistance. That heat is in the wire, usually a relatively long run buried and dissipated into the ground.

If someone wants reduce heat in their inverter to prolong life, it's better to be concerned about where it's mounted. Inside of a well insulated cool garage might be a good choice.

[foo1bar]

However, you need to understand that inverters are very sensitive to voltage drop and their algorithms sense the voltage drop and adjust the performance partly based on voltage drop.

Inverters adjust to the max power point - they do not know the difference between resistance within the modules and resistance within the wiring between the inverter and the modules. They can see the voltage/current curve of the sytem and pick the best operating point - and while the wiring affects that point the inverter must be able to handle different panels and different conditions for the panels (ex. bypass diodes being used) which also affect that curve.

Too much drop will cause the internal components to run hotter and the inverter will produce less. when the internals run hotter than optimal it is obvious that in a hotter environment internals will fail sooner and shorten the life of the equipment.

There might be some truth to that - I can see some of the heat of the wire being dissipated into the inverter - so more heat in the wire would mean more heat being brought into the inverter. How much would be very very dependent on a lot of things - like wire dimensions, length, how the wire is thermally insulated and the environment it's exposed to (ex. wires in a conduit on a roof would probably gain heat from the sun, while wires in a buried conduit would dissipate e heat to the ground)

I think most of what you've shared about inverter manuf. concerns about voltage drop is on the AC side though.
And there they are concerned about getting out of range of the expected grid voltage. And since they will automatically turn off if/when the voltage is too high, that means when they're doing max generation they are more at risk of hitting that point and turning off - which the customer will be unhappy with.

[foo1bar]

Ok - 50kW system producing 50kW most of the day because it's using trackers (not what you would actually see for someone's offgrid system or even any residential system - but sure let's use the numbers you supplied)

0.56kW extra lost to voltage drop (comparing the two possible configurations.)

Let's say a day is 8 hours and 365 days in the year - so 2920 hours per year

50kW * 2920 = 146000kwh
0.56kW * 2920 = 1635kwh

In 25 years the total losses are 0.56kW * 2920 * 25 = 40880 kwh

And 40880kwh is 28% of one year's production (146MWh)

BTW you can use whatever number of hours per year you want - it cancels out and comes out that still only 28% of a year's worth of production is lost over that 25 year span.

So in short - 1% of production loss is *AT MOST* ~25% of a year's production when looking at a 25 year span.
2% of production lost is *AT MOST* 50% of a year's production when looking at a 25 year span.
And most likely it will be significantly less because most people don't use trackers and most of their power generation is not at that max current condition. (Even with trackers, you're not going to be at that spot most of the year - but you will certainly be a lot closer to it for a much larger amount of time)

[ncs55]

foo1bar, consider this. Assume a 50 kW PV system at 400 Vdc with 1,000 feet from combiner box to inverter. The minimum dc output-circuit conductor is 4/0 for ampacity, resulting in 3.8% voltage drop, or 1,900 watts of lost power at full-power conditions. Upsizing to 300-kcmil conductors would result in 2.68% voltage drop, or 1,340 watts of lost power, which is a net gain of 560 watts under full power. full power can be all day long if the modules are mounted on trackers.

[ncs55]

DaveD2, First of all 0.5% VD is excellent and your reasoning of not wasting power is spot on my friend. However, you need to understand that inverters are very sensitive to voltage drop and their algorithms sense the voltage drop and adjust the performance partly based on voltage drop.
en-phase says this about voltage drop: The total VRise in the AC wiring should be less than 2% in all wire sections from the point of common coupling (PCC) to the last microinverter on each branch or sub-branch circuit as described in Voltage Rise by Wire Section on page 3. A good practice is to maintain less than 1% VRise in the Engage Cable. https://enphase.com/sites/default/fi...Vdrop_M250.pdf Please read the whole article as it goes into more depth.
SMA has this note at the bottom of their voltage drop calculator. IMPORTANT: SMA America, Inc., recommends a maximum of 1% to 1.5% voltage drop from the Main Service panel to the inverter. This calculator, like the String Sizing Program, is intended to provide an estimate of system performance, and should not be used to guarantee or imply any guarantee about system performance. The code that forms the basis of this Calculator was used with permission from the Computer Support Network.
I read a tech brief from Schneider that talked specifically about voltage drop and how the inverter senses and adjusts to the rise and fall of voltage drop throughout the day. there was a reference in that paper about voltage drop and what happens inside of the inverter when the voltage drop is out of the range of the inverter. Too much drop will cause the internal components to run hotter and the inverter will produce less. when the internals run hotter than optimal it is obvious that in a hotter environment internals will fail sooner and shorten the life of the equipment. I cannot find that paper right now as I read it when going through training for repair of these units. Considering that I repair utility scale inverters and the first question they ask is about the system size and module stringing, the second question they ask is concerning system grounding and the third question is what are the wire sizes on both AC and DC sides. I have asked them about why they want to know about the wire size and the techs will always refer to voltage drop and internal heat as a way to determine what caused the failure and eliminate the common mistakes concerning voltage drop. Ask them further to explain and they will always tell you that there is a voltage drop range and it will cause excessive heat. common sense tells anyone that electronic components fail quicker when exposed to a higher heat than they are designed for. So, again voltage drop will hurt an inverter and eventually kill it. I do not care if you believe this or not because it is true and I have had to change out DC wiring, under the manufacturer recommendation that met the NEC 2% ruling in order to permanently fix inverter failure. You need to understand the internal workings of the inverter to understand why this is so important for the life of the product. You will not see this in an installation manual because as a system designer or installer it is assumed that you already have this basic knowledge. Cheers.

[foo1bar]

1% vs 3% voltage drop (at max current) will not result in even close to a full year's worth over 25 years.
At most it would be 2% * 25 = 50% of a year.
But since you would not be at max current the vast majority of the time the overall impact would be even less.
(ex. if it's 5kW system, when the system is doing only 500W of power instead of 5000W you would have ~0.1% voltage drop instead of ~0.3% voltage drop instead of 1% vs 3%. It wouldn't be linear - but it's close enough that you can see how you're going to be significantly less than a year's worth of production)

[DaveDE2]

@ncs55
Don't get me wrong, minimizing voltage drop is probably the least expensive thing an installation can have and I fully concur with that. My new systemwill have <0.5% DC voltage drop. It uses two of the new Sunnyboy 6.0kW inverters. I just received them today and nowhere in the manual does it say anything about recommended voltage drop. It says follow NEC rules, and then goes on to say any wire from awg14 to awg8 is allowable. It makes sense however that any inverter manufacturer would recommend low voltage drop, afterall they have gone to great lengths to maximize efficiency so needlessly throwing away more power would be a shame.
Having said that, voltage drop will not hurt an inverter. In fact, one of the jobs of the inverter is to handle voltage drop. Think about what happens when the sun first comes up or when one of the panels is partially shaded and the bypass diodes kick in. String voltages can and will vary widely, and voltage drop in the wires isn't even consuming the lion's share. This is not a problem for most inverters, especially the newer ones, as the SB6.0 can track a 100v to 550v input range.
I'm not an inverter repairman but I've been an electrical engineer for over 30 years and have designed hundreds of circuits, circuit boards, power supplies, etc.

Believe what you will, but I happen to seriously doubt your results have anything to do with voltage drop. Cheers.

[ncs55]

That document was what prompted me to ask their tech about voltage drop years ago. It does not affect their warranty that I know of and I have never had them try and get out of a warranty. Solid company with an excellent product IMO. If you add up the difference in 1% vs 2-3% over the life of the system say 25 years, the harvest gained by 1% or less is about an extra year of production or more depending on the system size and more than pays for itself. Fixing a premature inverter failure out of warranty will cost the consumer more than it would cost to simply plan for the 1%. The cost of wire is a small percent in a system design.

[SunEagle]

Nonsense.

A 2-3% voltage drop results in power/voltage loss in the wires from the panels to the inverter and nothing more. It in no way affects the inverter other than producing slightly less voltage into it.

Wire sizing above and beyond NEC requirements is a tradeoff for the end user, going bigger saves a little power, going smaller wastes a little power.

Actually I have to agree with ncs55 concerning SMA. I did find an early document for a Sunny Boy SB 700U showing a chart for wire size and distances for voltage drops. There was a note that SMA recommends a max voltage drop of 1.5% for the wire used. I did not find anywhere in this manual what would happen if the 1.5% VD was exceeded or if SMA would void any warranty if that happened.

[ncs55]

I am sorry but you are incorrect, and it is not nonsense because you happen to disagree with my results, there is more happening than power/voltage loss going on inside of the inverters when they are experiencing voltage drop. Inverters are complex on the inside and certain voltage drop levels do cause heat and that takes out fets and other internal components. Do you repair inverters, and solar systems and communicate with the manufacturers regularly? I do, and I have seen this many times and verified it with the inverter manufacturers. I first learned this from repairing small commercial inverters 30-100k in size. It is the same in the residential units. If it were not true then please explain how the units that I have repaired / rewired, perform better, have not had the same issue and have run longer than before the fix? I monitor these systems after repair and have been in the field long enough to see the results. I am no inverter expert but I know what I see and what I am trained to do. And see the results for years after the repairs.

[DaveDE2]

A 2-3% voltage drop will cause more heat inside of the inverter and is one of the biggest mistakes that cause premature inverter failure. It is more expensive to have to replace an inverter than to simply follow a 1% or less rule.

Nonsense.

A 2-3% voltage drop results in power/voltage loss in the wires from the panels to the inverter and nothing more. It in no way affects the inverter other than producing slightly less voltage into it.

Wire sizing above and beyond NEC requirements is a tradeoff for the end user, going bigger saves a little power, going smaller wastes a little power.

[J.P.M.]

I bet there is also a relationship to the percentage of less nurses at hospitals to rise of &quot;accidental&quot; deaths.

If my wife's opinion in such things counts ( M.S.N., nurse practitioner + prior MBA and CPA), as a matter of common sense, the reduced staffing levels of health care professionals most certainly can contribute to infection rates and less patient care of all kinds. That level of staffing is a direct consequence of management decisions.

[ncs55]

SunEagle, I hear you. I have run across that same situation many times. Usually in smaller commercial systems 30-100kW. They spent everything they had to get the system installed and have no budget for maintenance and are counting on the production without the thought of a failure, to pay them back. It is hard for them to pay for just the materials sometimes. Some of the inverter boards can cost 2-3 thousand dollars and there are usually power supplies etc. that go with the boards that add an extra 3-6 hundred dollars each. The labor cost for repair is nothing compared to the materials. Boards and power supplies also usually fail due to grounding, or lack of, at the array. That can be very costly to repair properly. And in that case the labor is most of the cost. Picture example below of a grounding issue that caused many power supply failures before we got there and examined the array. What you see was not NEC and should have never made it past a final inspection for the time of the install. We had to ground the modules properly along with and before fixing the Xantrex 30-208 permanently. That is why you will see many comments in here from me about the importance of proper grounding of each module with a ground lug and bare copper connecting everything.