pv system connected to sub-panel

[Mike90250]

600Vdc Charge Controller, Morningstar also has one too.

[J.P.M.]

That was entirely my point (to give others food for thought). As I stated I have a string inverter and it was after a lot of research that I settled on Outback some of it here.

I would love to have a SolarEdge SE3000H with DC coupled VRLA batteries but such a beast does not yet exist
When it does come out, I will NOT be changing out my Outback for one because we seem to agree on some things like: don't fix what ain't broke.

Very Respectfully
Butch

Understood and ( perhaps) and maybe another one : Don't make things more complicated than necessary for the task. KISS.

[J.P.M.]

actually, my statement concerned the significant advantage of string over micro-inverters in being the ability to employ long dc-wire runs to avoid power loss. of course the ability to tag any digital or analog device with a unique identifier in order to provide RTA of this device presents 'an advantage' over the lack to do so.

no one actually answered the long wire run; although i now assume this is 'the' or one primary advantage.

As Butch writes, The long wire loss is mostly f(voltage), not whether the line is carrying AC or DC. Higher Voltage results in lower line loss.

For a given amount of power transmission, higher voltage reduces the line current which is what reduces the resistive losses in the conductor. One better way to make the losses as small as desired is to use appropriate wire thickness and or material as required by specs. Making the choice of micro inverter or optimizers vs, string inverter based on line loss due to voltage drop seems a bit odd.

[ButchDeal]

Butch: Thank you for the consideration, time and effort of your response. I doubt you and I will ever entirely agree on advantages/drawbacks of string vs. other methods of converting solar PV DC to AC, but I bet others reading this exchange may well enable them to be able to gain some useful information and food for thought to help make more informed decisions.

Respectfully,

J.P.M.

That was entirely my point (to give others food for thought). As I stated I have a string inverter and it was after a lot of research that I settled on Outback some of it here.

I would love to have a SolarEdge SE3000H with DC coupled VRLA batteries but such a beast does not yet exist
When it does come out, I will NOT be changing out my Outback for one because we seem to agree on some things like: don't fix what ain't broke.

Very Respectfully
Butch

[J.P.M.]

Butch: Thank you for the consideration, time and effort of your response. I doubt you and I will ever entirely agree on advantages/drawbacks of string vs. other methods of converting solar PV DC to AC, but I bet reading this exchange may well enable others to gain some additional useful information and food for thought to help make more informed decisions.

Respectfully,

J.P.M.

[ButchDeal]

actually, my statement concerned the significant advantage of string over micro-inverters in being the ability to employ long dc-wire runs to avoid power loss. of course the ability to tag any digital or analog device with a unique identifier in order to provide RTA of this device presents 'an advantage' over the lack to do so.

It isn't that they are DC power runs but that they are higher voltage. This can be done with both String and Optimized systems. Though not all string systems use high voltage. Most grid tie only systems do but most bimodal and off grid systems use 150V strings so not an advantage there (Schneider being the exception with their 600V Charge Controller). Though I hear OutBack is working on a 300V ChargeController.

no one actually answered the long wire run; although i now assume this is 'the' or one primary advantage.

If by " 'the' or one primary advantage" you mean the one or primary advantage that you are concerned with sure. Most people would consider this pretty low on the comparison chart unless they have very long wire runs.

[Jest Waitn]

You insisted on this by pushing for folks to weigh in on differences between string inverters and microinverters. You got your wish! Differences in monitoring, and the value of that, is an important part of the discussion.

actually, my statement concerned the significant advantage of string over micro-inverters in being the ability to employ long dc-wire runs to avoid power loss. of course the ability to tag any digital or analog device with a unique identifier in order to provide RTA of this device presents 'an advantage' over the lack to do so.

no one actually answered the long wire run; although i now assume this is 'the' or one primary advantage.

[Jest Waitn]

Understood (I think).

I've found terminology use and assuming everyone is on the same page with definitions can be confusing if not carefully explained.

Example: Up to about the mid point of my engineering career, I'd spent the better part of my time designing heat transfer equipment of all sorts, from power boilers to cooling towers and a lot of applications in between. So, once upon a time, some purchasing slug at a compressor co. my outfit was trying to garner business from challenged my design credentials and experience on a project proposal because there was no mention of "cooler design" on my resume, coolers being the generic term the compressor industry commonly uses for intercoolers and aftercoolers used on their products. He, not understanding that about half or more of the equipment I'd designed up to that point involved removing heat from a fluid or a process, and thus "coolers", didn't know that a "cooler" is an application more than an equipment classification, or that a "cooler" is actually more generally described as a heat exchanger. A short conversation cleared up the confusion, but his ignorance and myopic assumption caused an initial problem that might have caused a lost opportunity if not caught.

In this thread, I should have been more aware of similar terminology confusion and the problems imprecise writing and assumptions can cause. I'll make more effort to be more careful in the future.

cool.

[J.P.M.]

You said "if for no other reason than their being more of them per install"

Since you were talking about "modules", I assumed you were using the usual definition of that term. A module is what is sometimes called a "panel" - a collection of PV cells, usually 60 or 72 of them in an aluminum frame. A "module" is a good term for them as it is short and descriptive of it yet is a distinct term from the thing you put circuit breakers into.

My best guess is that for a situation with some shade the micro-inverter or optimizer setup gets you a few percent benefit vs. a string inverter.
Obviously it's going to change based on what kind of shading, etc.
I don't think I've seen "hype" that a 50% shaded array you'll get 50% of the unshaded output.
My guess is that you'd see less than 25% of the output with a string inverter (depending heavily on the type of shading)
And you'd see a significant increase in production with microinverters or optimizers.
BUT it'd really really really depend on the type of shading and the topology of the system.

Understood (I think).

I've found terminology use and assuming everyone is on the same page with definitions can be confusing if not carefully explained.

Example: Up to about the mid point of my engineering career, I'd spent the better part of my time designing heat transfer equipment of all sorts, from power boilers to cooling towers and a lot of applications in between. So, once upon a time, some purchasing slug at a compressor co. my outfit was trying to garner business from challenged my design credentials and experience on a project proposal because there was no mention of "cooler design" on my resume, coolers being the generic term the compressor industry commonly uses for intercoolers and aftercoolers used on their products. He, not understanding that about half or more of the equipment I'd designed up to that point involved removing heat from a fluid or a process, and thus "coolers", didn't know that a "cooler" is an application more than an equipment classification, or that a "cooler" is actually more generally described as a heat exchanger. A short conversation cleared up the confusion, but his ignorance and myopic assumption caused an initial problem that might have caused a lost opportunity if not caught.

In this thread, I should have been more aware of similar terminology confusion and the problems imprecise writing and assumptions can cause. I'll make more effort to be more careful in the future.

[sensij]

not to rain on anyone's parade, but this thread seems to have been hijacked. not that it is irrelevant to PV systems, but perhaps this 'departure' might be moved to the new topic?

You insisted on this by pushing for folks to weigh in on differences between string inverters and microinverters. You got your wish! Differences in monitoring, and the value of that, is an important part of the discussion.

[Jest Waitn]

not to rain on anyone's parade, but this thread seems to have been hijacked. not that it is irrelevant to PV systems, but perhaps this 'departure' might be moved to the new topic?

[ButchDeal]

1.) I believe some of that confusion centers around the use of the word "module". I'll need to own that your use of the word "module" was perhaps incorrectly assumed and interpreted by me to refer to either micro inverters or optimizers as in a SolarEdge system. If I misinterpreted your meaning, I apologize.

But I'd be less than candid if I didn't say that IMO, you could have been a bit more on topic or a bit more specific when referring to equipment other than the subject of the post which I feel was clearly confined to roof top inverter components.

Possibly. I and a few others stick to the nomenclature that modules are devices made of PV cells usually in groups 60 or 72, panel is a contiguous group of modules in a plane, array is a group of panels at a single site with one interconnect or controlling mechanism.

2.) The crux of my original point(s) was about micro inverters, and to a large extent, optimizers as well, in that a single device failure of a micro or optimizer on an array that would cause one panel's output to get knocked out, and such a relatively small reduction is likely to be less catastrophic in terms of array output than if an if a string inverter on a similar size and paneled system fails, knocking out the entire array's output. And so, because of that reduced noticeability, may well, without monitoring by either the user or the vendor, also stay unnoticed for some time, and with a higher probability of going unnoticed for longer than if a string inverter died. Such long times of unnoticed reduced output can thus have an effect on annual production. Right now, vendor monitoring may be good from decent vendors, but as (and if) things head south, vendor monitoring quality may suffer. I have little confidence in user monitoring unless a whole system stops. That's why I think individual panel monitoring capability is mostly a red herring.

I'll stick with the other points I made about rooftops being a harsh and unnecessary environment for electronics making them more failure prone; more of them, and so having a higher probability of failure due to more failure points; and lesser accessibility of rooftop equipment vs. ground accessible string inverters as points against micros and other electronic equipment on a roof.

I tend to agree here IF no one has taken the simple step to set up automated monitoring. It is relativly simply for an automated system with PV module level monitoring, to tell if there is a failure on a single micro inverter or optimizer and send an alert via text, email, or smart phone alert. Without taking the step to enable this the lose of a single PV module would be hard to notice on a non-string system. On a string system, it would be noticed at the next utility bill most likely for a diligent homeowner (not many are even this diligent).

3.) When you responded to my 1:02 P. M. Re: "adding to the chorus" post, it kind of read to me that you were disagreeing with my statement that the novelty of individual panel monitoring wears off quickly, and in doing so, implying that your were agreeing with the rest of the portions of my 1:02 P.M. post.

I was stating that looking at the module level output gets old but that having a system that knows about the module level output allows simple automated alerts such that the system can easily tell the homeowner (or installer, or broker etc) that there may be a problem. I get a few false alerts each day particularly in winter due to snow fall but SolarEdge makes it easy to clear those out in a few minutes. I have talked to them about adding some snow "smarts" to the alerts, they are working on it.

4.) When you stated you have been told you have "the most module replacements by far...", that is where I (looks like now, in hindsight) erred, in the sense that I assumed you were referring to micro inverters or optimizers replacement as I was, not entire panel/optimizer or panel micro combinations, or individual solar panels.

5.) However, the rest of my point 1 of my 2:02 P.M. post is essentially unchanged in that if one vendor has more replacements than other vendors of similar or even identical equipment, why would someone not logically come to at least one conclusion of several, that differences in failure rates of different vendors may lie in places other than the equipment ? An example: If vendor A and vendor B both install 1,000 M250 micro inverters over the same time period and calendar dates, and vendor A reports ten M250's need replacing, while vendor B reports two M 250's need replacing/ What's one possible conclusion ? Valid, correct, fair or not, one conclusion that's easier to say and spread than disprove is that, given that same equipment, the vendor with the higher replacement rate doesn't do as good an installation. I'm not calling you a bad vendor - FWIW, I believe you are ethical, honest and knowledgeable. I've seen the dirt vendors and damn them, and I believe I know the difference. I'm only asking what one semi logical conclusion to such a statement as you made about your seeming agreement with someone who said your outfit has the highest replacement rate by far. Why is that?

SImple, we are national. We use MANY installers and multiple distributors of the equipment. We have had a very few set of isolated failure clusters and they turned out to be installers with poor practices, which we have discontinued using those installers. We have no problem sending a different installer out to check an installers work when something like this is suspected, particularly on newer installers. In your example If Vendor A is checking the monitoring and Vendor B is not (like many installers tend to do) would this change your evaluation.
Note that with SolarEdge they have recently (about 2 years ago) added a feature of association such that individual site monitoring can be shared fully between two companies such are ours and our installers. Before that we had to set up SUB accounts for installers. With this older sub system, it lets us see EVERY site the installer has ever set up so I know that they do not monitor them as most of them are not fully set up, don't have customer access, have no alerts configured, and even some failures for months/ years (but no one is looking).

6.) Point 3 of my 2:02 post was almost a rhetorical question: If you have higher replacement rates of equipment, and it's assumed those higher rates are from problems discovered from monitoring, and, assuming replacement rates among (quality) vendors to be approx. the same by virtue of similar equipment and decent installation quality, would that not mean that there may well be a lot of equipment out there that is operating less than optimally but not monitored ? To the extent that may have some validity, see my point 2 of this post.

This is my point, there is a lot of equipment out there that is not operating optimally due to diode failures, wire failures, etc. Hard to tell. From the full access we have with some of our installers, I have access to several community scale projects that a few have worked on that have whole arrays down. Yes these are being "monitored" in the sense that they are sending data to the cloud but not in the since that anyone is looking at it or automated alerts are going out. One has had an entire inverter offline for months, another an entire string (half of one inverter), several have a smattering of failed pv modules. These are all large jobs using 3 phase 33kw inverters.
BTW when I say failed PV modules I am guessing but it is a fairly educated guess. The optimizer is working still so the PV module is putting out some power, but at a level below the neighboring modules by a factor of 1/3 or 1/4, which tends to indicate failed wiring or diodes. There are some that the optimizer is not reporting and these could be either the optimizer or PV module, but in our experience, is more likely to be (in order of probability): wiring, PV module, optimizer.

Other failures we see occasionally are Arc Fault which the inverters will shut down and stay down for arc faults till physically reset. These have so far always been wiring issues, one case was a poor connection in the inverter, and all the rest have been rodent damage to wires. We had a similar arc fault on a fronius, turned out to be improperly torqued wiring, but that was the day after turn on where often wire damage shows up months or years latter. (Yes we monitor Fronius systems as well, in fact we monitor nearly all the systems including OutBack, SMA, Enphase, Fronius, and SolarEdge).

7.) I don't know where or how I gave Foo1bar the impression - except perhaps in the confusion of the terminology discussed above - that I stated or implied that string inverters would allow smaller arrays for the same output as micro or optimizer equipped arrays. Most of us, including me, understand there is good design logic that leads to smaller arrays being possible with micro or optimizer equipped systems over sting inverter systems in applications where system shading is moderate to substantial.

I would argue however, and as a bit of an aside, that, because a micro or optimizer equipped system cannot create sunlight, it's not possible, for example, to have the instantaneous output of a 50 % shaded array to be restored to that of an identically oriented and sized unshaded array simply by using micros or optimizers. Seems to me that's somehow left unsaid and untouched for the solar ignorant to infer on their own.

I didn't get that impression.

8.) Nor did I ever state or mean to imply that the type of inverter method employed will have any effect (that I'm aware of anyway) on solar panel longevity or performance.
Maybe it's time for a sticky on micro inverter pros and cons.

I agree on the longevity but
I would say that optimizers can improve unshaded performance over time.
In the simplest account this comes into play with the fact that PV modules have a rating of X watts -0,+y , That difference will be better taken advantage of with optimizers (the +y part). Further they will not soil or degrade at a constant rate over time so the optimizers will continue to pull the max power from each PV module as the differences between them increases over the years.
This improved performance is not great but measurable. In my personal self install we spent a bit of time balancing the 4 strings of 3 modules such that each string would be as close as possible to each other in performance, few installers do that on string installs, and to be fair the gain is small. Now that my system is 3 years old, it is unlikely to be still as balanced since the modules likely degraded slightly differently.

I don't like tigo systems currently (maybe SMA can improve them). The communications are problematic and they complicate the installer rather then simplify it (like solaredge does). Our CEO and co founder put in a schneider bimodal system with Tigo optimizers (his is heavily shaded) at the same time as my Outback. Keeping those tigos reporting is a pain in the a$$, it is the main reason we do not install them and settled on Outback (outback does not have module level optimization but has better monitoring than schneider and simpler installs for us).

[ButchDeal]

One of my points might be better explained this way: With a micro inverter system equipped array of 16, ea. 300 Watt modules, one micro inverter failure may well not be noticed as quickly, or at all maybe as if the inverter dies on a similar paneled but equipped with a string inverter.

With no monitoring or alerting I agree. The string inverter system will have more lose in production likely noticed on the next bill. With automated alerts though the micro/optimized system would alert at most 48 hours though. Enphase does not make this alerting easy though and few people take the time to add an email address to solaredge to get the preconfigured alerts or adjust them ( they are highly configurable)

[foo1bar]

7.) I don't know where or how I gave Foo1bar the impression - except perhaps in the confusion of the terminology discussed above - that I stated or implied that string inverters would allow smaller arrays for the same output as micro or optimizer equipped arrays.

You said "if for no other reason than their being more of them per install"

Since you were talking about "modules", I assumed you were using the usual definition of that term. A module is what is sometimes called a "panel" - a collection of PV cells, usually 60 or 72 of them in an aluminum frame. A "module" is a good term for them as it is short and descriptive of it yet is a distinct term from the thing you put circuit breakers into.

My best guess is that for a situation with some shade the micro-inverter or optimizer setup gets you a few percent benefit vs. a string inverter.
Obviously it's going to change based on what kind of shading, etc.
I don't think I've seen "hype" that a 50% shaded array you'll get 50% of the unshaded output.
My guess is that you'd see less than 25% of the output with a string inverter (depending heavily on the type of shading)
And you'd see a significant increase in production with microinverters or optimizers.
BUT it'd really really really depend on the type of shading and the topology of the system.

[J.P.M.]

Butch: Further to my prior posts on this thread, and others from other posters, there looks to me like a lot of confusion about what I wrote in my 04/27/17, 2:02 P.M. response to your 1:12 P.M. post of the same date. Maybe the following will help clear up what I wrote and what I was trying to get at.

1.) I believe some of that confusion centers around the use of the word "module". I'll need to own that your use of the word "module" was perhaps incorrectly assumed and interpreted by me to refer to either micro inverters or optimizers as in a SolarEdge system. If I misinterpreted your meaning, I apologize.

But I'd be less than candid if I didn't say that IMO, you could have been a bit more on topic or a bit more specific when referring to equipment other than the subject of the post which I feel was clearly confined to roof top inverter components.

2.) The crux of my original point(s) was about micro inverters, and to a large extent, optimizers as well, in that a single device failure of a micro or optimizer on an array that would cause one panel's output to get knocked out, and such a relatively small reduction is likely to be less catastrophic in terms of array output than if an if a string inverter on a similar size and paneled system fails, knocking out the entire array's output. And so, because of that reduced noticeability, may well, without monitoring by either the user or the vendor, also stay unnoticed for some time, and with a higher probability of going unnoticed for longer than if a string inverter died. Such long times of unnoticed reduced output can thus have an effect on annual production. Right now, vendor monitoring may be good from decent vendors, but as (and if) things head south, vendor monitoring quality may suffer. I have little confidence in user monitoring unless a whole system stops. That's why I think individual panel monitoring capability is mostly a red herring.

I'll stick with the other points I made about rooftops being a harsh and unnecessary environment for electronics making them more failure prone; more of them, and so having a higher probability of failure due to more failure points; and lesser accessibility of rooftop equipment vs. ground accessible string inverters as points against micros and other electronic equipment on a roof.

3.) When you responded to my 1:02 P. M. Re: "adding to the chorus" post, it kind of read to me that you were disagreeing with my statement that the novelty of individual panel monitoring wears off quickly, and in doing so, implying that your were agreeing with the rest of the portions of my 1:02 P.M. post.

4.) When you stated you have been told you have "the most module replacements by far...", that is where I (looks like now, in hindsight) erred, in the sense that I assumed you were referring to micro inverters or optimizers replacement as I was, not entire panel/optimizer or panel micro combinations, or individual solar panels.

5.) However, the rest of my point 1 of my 2:02 P.M. post is essentially unchanged in that if one vendor has more replacements than other vendors of similar or even identical equipment, why would someone not logically come to at least one conclusion of several, that differences in failure rates of different vendors may lie in places other than the equipment ? An example: If vendor A and vendor B both install 1,000 M250 micro inverters over the same time period and calendar dates, and vendor A reports ten M250's need replacing, while vendor B reports two M 250's need replacing/ What's one possible conclusion ? Valid, correct, fair or not, one conclusion that's easier to say and spread than disprove is that, given that same equipment, the vendor with the higher replacement rate doesn't do as good an installation. I'm not calling you a bad vendor - FWIW, I believe you are ethical, honest and knowledgeable. I've seen the dirt vendors and damn them, and I believe I know the difference. I'm only asking what one semi logical conclusion to such a statement as you made about your seeming agreement with someone who said your outfit has the highest replacement rate by far. Why is that?

6.) Point 3 of my 2:02 post was almost a rhetorical question: If you have higher replacement rates of equipment, and it's assumed those higher rates are from problems discovered from monitoring, and, assuming replacement rates among (quality) vendors to be approx. the same by virtue of similar equipment and decent installation quality, would that not mean that there may well be a lot of equipment out there that is operating less than optimally but not monitored ? To the extent that may have some validity, see my point 2 of this post.

7.) I don't know where or how I gave Foo1bar the impression - except perhaps in the confusion of the terminology discussed above - that I stated or implied that string inverters would allow smaller arrays for the same output as micro or optimizer equipped arrays. Most of us, including me, understand there is good design logic that leads to smaller arrays being possible with micro or optimizer equipped systems over sting inverter systems in applications where system shading is moderate to substantial.

I would argue however, and as a bit of an aside, that, because a micro or optimizer equipped system cannot create sunlight, it's not possible, for example, to have the instantaneous output of a 50 % shaded array to be restored to that of an identically oriented and sized unshaded array simply by using micros or optimizers. Seems to me that's somehow left unsaid and untouched for the solar ignorant to infer on their own.

8.) Nor did I ever state or mean to imply that the type of inverter method employed will have any effect (that I'm aware of anyway) on solar panel longevity or performance.

Maybe it's time for a sticky on micro inverter pros and cons.