Upgrading panels

[caribconsult]

A few questions about upgrading:

We currently have 250w panels brand unknown but available, with enphaseM250 inverters at each panel and an ENVOY on the line. The max throughpower for the inverters is 250W, so if we wanted to replace all the panels with higher output ones, say 375w, we'd need to place the inverters as well? And the ENVOY?

Because the wattage is beyond their capability?

And they 'd have to be compatible with the existing control of the gateway, I suppose?

 

[gadfly56]

A few questions about upgrading:

We currently have 250w panels brand unknown but available, with enphaseM250 inverters at each panel and an ENVOY on the line. The max throughpower for the inverters is 250W, so if we wanted to replace all the panels with higher output ones, say 375w, we'd need to place the inverters as well? And the ENVOY?

Because the wattage is beyond their capability?

And they 'd have to be compatible with the existing control of the gateway, I suppose?

Is this the same system with which you took us on that epic journey regarding Tesla and the software debacle? I go with the idea that "If in ain't (mostly) broke, don't fix it."

 

[jaggedben]

A few questions about upgrading:

We currently have 250w panels brand unknown but available, with enphaseM250 inverters at each panel and an ENVOY on the line. The max throughpower for the inverters is 250W, so if we wanted to replace all the panels with higher output ones, say 375w, we'd need to place the inverters as well?

Depends on the actual specs of the module. It is perfectly fine to have an oversized module, to a point. In fact it's standard practice.
Enphase has a compatability tool here: https://enphase.com/en-us/support/module-compatibility
As an example, I believe LG375s (60-cell) are compatible with the M250.
Note there are two versions of the M250, one that is compatible with 72 cell modules and one that is not.

As long as the modules are compatible, the only downside to oversizing modules by so much is that you get clipping. i.e. the inverter still outputs only 250W even when much more is available. But since the panel doesn't always output its rating, and you get more energy at those times, some clipping is usually accepted. Probably I wouldn't put more than a 320W module on an M250 but it's your choice.

And the ENVOY?

You don't have to replace the Envoy if you don't replace the micros. However, the Envoy that is compatible with the M250 is not compatible with IQ micro-inverters currently available, so if you upgraded to IQs then yes, you'd need a new Envoy. In other circumstances it might not be necessary.

And they 'd have to be compatible with the existing control of the gateway, I suppose?

If you got new microinverters and a new Envoy they would need to be programmed similar to your existing setup, which I assume is not factory defaults.

 

[caribconsult]

Is this the same system with which you took us on that epic journey regarding Tesla and the software debacle? I go with the idea that "If in ain't (mostly) broke, don't fix it."

Yes, it is the same one, and your point is well-noted. Their customer support continues to be soporific, but through the help of their local crew we get excellent tech support, now. So I'm not inclined to mess with it.

 

[caribconsult]

Depends on the actual specs of the module. It is perfectly fine to have an oversized module, to a point. In fact it's standard practice.
Enphase has a compatability tool here: https://enphase.com/en-us/support/module-compatibility
As an example, I believe LG375s (60-cell) are compatible with the M250.
Note there are two versions of the M250, one that is compatible with 72 cell modules and one that is not.

As long as the modules are compatible, the only downside to oversizing modules by so much is that you get clipping. i.e. the inverter still outputs only 250W even when much more is available. But since the panel doesn't always output its rating, and you get more energy at those times, some clipping is usually accepted. Probably I wouldn't put more than a 320W module on an M250 but it's your choice.



You don't have to replace the Envoy if you don't replace the micros. However, the Envoy that is compatible with the M250 is not compatible with IQ micro-inverters currently available, so if you upgraded to IQs then yes, you'd need a new Envoy. In other circumstances it might not be necessary.



If you got new microinverters and a new Envoy they would need to be programmed similar to your existing setup, which I assume is not factory defaults.

So what you are saying is that I probably could put bigger panels on the M250 units, but I wouldn't get much more output that I'm getting now, since there is a 250w output cap on each inverter, but with the bigger panels, I might get that high limit more often, in lower light, because more sunlight is being harvested, do I understand that right? The array we have now is good for 5.1Kw, and the PW is generally charged by noon-1pm. I think it's enough.

 

[jaggedben]

So what you are saying is that I probably could put bigger panels on the M250 units, but I wouldn't get much more output that I'm getting now, since there is a 250w output cap on each inverter, but with the bigger panels, I might get that high limit more often, in lower light, because more sunlight is being harvested, do I understand that right? The array we have now is good for 5.1Kw, and the PW is generally charged by noon-1pm. I think it's enough.

I think you are getting the concept but probably need to adjust your thinking for certain panel wattages. Right now with 250W panels your inverters will never clip. Thing is, with 270W panels, they would probably never clip either, and you'd get approximately 8% more energy production. That's because solar panels usually underperform their nameplate rating in the real world due to soiling, high temps, and so on, not to mention all the times that insolation is less than full sun. So you have to go significantly higher with module rating to actually get a meaningful amount of clipping. And even beyond that point, you'll still get more energy harvest, although the returns will start to diminish at an ever increasing rate. Enphase has a white paper on this here, but it applies to all inverter types (in fact, even more so to other types than to micros). Anyway, for a rule of thumb, a DC-to-AC ratio of 1.2 (which would be a 300W module on your M250s) is considered fine and standard in residential solar.

With that said, you are probably right that your current array is good enough. And dumping perfectly good panels to replace with the latest and greatest is generally not worth the money. If you really need more energy harvest and have more roof space, probably better to add a few additional panels.

 

[caribconsult]

I think you assess the situation like I do....first and foremost, if it ain't broke, don't break it. Our current array is fairly well matched to our needs and capacity. On a reasonably sunny day, the PW is 100% charged by noon-1pm, on cloudy or hazy days, it may only get to the upper 90's but that's good enough to last through the night and hopefully there's sun the next day, since our house consumption is pretty reasonable. And I can zap it a bit with some grid if I want to bring the battery up to where it should be. Plus, the Envoy works with the M250 inverters and upgrading to the IQ's would add another $3K just for the inverters (from Enphase quote). The compatible Envoy I suppose is another $5-600, plus installation. Don't need it. About the envoy: I can read it two ways: on my LAN, since it has an IP address, or via the Enphase MyEnlighten website, but I get two different sets of numbers. Which to believe? The GUI might say 21 inverters detecting, 19 producing power, while the Enlighten site says 12 detected, 8 reporting or some stuff like that that doesn't add up. I think the GUI numbers are more accurate but the website lets you look at each panel's chart to see if something is amiss. Any opinion on Envoy? It has been quite an educational experience learning how this system works.

Now about the clipping, meaning the tops of the sine wave are clipped, approaching square waves if taken to the limit, and not good for household use, is that what we're talking about here? That would present an AC waveform that might not work well with all things? Somethings want pure sine wave. What risk is enountered if your AC is clipped sine waves? I suppose when the battery takes over after sundown, the waveform reverts to pure sine wave? Thanks as usual for your pointed advice. Much appreciated

 

[gadfly56]

I think you assess the situation like I do....first and foremost, if it ain't broke, don't break it. Our current array is fairly well matched to our needs and capacity. On a reasonably sunny day, the PW is 100% charged by noon-1pm, on cloudy or hazy days, it may only get to the upper 90's but that's good enough to last through the night and hopefully there's sun the next day, since our house consumption is pretty reasonable. And I can zap it a bit with some grid if I want to bring the battery up to where it should be. Plus, the Envoy works with the M250 inverters and upgrading to the IQ's would add another $3K just for the inverters (from Enphase quote). The compatible Envoy I suppose is another $5-600, plus installation. Don't need it. About the envoy: I can read it two ways: on my LAN, since it has an IP address, or via the Enphase MyEnlighten website, but I get two different sets of numbers. Which to believe? The GUI might say 21 inverters detecting, 19 producing power, while the Enlighten site says 12 detected, 8 reporting or some stuff like that that doesn't add up. I think the GUI numbers are more accurate but the website lets you look at each panel's chart to see if something is amiss. Any opinion on Envoy? It has been quite an educational experience learning how this system works.

Now about the clipping, meaning the tops of the sine wave are clipped, approaching square waves if taken to the limit, and not good for household use, is that what we're talking about here? That would present an AC waveform that might not work well with all things? Somethings want pure sine wave. What risk is enountered if your AC is clipped sine waves? I suppose when the battery takes over after sundown, the waveform reverts to pure sine wave? Thanks as usual for your pointed advice. Much appreciated

Regarding clipping, I believe that jaggedben is referring to the amount of energy harvested versus what's available. For a very simple example, let's suppose that the inverter is rated at 250 watts, and the panel is at 300 watts. In a perfect world where everything runs at 100%, the inverter will only be able to deliver 250 watts to your energy storage system, or your grid tie if you had one. The extra 50 watts the panel could supply gets "clipped".

 

[caribconsult]

Regarding clipping, I believe that jaggedben is referring to the amount of energy harvested versus what's available. For a very simple example, let's suppose that the inverter is rated at 250 watts, and the panel is at 300 watts. In a perfect world where everything runs at 100%, the inverter will only be able to deliver 250 watts to your energy storage system, or your grid tie if you had one. The extra 50 watts the panel could supply gets "clipped".

So it's wasted energy and there's seems to be no good reason to push into the clipping zone, would you agree? Match the panels to the inverters, don't push the inverter, is probably good advice for the longevity of the inverter, as well? And, no, we don't have a grid tie, or 'sell-back' system and we don't want one.

 

[gadfly56]

So it's wasted energy and there's seems to be no good reason to push into the clipping zone, would you agree? Match the panels to the inverters, don't push the inverter, is probably good advice for the longevity of the inverter, as well? And, no, we don't have a grid tie, or 'sell-back' system and we don't want one.

Well, I'd keep in mind his other very excellent points. If the panels get dirty, their output will drop. Likewise if it's hazy, or partly cloudy. To get good output most of the time, you will live with clipping some of the time. This is poking some of my OCD tendencies, where I'd start looking to see how many clear sky days there where in your locale vs hazy vs partly cloudy and try to balance that against potential wear on the inverter, cost of the panels, real estate they take on the roof, should you just add some more panels instead...ohhh, my head is starting to hurt now.

 

[electrofelon]

So it's wasted energy and there's seems to be no good reason to push into the clipping zone, would you agree? Match the panels to the inverters, don't push the inverter, is probably good advice for the longevity of the inverter, as well? And, no, we don't have a grid tie, or 'sell-back' system and we don't want one.

It is generally most economical to have some clipping. You can clip some, and the energy lost will never pay for what the next size up inverter would have cost (up to a point of course).

 

[jaggedben]

So it's wasted energy and there's seems to be no good reason to push into the clipping zone, would you agree? Match the panels to the inverters, don't push the inverter, is probably good advice for the longevity of the inverter, as well? And, no, we don't have a grid tie, or 'sell-back' system and we don't want one.

Read what I wrote again and the Enphase paper. Pushing into the clipping zone is fine and normal. It does not meaningfully affect inverter longevity.

 

[caribconsult]

I got it about the clipping part, but does the clipped waveform have any negative effect on anything you might plug in to a house system? Like TV's or LED lights? Or oscilloscopes? Longevity of the inverters isn't really an issue here, just a question of what the down side is to less than a good sine wave? And thanks so much for your input here. I learn a lot from your posts.

 

[electrofelon]

I got it about the clipping part, but does the clipped waveform have any negative effect on anything you might plug in to a house system? Like TV's or LED lights? Or oscilloscopes? Longevity of the inverters isn't really an issue here, just a question of what the down side is to less than a good sine wave? And thanks so much for your input here. I learn a lot from your posts.

There isnt actually any "clipping" going on, perhaps it is not the best word. What is happening is the inverter forces the system to produce less power by shifting the I-V curve off of the ideal point.

 

[jaggedben]

Yes it has nothing whatsoever to do with the AC waveform. It's about maximum output current. Read the Enphase paper.

 

[Carultch]

I got it about the clipping part, but does the clipped waveform have any negative effect on anything you might plug in to a house system? Like TV's or LED lights? Or oscilloscopes? Longevity of the inverters isn't really an issue here, just a question of what the down side is to less than a good sine wave? And thanks so much for your input here. I learn a lot from your posts.

What is "clipped", is the graph of power output as a function of time of day. Instead of being a round bell-like curve, it is flat-lined on top for a couple hours during the middle of the day, for all times where the bell would otherwise extend above the inverter's power limit.

The AC waveform is still as close to a sine wave as practical to achieve with the electronic filtering. The harmonic distortion is kept within the published %THD limit.

What ends up happening energy-wise, is sunlight "remains on the roof", in the form of hotter modules. The voltage of the panels is shifted away from the "sweetspot" at Vmp, close toward open-circuit voltage. The panels heat up, but not to any more than they are designed to withstand during an open circuit anyway.

 

[caribconsult]

What is "clipped", is the graph of power output as a function of time of day. Instead of being a round bell-like curve, it is flat-lined on top for a couple hours during the middle of the day, for all times where the bell would otherwise extend above the inverter's power limit.

The AC waveform is still as close to a sine wave as practical to achieve with the electronic filtering. The harmonic distortion is kept within the published %THD limit.

What ends up happening energy-wise, is sunlight "remains on the roof", in the form of hotter modules. The voltage of the panels is shifted away from the "sweetspot" at Vmp, close toward open-circuit voltage. The panels heat up, but not to any more than they are designed to withstand during an open circuit anyway.

Thanks for straightening that 'clipping' business out. I wrongly assumed what was meant was the traditional clipping of a sine wave you'd see when a circuit was overloaded or overdriven, and the the top of the wave (and the bottom) would be flattened...kind of on its way to a square wave, when that's not what is going on at all. I appreciate the explanation.