Category
- Products
-
- Wiki
- Contact us
Inverter manufacturers
sales@inverters.cn
Tag:
Anyways, please read the paper and form your own opinions, but my take on this paper is that it sets the minimum benefit of module level MPPT, and ANY additional variables will increase the benefit. Soiling, variable aging, and variable irradiance (shading) are very real considerations for any site which can have very significant effects on harvest. Additionally, without module level monitoring, there is no way to determine whether or not a modules are even meeting their standard deviation specifications.
Soiling and aging effects are especially hard to model, and the paper references establishes some techniques to quantify the potential benefit of microinverters under these conditions through measurements at an actual test site. Although the study did not use any global MPPT inverters, nonmonotoic conditions which "fooled" the inverter were excluded from the results.
Soiling and aging effects are especially hard to model, and the paper references establishes some techniques to quantify the potential benefit of microinverters under these conditions through measurements at an actual test site. Although the study did not use any global MPPT inverters, nonmonotoic conditions which "fooled" the inverter were excluded from the results.
100% failure is not an outlier in my experience. It may not be mechanical failure, but comms instead. I have seen very close to if not 100% of systems installed have some sort of issue, or multiple failure issues. No one really has the communications side of these things down yet. Communication, or inverter malfunction errors are typical on any micro installation I have seen. The costs of micro inverter versus central inverter replacement and micro service calls are much more expensive. I cannot imagine deploying micros on a commercial scale until a few things happen. The price per unit would have to fall dramatically in order to make up for the increased labor costs, and the communication issues would need to be orders of magnitude better.
We have installed a 140kw micro inverter system and it was at the customer request. The saving grace of the installation is that it is a carport structure and the micros are accessible via an 8' extension ladder. No modules have to be pulled in order to service a micro. I do not recommend micro inverters to anyone looking for more than a kw or so. That is also changing with newer small HF style central inverters with low MPPT ranges. That being said, the customer base Loves micro inverters. Many customers are sold on them before speaking to a sales rep, so someone on the micro side is doing something right! The shading argument is a little off in my opinion and there is a fine line between selling someone a system that has incremental shading, and selling someone a shady system..... System design with a micro can be more simple for a non technical sales person, but it can also lead to poor array placement if left unchecked by the technical support and engineering staff. There is a lot of, oh look a big roof with shade on it, lets use micros inverters happening out there right now. I am in agreement that they have a place in the pv world, but due diligence and robust system design should remain prime drivers on site selection.
We have installed a 140kw micro inverter system and it was at the customer request. The saving grace of the installation is that it is a carport structure and the micros are accessible via an 8' extension ladder. No modules have to be pulled in order to service a micro. I do not recommend micro inverters to anyone looking for more than a kw or so. That is also changing with newer small HF style central inverters with low MPPT ranges. That being said, the customer base Loves micro inverters. Many customers are sold on them before speaking to a sales rep, so someone on the micro side is doing something right! The shading argument is a little off in my opinion and there is a fine line between selling someone a system that has incremental shading, and selling someone a shady system..... System design with a micro can be more simple for a non technical sales person, but it can also lead to poor array placement if left unchecked by the technical support and engineering staff. There is a lot of, oh look a big roof with shade on it, lets use micros inverters happening out there right now. I am in agreement that they have a place in the pv world, but due diligence and robust system design should remain prime drivers on site selection.
First of all, the test procedure DID imitate a "Global MPPT" algorithm. All data points where the Fronius inverter was off-MPPT (i.e. when it was causing the "Christmas light effect") were excluded from the results (even though it's arguable that these are valid data points). NREL noted this in "Inverter MPPT Error" section of the study, stating that there would be another 2% energy harvest gain for the microinverter system in the moderate shade scenario, if they had included these MPPT errors.
Regardless, that "Global MPPT" algorithms or dual-MPPT channels address shade mismatch are misleading for two reasons:
(1) The string-level MPPT will still "turn-off" any shaded sections of the array (using a bypass diode), even though there is a lot of diffuse light still reaching the shaded section. The result is that the string-level MPPT is ALWAYS exacerbating the impact of shade.
(2) The "Global MPPT" is still limited to the input voltage range of the inverter, so the term "Global" is really just marketing spin. And, this has a further consequence that the effectiveness of "Global MPPT" algorithms will be more limited in situations where the string is short or the weather is hot.
To summarize, if you rely on bypass diodes and "Global MPPT" algorithms to address mismatch effects, you will ALWAYS be at a disadvantage to micro inverters. It's only a question of how much.
Regardless, that "Global MPPT" algorithms or dual-MPPT channels address shade mismatch are misleading for two reasons:
(1) The string-level MPPT will still "turn-off" any shaded sections of the array (using a bypass diode), even though there is a lot of diffuse light still reaching the shaded section. The result is that the string-level MPPT is ALWAYS exacerbating the impact of shade.
(2) The "Global MPPT" is still limited to the input voltage range of the inverter, so the term "Global" is really just marketing spin. And, this has a further consequence that the effectiveness of "Global MPPT" algorithms will be more limited in situations where the string is short or the weather is hot.
To summarize, if you rely on bypass diodes and "Global MPPT" algorithms to address mismatch effects, you will ALWAYS be at a disadvantage to micro inverters. It's only a question of how much.
Areas with shading, small systems (~5kW) and complex roof geometries are perfect candidates for a micro inverter system. Commercial systems, regardless of how much they are pushed, make no sense if you take ROI and LCOE into consideration in my opinion. Many of the arguments fall apart under the slightest scrutiny- seeing a rooftop full of naked rails with micro inverters attached to them should scare anyone away when looking at all those potential points of failure on a rooftop with no shading.
An outlier with respect to his 100% failure rate could have merit if he wasn't the only one suffering from failures or environmental issues. I was in Las Vegas yesterday and a few installers talked to me about their failure rates in the hot areas of the SW United States. Many micro inverter manufacturers rate their product to 65C ambient, but one installer showed me a FLIR image of the area underneath a micro inverter rooftop installation that read 85C! His micro inverters had been installed for over a year and he had lost a total of 3 MONTHS of production because the inverters disconnect due to high temperatures. This happens more often than people realize in these hot areas.
An outlier with respect to his 100% failure rate could have merit if he wasn't the only one suffering from failures or environmental issues. I was in Las Vegas yesterday and a few installers talked to me about their failure rates in the hot areas of the SW United States. Many micro inverter manufacturers rate their product to 65C ambient, but one installer showed me a FLIR image of the area underneath a micro inverter rooftop installation that read 85C! His micro inverters had been installed for over a year and he had lost a total of 3 MONTHS of production because the inverters disconnect due to high temperatures. This happens more often than people realize in these hot areas.
Tags: inverter, Micro inverter
A failure of a roof-mounted micro inverter requires a much larger effort to replace, compared with a wall-mounted string inverter. I expect the cost differential in labor to replace a roof-mounted micro inverter vs. a wall-mounted inverter would be substantial. This may factor into the overall system availability if a homeowner opts to not replace a single failed micro inverter right away.
I have also heard the opposite case to hold true for certain (primarily government) installs where the budget to purchase the PV system is available, but the budget for O&M down the road is zero. In this case, it is expected that the system output degrades gradually over time with individual component failures, versus a complete system failure should a central inverter fail with no budget for repairs.
I do find this installer's claims of having failures on 100% of their installed microinverter systems to be hard to believe. This seems like an outlier to me, not in line with anecdotal evidence that I am hearing from other installers. However, I'm not an expert in reliability, so I'm not going to get into it.
I have also heard the opposite case to hold true for certain (primarily government) installs where the budget to purchase the PV system is available, but the budget for O&M down the road is zero. In this case, it is expected that the system output degrades gradually over time with individual component failures, versus a complete system failure should a central inverter fail with no budget for repairs.
I do find this installer's claims of having failures on 100% of their installed microinverter systems to be hard to believe. This seems like an outlier to me, not in line with anecdotal evidence that I am hearing from other installers. However, I'm not an expert in reliability, so I'm not going to get into it.
Addressing "point d", CEC efficiency testing is conducted by an independent test laboratory, according to a protocol established by the Sandia National Laboratory. Here is a link to a webpage where you can download the test standard.
Addressing "point a" (and part of "point c"), I believe the test protocol has specific requirements around power factor and harmonic distortion. Please refer to the link above for that info.
Addressing "point b", the test of power conversion efficiency does not include MPPT efficiency, though I agree that this is equally important. In fact, this is why Enphase publishes MPPT efficiency on its product datasheet.
Addressing "point c", the M215's efficiency remains very high at low power levels thanks to a patented technology called "burst-mode". This technology enables Enphase Micro inverters to cycle on and off and to interleave power stages at very high speeds (on the scale of micro-seconds), in order to optimize the power conversion efficiency all the way down to a fraction of a watt. And, your reaction to these efficiency numbers shows just how innovative and unique burst-mode technology is-- it's something that doesn't exist in any other inverter (or microinverter), and is one of the (many) things that makes Enphase the efficiency leader in micro inverters.
Addressing "point e", I'm not sure where your temperature information came from. We have operating temperature information published on the M215 product datasheet.
Tags: inverter, Micro inverter
a) What kind of a load was used, will these inverters have such a high efficiency with inductive (refrigerators, washing machines) or diode-capacitive one (TV sets, computers, monitors etc) load or they just mean very rare nowadays pure resistive load?
b) Does this efficiency include efficiency of its MPPT or it is just an inverter efficiency?
c) In some cases such a high efficiency is achievable by reducing waveform quality cause the output capacitor takes some energy to recharge 120 times per second and an output coil also has its resistance, what about waveforms quality in the whole range of loads? Such a plain curve in case of Enphase M215-60-2LL (95-96% efficiency in a load range 10-100%) makes me to suspect they have a reduced output filter, but Powercom SLK-1500 looks more (85-96% efficiency with load range 10-100%) realistic - all right, a good output filter reduces efficiency in case of small loads.
d) by the way - who made all these measurements? The problem is that most of contemporary digital voltmeters and ampermeters give wrong results with non-sine voltage and current measurements, so the better filter - the worst efficiency will be shown. In some cases (Tom Bearden's MEG) they ever show "over-unity!"
e) Why the Enphase M215-60-2LL results are actual just for 25-40 Centigrades? What if somebody installs them at the roof, will it have these 25-40 Centigrades?
b) Does this efficiency include efficiency of its MPPT or it is just an inverter efficiency?
c) In some cases such a high efficiency is achievable by reducing waveform quality cause the output capacitor takes some energy to recharge 120 times per second and an output coil also has its resistance, what about waveforms quality in the whole range of loads? Such a plain curve in case of Enphase M215-60-2LL (95-96% efficiency in a load range 10-100%) makes me to suspect they have a reduced output filter, but Powercom SLK-1500 looks more (85-96% efficiency with load range 10-100%) realistic - all right, a good output filter reduces efficiency in case of small loads.
d) by the way - who made all these measurements? The problem is that most of contemporary digital voltmeters and ampermeters give wrong results with non-sine voltage and current measurements, so the better filter - the worst efficiency will be shown. In some cases (Tom Bearden's MEG) they ever show "over-unity!"
e) Why the Enphase M215-60-2LL results are actual just for 25-40 Centigrades? What if somebody installs them at the roof, will it have these 25-40 Centigrades?
Tags: inverter, Micro inverter
Micro Inverters are no the be all end all solution but I feel they do make sense in some smaller applications. For example in small residential systems they do make sense. For a one or 2 module gird tied system micro inverters are likely the only solution. As system size increases above a few kW Micro inverters become too costly.
Another important consideration is the type of installation. In the residential space the roof is typically complex, has obstructions or may be exposed to shading from nearby objects. In these types of scenarios distributed MPPT solutions (micro inverters and DC power optimizers) do make sense. The ability to use different string lengths, different mounting orientations, different size modules, and shade tolerance are all valuable tools in rooftop PV system design. I must disagree with the conclusion that distributed solutions never add energy, and that shading rarely occurs. Shading is common in the residential space and there have been several independent studies that show that where shading occurs distributed technologies add substantial added energy. This makes perfect sense because with distributed MPPT solutions the current of the entire string is not reduced by shading one module in the string.
On the other end of the spectrum are ground mounted utility scale systems. In these types of systems many of the design advantages offered by distributed technologies are of less valuable. String length, module orientation, and module type are all simple inputs to the system design equation. In utility scale system shading is normally not an issue since nearby objects that would create shade are very uncommon. Without shading, the potential for increased energy yield typically comes from module mismatch. With a well matched array the lower efficiency of micros compared to larger inverters makes it difficult to produce additional energy. DC optimizers fare better here since their higher efficiency means that the lost energy recovered from mismatch can be larger than the losses incurred by inserting the optimizers into the system. Well matched is a key phrase since many systems installed during the many "PV booms" around the world are far from well matched.
Another important consideration is the type of installation. In the residential space the roof is typically complex, has obstructions or may be exposed to shading from nearby objects. In these types of scenarios distributed MPPT solutions (micro inverters and DC power optimizers) do make sense. The ability to use different string lengths, different mounting orientations, different size modules, and shade tolerance are all valuable tools in rooftop PV system design. I must disagree with the conclusion that distributed solutions never add energy, and that shading rarely occurs. Shading is common in the residential space and there have been several independent studies that show that where shading occurs distributed technologies add substantial added energy. This makes perfect sense because with distributed MPPT solutions the current of the entire string is not reduced by shading one module in the string.
On the other end of the spectrum are ground mounted utility scale systems. In these types of systems many of the design advantages offered by distributed technologies are of less valuable. String length, module orientation, and module type are all simple inputs to the system design equation. In utility scale system shading is normally not an issue since nearby objects that would create shade are very uncommon. Without shading, the potential for increased energy yield typically comes from module mismatch. With a well matched array the lower efficiency of micros compared to larger inverters makes it difficult to produce additional energy. DC optimizers fare better here since their higher efficiency means that the lost energy recovered from mismatch can be larger than the losses incurred by inserting the optimizers into the system. Well matched is a key phrase since many systems installed during the many "PV booms" around the world are far from well matched.
When you hear about the suitability of a particular inverter technology for a particular project, it comes from the perspective of this hard-earned experience. To the best of my knowledge, no peer reviewed paper has demonstrated a performance advantage for micro inverters in commercial applications -- let alone an LCOE advantage (taking into account the considerable difference in price points for string or central inverters compared to micro inverters). Perhaps you can show me otherwise?
I have seen recent work presented by NREL which demonstrates the performance advantage of the micro-inverter architecture for shaded conditions. This analysis appears to be well done and we appreciate the rigorous approach. Again, though, I wonder if the performance advantage (3.7% in the case of light shading) can overcome the price premium for the microinverter system.
So we tend to think that other factors are at work in the inverter market. The perceived ease of use of micro inverters shouldn't be discounted, for example. In the fast-growing North American market there are many new entrants so this factor is important. We are excited to be launching our SB 240 micro inverter system this year and look forward to serving the needs of these new entrants to the market and those companies who have built their businesses around the characteristics of microinverter technology. We have a few tricks up our sleeves and are optimistic about our chances to compete in this segment. After all, our track record is pretty good.
However our firm opinion is that over time even these new entrants might seek ways to improve returns and move towards more established technologies with proven gains in LCOE or ROI.
BTW, this might also explain your comment about "integrators turning to micro inverters for large-scale projects." I wonder if you can point to (for example) any of the Top 15 commercial systems integrators who are using microinverter technology for large-scale projects? If one supposes that these large, sophisticated integrators might be using global best practices, then it might be important to note that these companies rely on central inverters or decentralized string inverter architectures to deliver leading returns to PV investors.
I have seen recent work presented by NREL which demonstrates the performance advantage of the micro-inverter architecture for shaded conditions. This analysis appears to be well done and we appreciate the rigorous approach. Again, though, I wonder if the performance advantage (3.7% in the case of light shading) can overcome the price premium for the microinverter system.
So we tend to think that other factors are at work in the inverter market. The perceived ease of use of micro inverters shouldn't be discounted, for example. In the fast-growing North American market there are many new entrants so this factor is important. We are excited to be launching our SB 240 micro inverter system this year and look forward to serving the needs of these new entrants to the market and those companies who have built their businesses around the characteristics of microinverter technology. We have a few tricks up our sleeves and are optimistic about our chances to compete in this segment. After all, our track record is pretty good.
However our firm opinion is that over time even these new entrants might seek ways to improve returns and move towards more established technologies with proven gains in LCOE or ROI.
BTW, this might also explain your comment about "integrators turning to micro inverters for large-scale projects." I wonder if you can point to (for example) any of the Top 15 commercial systems integrators who are using microinverter technology for large-scale projects? If one supposes that these large, sophisticated integrators might be using global best practices, then it might be important to note that these companies rely on central inverters or decentralized string inverter architectures to deliver leading returns to PV investors.
Claiming that today's micro-inverters---which use mixed-signal ASIC technology to operate at 96% efficiency and deliver utility interactive and wireless networking capabilities---is anything like the microinverters of the 1980's is absurd.
The rate of evolution in microinverters is faster than string inverters in every dimension (from performance and reliability to cost and features), which is the whole point of why end-customers are so interested.
I would hope that the rate of micro evolution is progressing as they have a ways to go to catch up with string inverters. They aren't as reliable, they aren't as efficient and they aren't as cost effective as string inverters for larger installations. It baffles me to hear about 600kW to 2MW micro installs. I find it hard to believe that those LCOE/ROI works out in favor of the customers, and I have yet to be proven wrong. I would be happy to sell people 4000 SB240's, but I feel compelled to warn against it for multiple reasons, mainly, because it doesn't make sense. Micros just aren't there- yet.
Also, I think you need to reread the article as he never tried to compare micros from 30 years ago to the ones today. He only mentioned that they have been around a long time.
The rate of evolution in microinverters is faster than string inverters in every dimension (from performance and reliability to cost and features), which is the whole point of why end-customers are so interested.
I would hope that the rate of micro evolution is progressing as they have a ways to go to catch up with string inverters. They aren't as reliable, they aren't as efficient and they aren't as cost effective as string inverters for larger installations. It baffles me to hear about 600kW to 2MW micro installs. I find it hard to believe that those LCOE/ROI works out in favor of the customers, and I have yet to be proven wrong. I would be happy to sell people 4000 SB240's, but I feel compelled to warn against it for multiple reasons, mainly, because it doesn't make sense. Micros just aren't there- yet.
Also, I think you need to reread the article as he never tried to compare micros from 30 years ago to the ones today. He only mentioned that they have been around a long time.
This is a recurring question for me in my Soar Academy classes and is
gaining traction. I want to add that the SMA Sunny Island can be used
with any solar/wind inverter regardless of make. SMA is coming out with
our own micro inverter after SPI and it will be able to be used with the
SI45/50/6048U inverters, although initially, it will not respond to the
Sunny Island frequency shifts.
That being said, the only issue with a non-SMA inverter will occur when the following three conditions exists:
1. The grid is out (or in a standalone system)
2. The batteries are fully topped off
3. There is no demand on the protected loads panel
If these three conditions are met, the Sunny Island will increase the micro grid frequency to throttle the RE inverters down to 0W to prevent battery overcharge. With a non-SMA inverter, it will not respond to these frequency shifts and as soon as the frequency hits the UL1741 limit of 60.5Hz it will immediately disconnect from the grid. Worst case, you will have to wait 5 minutes when the Island recognizes a demand and the isolation relays in the inverters will be overexercised. Not a big deal really.
That being said, the only issue with a non-SMA inverter will occur when the following three conditions exists:
1. The grid is out (or in a standalone system)
2. The batteries are fully topped off
3. There is no demand on the protected loads panel
If these three conditions are met, the Sunny Island will increase the micro grid frequency to throttle the RE inverters down to 0W to prevent battery overcharge. With a non-SMA inverter, it will not respond to these frequency shifts and as soon as the frequency hits the UL1741 limit of 60.5Hz it will immediately disconnect from the grid. Worst case, you will have to wait 5 minutes when the Island recognizes a demand and the isolation relays in the inverters will be overexercised. Not a big deal really.
Tags: inverter, Micro inverter