oversizing inverters/PV's - clipping

daemeonr

New Member
Joined
Oct 5, 2020
Messages
5
I *think* I am starting a new thread, would the moderator advise if I did not?

(copied from an existing thread, with typos fixed ;{) I joined this website to warn all of you about the marketing hype and misrepresentation that Enphase is doing with their clipping vs. "supported" panels, e.g. the IQ7+ and "450w" panels. AND THEY KNOW IT. Your mileage may vary, but for me, buying the Enphase microinverters was a COSTLY mistake (I am fighting enPhase to get a RMA refund!). For those who do not have a mix of string & microinverters generating power, hence may have a reference system to actually compare the lost power of an Enphase setup vs. a modern string inverter, I hope this info is helpful.

On my (large) roof I have an existing system with a brand new high efficiency string inverter, and a new Enphase system. I was therefore able to actually compare the relative power output, and the shocking power (and $$$) loss due to the Enphase system.

I live in the San Francisco Bay Area in an 150 year old house on top of a volcanic cone. 100-150 feet higher than surroundings, no trees or PV array shadowing.

South West Roof: 16 years ago I put up 16 200w Kyocera panels with a Xantrex 3.2kw string inverter. A couple of months ago (one month out of warranty but 4 months after provable failure) my 3-times replaced Xantrex string inverter died again. Replaced it with a brand new SMA 3.8 transformerless unit (single string), street price $1300.

South East Roof: 8 weeks ago installed 16 405w Jenko panels with 16 enPhase IQ7+ microinverters. The clipping from this junk is costing $1-3k per year, and I am fighting with Enphase to get a credit when I rip them out. ... They clip multiple kilowatts on multiple days, even the hot ones we have had here in the San Francisco bay area (and the panels still have a layer of soot on them!!!). I calculated (conservatively, allowing for utility's time of use pricing) that I am losing $1-3k per year. BUYING THE ENPHASE'S WAS A MISTAKE given my 405w configuration. The power loss would pay for a 10-year warranty SMA 7.7 in less than every two years!!!!


The Enphase array points south east the SMA array points south west. No trees, now shadowing; get breezes pretty continuously off the bay, arrays are 100-200' above any houses between them and the SF Bay, less than 100' to main panel. So ... the SMA is a 1.3% (97%) more efficient than the Enphase, the power peaks (per SMA/enLighten graphs) 1:45 later on the SMA than the Enphase PV array. This allows me to compare the clipping of the Enphase vs. the curve of the SMA (not actual power, but the curve shapes!). FYI, it looks like the Kyocera's may have degraded 5-7%, but the margin of error is great enough that I cannot get closer than that. Note that a UC Davis study of 20-30 year old panels showed 4-5% degradation relative to as-new flash! Here again we have Enphase claiming "20% power loss over the 25 year life of the panels" which is PROVABLY false.

Results: with these 405w panels, even with the fire weather, red sky, haze, etc. I see clipping on the Enphase of 1.5-3 hours/day for 24 days out of the past 40. Using the ration between the SMA & the Enphase power out (when there is a bit of fog/overcast and no clipping) suggests the Enphase clips about 1-3kw PER DAY (1kw @ 1.5 hrs clipping, 3kw @ 3 hrs clipping), and has done so 24 days out of the past 40. Note the clipping even occurred at 95 degrees ambient temp (did not check the panel temps) ... therefore if you get a lot of sun, especially with breezes to cool the panels, do NOT exceed the clip output by more than 10% (for 25 year life) plus maybe 20%. Remember that you have to pay $500 for the Enphase monitor, then $120-140 per IQ7+. Recall that wire size is a function of current, not voltage, so the strings (400v) are less of an issue than the microinverter (110v).

The Enphase's are expensive compared to "street price" of SMA's. I believe the Enphase inverters are generally a very expensive marketing gimmick all built around the "oversizing" marketing FUD. Yes oversizing makes sense, but claiming support for a 440w panel but clipping to 290w (IQ7+) is darned near fraud.

FYI, you don't care about the number of cells (60, 72, 144 1/2 cells, etc.), you only care that the VoC (Voltage Open Circuit) is not exceeded, unless you live somewhere that you get really bright midwinter sunny days (voltage increases as the temp drops for a given sun energy). Watch out for the "500w dual sided" panels though. They are tricky

So my takeaways:
  1. If you have shade, lots of overcast, etc. such that your panels "nearly never" attain rated power, and if you get a great price on high power panels, then consider oversizing.
  2. Compare the "street" rather than "mfg list" price: the fixed cost of the Enphase Envoy at $600 plus $133 per IQ7+ vs. e.g. an SMA Sunny Boy at $1200 for a 3.8/$1400 for a 7.7kw.
  3. One sees (e.g. Jenko) 405w panel priced such that the IQ7+ is almost as much as the panel. But adding a panel to a string inverter is pretty cheap
  4. Your panels can lose about as much capacity from a layer of dirt/soot as they will deteriorate over time (based on the UC Davis analysis)
  5. Derating of the inverters applies equally well to either string or micro-inverters, SMA's handle over current fine (they clip by design), but must NEVER exceed rated input voltage! Note that Voc (Voltage Open Circuit) will increase in really cold sunny weather. I do not have that issue here.
  6. Consider the repairability of the components. I chose SMA both because of their reputation and how easy it is to get them repaired if out of warranty.
  7. I do not know the real failure rate of a string inverter, but it costs $500-750 (plus shipping) to have an SMA rebuilt. I contacted 2 repair facilities, lighting seems to be the cause of nearly all repairs. So invest in lighting protection. It is CHEAP INSURANCE.
  8. Assume warranties of 25 years life for the micro+Envoy, vs. 10 years for the SMA (unless you get the addl 5 year warranty from SMA for ~$300 I think?). This means that if Enphase clipping costs you more than $1500 in 15 years, the SMA string is cheaper (I am ignoring the discounted cash value of the added upfront cost of the Enphase)
  9. Yes, it is easier to identify which panel is not producing warranty'ed power with the IQ. What is that worth to you?
  10. Do you care that you only get single phase power from the IQ vs. 2 phase from an SMA? E.g. for charging batteries should you eventually need that? Your house is wired for two phase (sometimes called "split phase")
  11. FYI, my biggest electronics costs have been around my utility's providing out of phase power (voltage vs. current phase), overvoltage, and brownouts. The Xantrex turns out to have been poorly designed for dirty power. The SMA's seem to have consistently better designs in this area, but that does no good for my house ;{) I do not know about the Enphase micro's and Envoy.
 

Ampster

Renewable Energy Hobbyist
Joined
May 3, 2020
Messages
4,371
Location
Kenwood, California
I have not looked up the Enphase sizing tool to see what the tool suggested but how did you decide to go with that model inverter?
The reason I am asking is I bought some IQ-7s and I used the tool to find the right model to go with my Sunpower 305 Watt panels. I have the opposite issue, in that because the Sunpower panels are 96 cells the decision was driven by the voltage of the panels and I have inverters that will never see their maximum output.
 

daemeonr

New Member
Joined
Oct 5, 2020
Messages
5
I spoke with Enphase engineering today which tried to BS me. Did not work. Finally admitted (barely) that The Enphase sizing tool has a database of the various panel specifications. It references that then specifies inverters on the first the Voc, then whether the inverter INPUT can handle the panel's flash power. Apparently does not even take input of seasonal cold/w sun temperature.

So one gets the inverter that can handle the Voc, and the "suggested" power range. The fact is the selection tool does not even care about the PV's flash power except to assure that the MPP can adjust to that power range. In other words, Marketing drove engineering to make inverters to handle high input by moving the MPP, without engineering microinverters that handle the power.

I must commend Enphase. When faced with these irrefutable facts and their "optimistic" calculations, not to mention the very outdated and then-relevant "white paper", Enphase is working to swap the 7+ with 7A's.

The engineering design of Sunpower panels is rather, er, interesting. They designed the panels to be used with a custom branded SMA string inverter, and saved money by configuring the panels to have higher voltage for the watts. When poor souls like you buy the panels, you are forced to an oversized microinverter. Not sure when you bought the 7's or what you paid for them.
 

Ampster

Renewable Energy Hobbyist
Joined
May 3, 2020
Messages
4,371
Location
Kenwood, California
The engineering design of Sunpower panels is rather, er, interesting. They designed the panels to be used with a custom branded SMA string inverter, and saved money by configuring the panels to have higher voltage for the watts. When poor souls like you buy the panels, you are forced to an oversized microinverter. Not sure when you bought the 7's or what you paid for them.
The Sunpower panels I purchased were used and they cost me $90 each. They came off a commercial building that was being reroofed. Seven of them are on a string with my Skybox inverter and are working fine. The other four of them were hooked up to IQ7s on that part of a west facing patio cover that had shade. I think I paid $120 or $130 each for the IQ7s two years ago. I have no complaints about about the IQ7s or the Sunpower panels. I have another installation of new Sunpower panels on a rental property that I used to live in. Based on that experience I knew I could take a chance on used ones. Like the ones with shade they are on a west facing patio cover and I am happy with the production.
 
Last edited:

Ampster

Renewable Energy Hobbyist
Joined
May 3, 2020
Messages
4,371
Location
Kenwood, California
In my earlier post I did not intend to imply that I was unhappy in any way with the performance of the Sunpower panels and the IQ7s. My TOU rate in the afternoons is $0.50 per kWh and those panels and micros have almost paid for themselves in the production that I have sold to the grid in the past two years.
 

treydot

New Member
Joined
Jul 18, 2021
Messages
1
I *think* I am starting a new thread, would the moderator advise if I did not?

(copied from an existing thread, with typos fixed ;{) I joined this website to warn all of you about the marketing hype and misrepresentation that Enphase is doing with their clipping vs. "supported" panels, e.g. the IQ7+ and "450w" panels. AND THEY KNOW IT. Your mileage may vary, but for me, buying the Enphase microinverters was a COSTLY mistake (I am fighting enPhase to get a RMA refund!). For those who do not have a mix of string & microinverters generating power, hence may have a reference system to actually compare the lost power of an Enphase setup vs. a modern string inverter, I hope this info is helpful.

On my (large) roof I have an existing system with a brand new high efficiency string inverter, and a new Enphase system. I was therefore able to actually compare the relative power output, and the shocking power (and $$$) loss due to the Enphase system.

I live in the San Francisco Bay Area in an 150 year old house on top of a volcanic cone. 100-150 feet higher than surroundings, no trees or PV array shadowing.

South West Roof: 16 years ago I put up 16 200w Kyocera panels with a Xantrex 3.2kw string inverter. A couple of months ago (one month out of warranty but 4 months after provable failure) my 3-times replaced Xantrex string inverter died again. Replaced it with a brand new SMA 3.8 transformerless unit (single string), street price $1300.

South East Roof: 8 weeks ago installed 16 405w Jenko panels with 16 enPhase IQ7+ microinverters. The clipping from this junk is costing $1-3k per year, and I am fighting with Enphase to get a credit when I rip them out. ... They clip multiple kilowatts on multiple days, even the hot ones we have had here in the San Francisco bay area (and the panels still have a layer of soot on them!!!). I calculated (conservatively, allowing for utility's time of use pricing) that I am losing $1-3k per year. BUYING THE ENPHASE'S WAS A MISTAKE given my 405w configuration. The power loss would pay for a 10-year warranty SMA 7.7 in less than every two years!!!!


The Enphase array points south east the SMA array points south west. No trees, now shadowing; get breezes pretty continuously off the bay, arrays are 100-200' above any houses between them and the SF Bay, less than 100' to main panel. So ... the SMA is a 1.3% (97%) more efficient than the Enphase, the power peaks (per SMA/enLighten graphs) 1:45 later on the SMA than the Enphase PV array. This allows me to compare the clipping of the Enphase vs. the curve of the SMA (not actual power, but the curve shapes!). FYI, it looks like the Kyocera's may have degraded 5-7%, but the margin of error is great enough that I cannot get closer than that. Note that a UC Davis study of 20-30 year old panels showed 4-5% degradation relative to as-new flash! Here again we have Enphase claiming "20% power loss over the 25 year life of the panels" which is PROVABLY false.

Results: with these 405w panels, even with the fire weather, red sky, haze, etc. I see clipping on the Enphase of 1.5-3 hours/day for 24 days out of the past 40. Using the ration between the SMA & the Enphase power out (when there is a bit of fog/overcast and no clipping) suggests the Enphase clips about 1-3kw PER DAY (1kw @ 1.5 hrs clipping, 3kw @ 3 hrs clipping), and has done so 24 days out of the past 40. Note the clipping even occurred at 95 degrees ambient temp (did not check the panel temps) ... therefore if you get a lot of sun, especially with breezes to cool the panels, do NOT exceed the clip output by more than 10% (for 25 year life) plus maybe 20%. Remember that you have to pay $500 for the Enphase monitor, then $120-140 per IQ7+. Recall that wire size is a function of current, not voltage, so the strings (400v) are less of an issue than the microinverter (110v).

The Enphase's are expensive compared to "street price" of SMA's. I believe the Enphase inverters are generally a very expensive marketing gimmick all built around the "oversizing" marketing FUD. Yes oversizing makes sense, but claiming support for a 440w panel but clipping to 290w (IQ7+) is darned near fraud.

FYI, you don't care about the number of cells (60, 72, 144 1/2 cells, etc.), you only care that the VoC (Voltage Open Circuit) is not exceeded, unless you live somewhere that you get really bright midwinter sunny days (voltage increases as the temp drops for a given sun energy). Watch out for the "500w dual sided" panels though. They are tricky

So my takeaways:
  1. If you have shade, lots of overcast, etc. such that your panels "nearly never" attain rated power, and if you get a great price on high power panels, then consider oversizing.
  2. Compare the "street" rather than "mfg list" price: the fixed cost of the Enphase Envoy at $600 plus $133 per IQ7+ vs. e.g. an SMA Sunny Boy at $1200 for a 3.8/$1400 for a 7.7kw.
  3. One sees (e.g. Jenko) 405w panel priced such that the IQ7+ is almost as much as the panel. But adding a panel to a string inverter is pretty cheap
  4. Your panels can lose about as much capacity from a layer of dirt/soot as they will deteriorate over time (based on the UC Davis analysis)
  5. Derating of the inverters applies equally well to either string or micro-inverters, SMA's handle over current fine (they clip by design), but must NEVER exceed rated input voltage! Note that Voc (Voltage Open Circuit) will increase in really cold sunny weather. I do not have that issue here.
  6. Consider the repairability of the components. I chose SMA both because of their reputation and how easy it is to get them repaired if out of warranty.
  7. I do not know the real failure rate of a string inverter, but it costs $500-750 (plus shipping) to have an SMA rebuilt. I contacted 2 repair facilities, lighting seems to be the cause of nearly all repairs. So invest in lighting protection. It is CHEAP INSURANCE.
  8. Assume warranties of 25 years life for the micro+Envoy, vs. 10 years for the SMA (unless you get the addl 5 year warranty from SMA for ~$300 I think?). This means that if Enphase clipping costs you more than $1500 in 15 years, the SMA string is cheaper (I am ignoring the discounted cash value of the added upfront cost of the Enphase)
  9. Yes, it is easier to identify which panel is not producing warranty'ed power with the IQ. What is that worth to you?
  10. Do you care that you only get single phase power from the IQ vs. 2 phase from an SMA? E.g. for charging batteries should you eventually need that? Your house is wired for two phase (sometimes called "split phase")
  11. FYI, my biggest electronics costs have been around my utility's providing out of phase power (voltage vs. current phase), overvoltage, and brownouts. The Xantrex turns out to have been poorly designed for dirty power. The SMA's seem to have consistently better designs in this area, but that does no good for my house ;{) I do not know about the Enphase micro's and Envoy.
This video provides an excellent explanation about Enphase microinverter clipping and how it actually affects power output. Bottom line is that the increased efficiency of energy production using Enphase microinverters when solar irradinace is not at peak more than compensates (generally) for the loss of energy production caused by clipping during peak hours. For example during morning and afternoon hours, under cloudy or partly cloudy conditions, etc. I know this is much more complex than I have just described, but Dave Jones' video gives an in-depth analysis of why you should use panels that are higher rated than the inverter's capacity. Well worth the watch:
 
Last edited:
Top