I can conceive of how this would contribute to component stress, but it is very common practice and gets into a more efficient operating range for the inventer. My DC input is well here the AC rating for my inverter and failed within two years
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Is this clipping?
- Thread starter uscpsycho
- Start date
charlesj
Active Member
I wonder where the inverter companies got the data that >1:1 , up to 1.5 is ok without qualifications.
"Clipping", is more common than apparently accepted. This for many reasons:
Take two homes:
1) I have only so much acceptably oriented panels. My inverters are sized such that I will never ever clip.
2) My son has a simply massive amount of ideally oriented panel space. He's way over-paneled relative to the incentives and constraints in our area, but we bring him right to the max by clipping. He does not really clip that often, but all the rest of the time he's producing a ton of power. It's laughable how much power he makes in a year.
- As PvWatts often shows:
- What you're going to get and what your panels are capable of are often vastly different things.
- Many folks will receive their "peak" power level for exceptionally short periods of time and or season.
- Some utilities and or area's limit the size of your panel array, and or make imposing to pass certain sizes. In these situatons clipper is a well accepted method of chopping off the "max rated" power, yet because of the larger panel field now allowed, a much greater annual output.
Take two homes:
1) I have only so much acceptably oriented panels. My inverters are sized such that I will never ever clip.
2) My son has a simply massive amount of ideally oriented panel space. He's way over-paneled relative to the incentives and constraints in our area, but we bring him right to the max by clipping. He does not really clip that often, but all the rest of the time he's producing a ton of power. It's laughable how much power he makes in a year.
Sarcasm? I can't really tell in this instance. Inverter companies are the designers and manufactures of their equipment and are the ones setting the limits for max DC input power and max AC output power. If their devices fail while within the operating specs then they are on the hook for warranty fixes, so it is in their best interest to define specifications that don't damage their devices.
Here are Solaredge specs:
| SE3000H-US | SE3800H-US | SE5000H-US | SE6000H-US | SE7600H-US | SE10000H-US | SE11400H-US | |
| Max AC Ouput | 3000 | 3800 | 5000 | 6000 | 7699 | 10000 | 11400 |
| Max DC Input | 4560 | 5900 | 7750 | 9300 | 11800 | 15500 | 17650 |
| DC/AC Ratio | 1.52 | 1.55 | 1.55 | 1.55 | 1.53 | 1.55 | 1.55 |
Mostly up to 1.55 DC/AC ratio with two at 1.52 and 1.53.
Last edited:
charlesj
Active Member
I am sorry. Yes, it was sarcasm.
Responded to
"Overdriving is BS. It will cause the inverter to fail."
This is what my system did every day until it stopped working. 9.6kw system with two powerwalls. One of those is a Powerwall + with a 7.6kw inverter. Every single day my production would clip at 7.7kw, until it stopped working on August 14th around noon. The inverter shut off, so we’ve assumed it broke. Tesla finally sent me an email yesterday to schedule a technician to come out on November 3rd. I went off when I found out if will be another two months before they even send someone to check it out. When I asked if they were going to show up with a new inverter and fix it on November 3rd they said according to what they can see on their end arching occurred. They didn’t tell me where, but my theory is that it has something to do with constant clipping. The fact that we hit 7.7kw almost every single day for multiple hours makes me think that caused the arching that then caused our solar to shut off.
Has this happened to anyone else?
Matt-FL
Member
My system clips on a daily basis (when sunny, of course), been installed for just over a year. I have 20.4kW (51 Tesla 400w panels) and two of the prior generation 7.6kW inverters. (Original plans were PW+ inverters, but my utility blocked Tesla from installing, so they redesigned with the older gen).
My system maxes out at 15.6kW for steady run, but on partly cloudy days, it will jump to 17.2kW for brief periods.
I have not experienced any arcing or other issues from the system. It's been rock solid since it was installed.
My system maxes out at 15.6kW for steady run, but on partly cloudy days, it will jump to 17.2kW for brief periods.
I have not experienced any arcing or other issues from the system. It's been rock solid since it was installed.
That does not look like clipping to me. When mine does, it is 100% flat lined at the top
Clipping is 100% safe.
Inverters are NOT designed to only do max output for a short period of time. Inverters have to pass dealing with max output for many hours at a time, I would be surprised if they were not tested for the almost impossible 24/7 max output for weeks on end.
Plus everything in a solar install is considered a continuous load, and this the breakers and wires are sized at an 80% de-rating.
We have run our two 10K solar edge inverters clipping regularly for 2-3 months of the year.
I am not saying that if there were marginal parts, that running at max output for several hours may not help push them over the edge, but the reality is you want it pushed over the edge now, while in warranty, vs holding on just long enough for the warranty to expire.
-Harry
I’ve posted on a couple threads about clipping in afternoon over last month. Then they replaced PW+ inverter last week. Everything fixed.
Then today they rolled me to FW 23.28.1. And I lost production by about 50%, which I later found out is so the system can tests one string at a time…and apparently this was due to some arc fault issues.
Starting to think these things are all connected now.
IMHO that could/should be re-balanced so that both inverters are better optimized, assuming that both are 7.6KW
Again, it's entirely possible that the multiple hours of full load is putting more stress on marginal components and helping them fail.
That is not how the inverter was designed, tested, or certified, and IMHO you want that failure to happen and the parts/inverter to get replaced.
-Harry
Clipping one is 7.7 and the other 7.0. Agreed moving a couple of panels over would be nice but at this point will be slightly involved unless I want cables crossing above ground on the pole mounts. Interestingly, the graph is from late May. In early April, the 7.0 does get close to the 7 line.
7.7 and 7.0, I could see it not being worth it to move panels around.
I still don't see that amount of clipping causing any issues for a properly built, not defective, Inverter.
Of course I have also been involved in electronic devices that were 10x over rated and still hit a fault condition we did not expect of 100x rated. This was not in Solar equipment.
-Harry
Originally it was going to be 2x 7.7s. Certified Tesla PW installer erroneously thought 2 PWs had a 15kW limit, so inverters got switched to 7.0 and 7.7. Lo and behold, 2 PWs couldn’t handle 14.7 and the system would shut down. So Tesla installer added another PW at their expense and thus far things have been mostly humming along. Would’ve been nice to have the second 7.7 inverter or strings better balanced but I’ve learned not to sweat it, especially as we’re just about breaking even on our annual electric bill.
Yes that is clipping. When clipping, the inverter is at 100% effort. The higher the DC:AC ratio the higher the likelihood of clipping. I bet the cooing fan on the inverter was running pretty hard too.
I have a 12.24 kW system with two powerwalls, on is a PW+. You probably have the same. Last summer, after months of clipping for hours everyday, the inverter (not even a year old) finally gave up the ghost. Having only one inverter, my solar system was inop. Tesla customer service is the worst, and our karma is not great. They finally came after a few weeks to diagnose. Yes, the inverter needed replacement. This had to be ordered and it would be a few more weeks.
As fate would have it, the replacement was scheduled the day Hurricane Ian made landfall (the eye went over my house). The panels, less one struck by a piece of roof tile, hung tough and were ready to go, but not without the inverter. I had the event many of of us buy solar with battery backup for and it didn't work (and the powerwalls, at 100% charge didn't work either). Tesla customer service is the worst. They could have come the day after, but it took them a week to get around to it. FPL got the power restored before Tesla fixed my system.
Lessons learned:
-- Inverters are the weakest link in the solar system. String or mini, inverters fail. Asking any piece of equipment to run at 100% rated capacity, which is what is happening when the system clips, increases the probability of failure.
-- DO NOT build a mission critical system with a single point of failure. Tesla does this when they provision only one inverter.
-- KEEP the DC:AC ratio as low as possible. DC:AC > 1 is to save cost. There is some argument that inverter efficiency increases with load, but I haven't seen any data. A low ratio allows the system to operate at a more sustainable level and give you headspace in a string system should you choose to just add panels.
I added a 7.6 kWh inverter. The system load is now split. Neither inverter comes close to clipping and over system production has increased ~20%. The inverters run at a more sustainable level. The redundancy gives me peace of mind.
Good luck.
dhrivnak
Active Member
We almost always overdrive and that is fully supported by the inverter manufacturers. With overdriving DC to AC you can lose a little by clipping but the system wakes up earlier, produces longer and runs at a higher efficiency for most of the day. I am 25% oversized with DC and have been for over 7 years now.
Sorry about having the equipment and not being able to use it. That hurts.
Just to clear up the "more efficient" claims, we are talking about a few percent difference from something like 40% to 100% loading of an inverter. Not a lot.
e.g. Two Reasons You Must Look At Efficiency Curves When Choosing Your Solar Inverter with two great figures;
and more importantly;
Just because you are getting "the most efficient" inverter says nothing about whether that inverter will produce the most power on your panels over a year. (Above the green version is "more efficient", but it will produce less overall power than the red version over normal load conditions.)
The bit that does not get enough air time, in my opinion, is that all inverters become less efficient as they heat up. That alone argues for having your standalone inverter spend more time at less than 100%. Again, the temperature effect amounts are a few percent to ten percent or so. The effect is much less pronounced for microinverters which are designed for much higher operating temperatures. (But it is still a real effect.)
If you are sweating efficiency, it normally requires a deep dive into the efficacy plots for a given inverter model by power, and by temperature. Different designs lose efficiency in different ways, and I think that modeling the efficiency effects at any given site is not straightforward. I would say that at some point, one pretty rapidly hits the point of diminishing ROI, especially with standalone inverters where the capital cost is high.
I think that if I were dependent upon a stand alone inverter, I would own a spare. The old adage that for critical items one equals zero, and two equals one applies to standalone inverters in my book. Or have a standby generator to be able to step in and cover for you.
All the best,
BG
I have yet to see any documentation mentioning "overdrive". I think it is a myth. How can an inverter with a rate max output of X do more than X?
Thank you for the well articulated response.Sorry about having the equipment and not being able to use it. That hurts.
Just to clear up the "more efficient" claims, we are talking about a few percent difference from something like 40% to 100% loading of an inverter. Not a lot.
e.g. Two Reasons You Must Look At Efficiency Curves When Choosing Your Solar Inverter with two great figures;
and more importantly;
Just because you are getting "the most efficient" inverter says nothing about whether that inverter will produce the most power on your panels over a year. (Above the green version is "more efficient", but it will produce less overall power than the red version over normal load conditions.)
The bit that does not get enough air time, in my opinion, is that all inverters become less efficient as they heat up. That alone argues for having your standalone inverter spend more time at less than 100%. Again, the temperature effect amounts are a few percent to ten percent or so. The effect is much less pronounced for microinverters which are designed for much higher operating temperatures. (But it is still a real effect.)
If you are sweating efficiency, it normally requires a deep dive into the efficacy plots for a given inverter model by power, and by temperature. Different designs lose efficiency in different ways, and I think that modeling the efficiency effects at any given site is not straightforward. I would say that at some point, one pretty rapidly hits the point of diminishing ROI, especially with standalone inverters where the capital cost is high.
I think that if I were dependent upon a stand alone inverter, I would own a spare. The old adage that for critical items one equals zero, and two equals one applies to standalone inverters in my book. Or have a standby generator to be able to step in and cover for you.
All the best,
BG
The rest of my story. I had 2 Honda 2k's that I could connect to get 30a. Of course I gave them to my father in law who lives 600 miles away. When it became apparent that the power outage could last a while, I went to home depot and braved the long line at 6a to get a generator. Rounding the corner to my house, I stopped and talked to a neighbor, who said he had an extra generator I could borrow! Following Ian, it was pure comedy concerning my 1st world problems. The Lord took care of the big stuff. Hose and roof made it though 6+ hours of 140+mph winds with a wind direction that abated any surge flooding. Blessed.
That's a good article. I've never seen an efficiency curve for a Tesla inverter. From the PW+ specs - "Solar Generation CEC Efficiency 97.5% at 208 V, 98.0% at 240 V". Assume this is at 100% load. No mention of ambient temp and associated drop in efficiency. What is the MTBF as a function of inverter load?
Efficiency gains would be hard to offset clipping loss. No clipping now and I think I gain ~15% daily in output. Probably not enough to justify cost, but redundancy and reliability increases are hard to quantify, but not zero. A little bit of future proofing too, should I need to add more panels.
When designing systems, estimates are based on NMOT irradiance 800 w/m2? STC is 1000. In S. Fla, irradiance can and does exceed STC. This compounds the clipping.
Regards.
Similar threads
