Acceptable Amount of Clipping

[Becky]

We have a 5.44KWp system that was installed last year. We have a power limit on our inverter at 4KWh, which in practice stops output around 3.8 or so.

I live in a very sunny part of California with few cloudy days. Our system faces south (a smidge towards southwest) with largely unobstructed views to the sun most of the day.

As I really dig into this, I see the clipping happens on a near daily basis from 11AM - 3PM most days from March through June. I've tried to read up on this, ideally I wanted to find a table or chart somewhere that matches system size with inverter limits, but it seems like there is a lot that goes into this and it's not an easy calculation. I understand clipping on a high level, and I get why it makes sense, but when I try to apply those explanations to our setup it doesn't seem correct to me. We have significant daily clipping for four months a year.

Given this, can anyone say whether this is to be expected? I'm feeling like I'm wasting money on panels that aren't being used to their full capacity.

Please let me know if you need any additional information, I included a screen shot of a typical output week from SolarEdge monitoring. I just don't know how to calculate this myself, and unfortunately do not trust our installers. I've honestly tried to find the answers myself, but I'm at a loss.

Thank you!
Output.JPG
Becky

[Ampster]

I have a 5.7 kW system with probably the same inverter. Mine is 15 months old and I see some clipping also. There are various theories that support any conclusion you want to arrive at. I am happy with my system but I did make the installer run some simulations if I had a larger inverter and the results were right now I may be losing 400 kWhr per year. That amounts to $160 per year and will fall off as the panel production degrades over time. FWIW my AC to DC ratio is 1.53 to 1 and yours is 1.42 to 1. Solaredge does not recommend a ratio above 1.55 to 1 for the HD Wave series of inverters. Is yours an HD Wave inverter?

[ButchDeal]

Is all your solar facing the same azimuth?

I wouldn't install a 5.44kw solar array in a 3.8kw ( SE3800H) inverter.
The difference in cost from the 3.8kw to 5kw inverter is less than $150, so why do it.

[Becky]

Some additional information, sorry, should have included the inverter info in the original post.

Our inverter is a HD Wave SE6000H, this 4.0 limit is a setting as far as I can tell, not actually a true limitation from the hardware. I found it scrolling through the menu on the inverter itself and it seems the most likely explanation for why our 5.4 system never generated more than ~3.8. That is probably my biggest point of confusion, this limit was assigned and I don't understand why.

All panels are facing the same way at the same angle.

[Ampster]

Some additional information, sorry, should have included the inverter info in the original post.

Our inverter is a HD Wave SE6000H, this 4.0 limit is a setting as far as I can tell, not actually a true limitation from the hardware. I found it scrolling through the menu on the inverter itself and it seems the most likely explanation for why our 5.4 system never generated more than ~3.8. That is probably my biggest point of confusion, this limit was assigned and I don't understand why.

All panels are facing the same way at the same angle.

You should contact your installer and ask them why. Could it be a limitation imposed by your power company? I have never heard of that being done before your post. Based on your inverter of 6000 and system size your AC to DC ratio is less than one to one.

[discodanman45]

This is strange. Just like Ampster said, contact the installer. Limiting your output to 4.0 kW does not make any sense.

[Becky]

I have, and they've brushed off my concerns saying that clipping is normal and a system never always achieves peak output. I have a call with them tomorrow, I'm just trying to see if my concerns are legit before I keep pressing them. I don't want to be the jerk here if this really is normal. It's a very uncomfortable place to be as a consumer when they can just spew technical jargon at you and I feel like we're at a disadvantage here in terms of information. I just know something about it doesn't feel right to me.

[J.P.M.]

I agree with Ampster that it's a smart idea to contact your installer soon.

Before you do that, I'd answer the following questions and then ask the same questions of the installer (For a lot of reasons, it's always better to know the answer to the questions you're asking in situations like this) :

1.) What is the actual size/model of the inverter ?

2.) If the inverter is a 6000H, why was it oversized compared to the 5.44 STC kW of panels installed ?

3.) If the inverter is a 3800, was it undersized ?

Looks to me you didn't waste money on extra array size. If an array is correctly sized to a load, that's the priority. The inverter sizing follows and is usually dome to make the most use of the array's output. Looks like somehow you don't have enough inverter for the array you have. Sort of like having a governor to limit the output of a racecar - unless maybe you have more than one array orientation such as one facing east and one facing west for example.

As for a way to estimate clipping: Because of the variation in the weather day/day, there is no way to get what's probably no better than +/- ~ 10 % SWAG on clipping on an annual basis, and probably not much better than +/- 20 to 30% for any one month.

However, there is a way to get a SWAG of some sort.
- Run PVWatts with the correct orientation for your array, 10 % system losses and the correct AC to DC ratios. If you're not familiar w/PVWatts, spend 20 minutes reading all the help/info screens before you do any runs. It's a fast learning curve. For confirmation of your input, holler back with your zip and array orientation. I and/or others here can confirm. Speaking only for myself, I won't do the work for you, but I will check it.
- Take the hourly output option. (and don't be intimidated by the spreadsheet size - it's (24 * 365) = 8,760 rows, but rather easy to handle.

- Now for the trick: Subtract each hour's modeled output from your observed max. hourly output, and put the result in another column. Set the column to dump the positive values.
Basically, subtract the model's output from that highest ever hourly value and keep only the negative values.
For example, if the model shows an hourly value of, say, 3,900 W for any particular hour, and your max. observed hourly value since startup is, say, 3,800 W, that difference is 3,800W - 3,900W = - 100W.
- Then, sum all the negative values in that difference column.
- The total of that difference column is a SWAG at annual clipping for your system as the PVWatts model knows about your system.

The method is no good at all for specific hours and maybe good +/- 30% or so for any 30 day period, but might be maybe +/- 10 % or so for yearly total clipping, and so, in that limited sense, may have some use. That a lot of clipping probably takes place for a couple/three months/year will make even that 10 % hard to achieve.

There may be other tricks you can pull such as changing AC to DC ratios and maybe winding up with some possible explanation/thoughts on what the inverter you have may be doing and/or why.

Good luck.

[discodanman45]

I have, and they've brushed off my concerns saying that clipping is normal and a system never always achieves peak output. I have a call with them tomorrow, I'm just trying to see if my concerns are legit before I keep pressing them. I don't want to be the jerk here if this really is normal. It's a very uncomfortable place to be as a consumer when they can just spew technical jargon at you and I feel like we're at a disadvantage here in terms of information. I just know something about it doesn't feel right to me.

Clipping is normal if you had a SE3800 inverter with your system. If they are indeed limiting your SE6000 to 4.0 kW, this is unnecessarily clipping due to them limiting your inverter. You should have no clipping if you have a SE6000 inverter with your system. I would ask them how does a 6 kW inverter clip if you only have 5.44 kW of panels. Your inverter should never clip, unless it is a perfect cool windy day, with your setup. Can you verify from your inverter that they actually installed a SE6000?

[bcroe]

NO WAY should a 6KW net metering inverter clip at 4KW. My 7.5KW clip at about 7.6KW.
Bruce Roe

[inetdog]

NO WAY should a 6KW net metering inverter clip at 4KW. My 7.5KW clip at about 7.6KW.
Bruce Roe

Unless, of course, it was programmed to, as is the case here.
The principal question is why that setting was made.

Who knows? Maybe if the installer only billed for an SE3800, he decided to limit the 6000 to the same output.
More information is needed.

[Ampster]

Unless, of course, it was programmed to, as is the case here.
The principal question is why that setting was made.

Who knows? Maybe if the installer only billed for an SE3800, he decided to limit the 6000 to the same output.
More information is needed.

Not a bad strategy if there was room on the roof for additional panels. I wish my installer had done that. Another panel in my system would have put me over the max AC to DC ratio and the Solaredge configuration software would have suggested the 6000kW inverter.

[J.P.M.]

Not a bad strategy if there was room on the roof for additional panels. I wish my istallee had done that. Another panel in my system would have put me over the max AC to DC ratio and the Solaredge configuration software would have suggested the 6000kW inverter.

If a system is sized to meet an annual load in some way that makes sense, at least to the party paying for the system and the electric bills, and all other things being equal, why would it be an advantage to, in effect, undersize an inverter ? If you're going to intentionally clip output at the inverter, you'll wind up with less annual output and a flatter top more often for graphs of daily clear day output.

I supposed I could conjure up scenarios where that may be an advantage or perhaps necessary, but as Inetdog writes, seems to me more information is needed from the OP. Something doesn't seem correct or at least common.

[Ampster]

If a system is sized to meet an annual load in some way that makes sense, at least to the party paying for the system and the electric bills, and all other things being equal, why would it be an advantage to, in effect, undersize an inverter ? If you're going to intentionally clip output at the inverter, you'll wind up with less annual output and a flatter top more often for graphs of daily clear day output.

I supposed I could conjure up scenarios where that may be an advantage or perhaps necessary, but as Inetdog writes, seems to me more information is needed from the OP. Something doesn't seem correct or at least common.

I agree that aggressive DC to AC ratios don't make any sense for panels all facing in the same direction. I was referring to the concept of installing a 6.0 kW inverter but only charging for a 3.8kW inverter and later offering additional panels if there were room on the roof. We need more facts to really understand the thinking of the installer.

In my case the design engineer literally followed the Solaredge design software that said they could run a 5.7kW panel system on a 3.8kW inverter. I complained and got a $1000 reduction in my cost from the sales manager. The design engineer is no longer employed at the company. In my mind they missed a sales opportunity to get another $1,100 in revenue because if they had gone to a 6 kW system of panels their cost for hardware may have gone up $400 and the incremental cost of labor would have been small since they were already on the site. Their bid included an upgrade to the electrical panel to 200 Amps and they knew I had two EVs and was planning on adding additional electrical appliances..

[J.P.M.]

I agree that aggressive DC to AC ratios don't make any sense for panels all facing in the same direction. I was referring to the concept of installing a 6.0 kW inverter but only charging for a 3.8kW inverter and later offering additional panels if there were room on the roof. We need more facts to really understand the thinking of the installer.

In my case the design engineer literally followed the Solaredge design software that said they could run a 5.7kW panel system on a 3.8kW inverter. I complained and got a $1000 reduction in my cost from the sales manager. The design engineer is no longer employed at the company. In my mind they missed a sales opportunity to get another $1,100 in revenue because if they had gone to a 6 kW system of panels their cost for hardware may have gone up $400 and the incremental cost of labor would have been small since they were already on the site. Their bid included an upgrade to the electrical panel to 200 Amps and they knew I had two EVs and was planning on adding additional electrical appliances..

Well, at least we agree on something.

Without bustin' on the OP too much, I'm wondering why a trip to the inverter and a read/picture of the nameplate on the inverter can't be done and maybe end some speculation or provide more info as to what's going on.