I'm back ... where is roof mount pricing nowadays in San Diego area

[J.P.M.]

There are tow problems with your anecdotal analysis:
1) solaredge has the largest market share especially for the past few years, due to NEC2017 and a lot of new installers that never bothered to read manuals or get training.

2) primarily people without problems do not bother to post In a tech forum, and corollary, string inverters often have no monitoring so new customers do not know of problems.

End result is that there is a larger percentage of new solaredge customers, new solaredge installers ( solaredge is easier to install than strings but has different rules for string length), and mostly people with problems posting means that there could be less problems per install with solaredge systems but you still see more solaredge problem posts than other inverters at the same time.

I agree with some of what you're writing but not all of it.

If the traffic around this forum is any indication, a lot of new installers don't seem to have as many problems with other inverter equipment.

I won't try to separate fly crap from pepper because I have no numbers to back it up but I'd respectfully suggest the number/proportion of SolarEdge related problems seems higher to me than their market share would suggest. Maybe I'll keep an informal tally of posts of inverter problems for 30 days or so and see what turns up.

To the degree that reported problems w/SolarEdge stuff are out of proportion to their market share is a reasonably correct statement, it might also suggest a couple of other things:
1.) Installers may find other equipment less complicated/troublesome to install, or the SolarEdge manuals are less understandable.
2.) SolarEdge support is not as good as the support provided by other inverter mfg.

Solar edge might also consider being a bit more selective about who they allow to install their equipment, or maybe make training mandatory. On the other hand, if it's that complicated, maybe that's part of the reason for all the complaints. KISS.

The reality I've seen in my HOA on the ~ 150 or so installs is that no one, and I mean no one, except me bothers with two things:
1.) Cleaning their array.
2.) Looking at or even being aware of any system monitoring much less knowing what to look for.

To the degree my observations about user non involvement are a reasonable reflection of reality, and if folks in my HOA are somewhat representative of the rest of the great unwashed masses, it would seem that the main advantage of panel level monitoring is wasted. Seems a lot of complication that's contrary to the KISS concept for little actual benefit.

To your point #2 : Sometimes but not always. I remember in years past on this forum and still, how many happy owners of Sunpower equipment where praising that product to the heavens. They were in general happier than slopped hogs. Cost ineffectiveness aside, S.P. equipment is good stuff and while a problem appeared 1X/awhile, more posts about S.P. stuff seemed favorable than unfavorable, even if uninformed.

[ButchDeal]

I sort of agree, but not having had any dealings with the SolarEdge folks I can only go by what shows up in my HOA and as a larger sample, what shows up in these threads. As to what shows up on the forum, seems to me there's more written about SolarEdge equipment than other inverter equipment and anecdotally it seem to me most of that is not favorable.

There are tow problems with your anecdotal analysis:
1) solaredge has the largest market share especially for the past few years, due to NEC2017 and a lot of new installers that never bothered to read manuals or get training.

2) primarily people without problems do not bother to post In a tech forum, and corollary, string inverters often have no monitoring so new customers do not know of problems.

End result is that there is a larger percentage of new solaredge customers, new solaredge installers ( solaredge is easier to install than strings but has different rules for string length), and mostly people with problems posting means that there could be less problems per install with solaredge systems but you still see more solaredge problem posts than other inverters at the same time.

[J.P.M.]

I love this thread, whats I take from it is, kinda splitting hairs over performance versus cost, to me I dont like Solar Edge, their service department or their optimisers, the more optimisers we have in the world is more money its going to cost us in time because we have to pay our guys to go fix their crap..............

I sort of agree, but not having had any dealings with the SolarEdge folks I can only go by what shows up in my HOA and as a larger sample, what shows up in these threads. As to what shows up on the forum, seems to me there's more written about SolarEdge equipment than other inverter equipment and anecdotally it seem to me most of that is not favorable.

As an opinion only, if more folks knew more about the limitations of PV beyond the mostly B.S. bill of goods most swallow from peddlers and their green mafia shills, there would be fewer arrays on residential property and those that did exist would be, as a group, more favorably oriented in less shade, thus eliminating most of the advantage and thus the need for more complicated and more expensive solutions using micros and optimizers on what may well be marginal applications to begin with.

Seems to me the more complicated (read less reliable, more $$) micro or optimizer solutions for shade or what IMO amounts to a red herring with respect to panel level monitoring amount to putting lipstick on the pig of crappy applications or situations. More informed users would get rid of a fair amount of that.

[solar pete]

I love this thread, whats I take from it is, kinda splitting hairs over performance versus cost, to me I dont like Solar Edge, their service department or their optimisers, the more optimisers we have in the world is more money its going to cost us in time because we have to pay our guys to go fix their crap..............

[bcroe]

From some distance I view optimizers as having several purposes.

1. Get the most power possible from a string when some cells or panels are shaded. This is
partially done by letting all unshaded panels operate at their best MPPT instead of increasing
voltage to keep up with other unshaded strings. And the shaded panel may put out some
power if any cells are unshaded.

2. Set the total string voltage to a standard value the inverter DC input needs.

3. Provide simplified array wiring and efficient high voltage like a simple string.

4. Provide individual panel monitoring and control.

At this location 2, 3, and 4 are of no advantage over my simple string. That leaves the question
of shading. Close by small shadows (vent pipe) do not occur. If they did, I would expect keeping
panels away from them would be better than going for a optimizer system. Big shadows (sun
behind tree) do occur at some times. But a string is grouped so that a shadow on one part may
soon sweep across most of it. The effect tends to happen at day extremes, when shadows are
moving faster. So the time that an optimizer can do much is rather limited.

Sometimes I think of a DIY optimizer to bring half a string to max efficiency near sunset could
be useful. But it will probably never happen. Bruce Roe

[nwdiver]

Butch, I can't fault you for leading the horse to water.

If I had spent $4k on optimizers for my 22.77kW ground mount... what's the return on that $4k? 30 years? If I'm adding optimizers when adding panels yields more kWh/$ how is that serving my customers? Our jobs should be the lowest cost and most reliable kWh. Most people could not care less about module level monitoring.

[Mike90250]

Butch, I can't fault you for leading the horse to water.

[nwdiver]

I highly disagree and so do most of the documentation and examples. One thing you seem to muss is that it takes a lot of shadow to get the diode to kick in, up to that point the whole module and string would be effected in a string. Then after the diode kicks in, you most of the time have other cells in other parts of the pv module effected by the shadow as well, thus still effecting the string output.

Here is a nice write up from tigo even on comparison of bypass diodes to optimizers but results would be better with solaredge optimizers.
https://support.tigoenergy.com/hc/en...mizers-better-

In my recommendations and sites we have built yes.

I learned about bypass diodes the hard way. The first solar array I built was using laminates when I knew very little about solar. The laminates didn't include a J-Box or bypass diodes. It takes very little shade to trigger the bypass diode. Since I installed ~45 bypass diodes by hand I was able to test them and see for myself how much shade it takes to trigger them... its not much. 1 fully shaded cell will trigger the diode whether it's a string of 12 panels or just 1 panel.

'Shaded cells have reduced output current, which affects all cells in series unless optimized. Tigo optimizers allow string current to partially bypass shaded modules, which mitigates the effect of shaded modules on system production.'

Even in an optimized panel the 3 cell groups are in series. If one cell group is shaded and can only output 2A but the other 2 cell groups are unshaded and are outputting 8A the bypass diode will be triggered for the shaded cells and that cell group will be deleted. This effect will be the same for a string or optimizer. You have 1 MPP for that panel. 1 String in that panel. The current through all cells MUST be the same. It cannot be 8A AND 2A... Sure... if ALL cell groups are partially shaded and can only output 2A then that panel can contribute what... (2A)(30v) ~60w? Is it really worth the expense of an optimizer to harvest ~60w for the short period of time that ALL cell groups are shaded?

The 22.77kW array I installed would have required 66 optimizers costing ~$4k. It's currently producing ~120kWh/day on average. How much more production do you think I would get if I had chosen optimizers?

But... I'm installing my first SE system next week.... will be interesting to see how its performance compares to all my string systems... Still having sticker shock, paid ~$2840 for (2) 7.7kW SMA inverters for the 22.77kW ground mount. >$5 for a 11.4kW HD-Wave inverter plus 48 optimizers

[ButchDeal]

Yeah... shadows move... it's not like once triggered bypass diodes stay triggered. Once the shadow moves the array adjusts to the new location. My point is that a shaded cell group is going to effect a panel on a string inverter very similar to if there was an optimizer. The only time this would be different is if ALL cell groups are shaded, in this case the panel would be completely removed by a string but if optimized would still produce whatever a completely shaded panel can produce (not much).

I highly disagree and so do most of the documentation and examples. One thing you seem to muss is that it takes a lot of shadow to get the diode to kick in, up to that point the whole module and string would be effected in a string. Then after the diode kicks in, you most of the time have other cells in other parts of the pv module effected by the shadow as well, thus still effecting the string output.

Here is a nice write up from tigo even on comparison of bypass diodes to optimizers but results would be better with solaredge optimizers.

https://support.tigoenergy.com/hc/en-us/articles/211327298-Don-t-bypass-diodes-reduce-the-effects-of-shading-Why-are-optimizers-better-

Even if rapid shutdown isn't required such as for a ground mount?

In my recommendations and sites we have built yes.

[nwdiver]

thats the point, shadows move. There are few installs with static shadows throughout the day. cover part of a cell group and the diod is not likely to kick in, you need to cover a lot of the cell group.
And further diodes are not for optimization but to protect the cells. They do result is a very rudimentary shadow mitigation but just that VERY RUDIMENTARY shadow mitigation.
Optimizers and in particularly module level MPPT which you get with micros and SolarEdge optimizers, gives as much as each PV module can give in almost all conditions.

As for worth the cost, well your cost estimates are slanted but that is a different assessment. I believe module level monitoring and efficiency as well as NEC2017 requirements make them a good choice in all conditions ( I am not alone but also not unanimous in this). Few if any believe that string optimization is better in shadow conditions or even close.

Yeah... shadows move... it's not like once triggered bypass diodes stay triggered. Once the shadow moves the array adjusts to the new location. My point is that a shaded cell group is going to effect a panel on a string inverter very similar to if there was an optimizer. The only time this would be different is if ALL cell groups are shaded, in this case the panel would be completely removed by a string but if optimized would still produce whatever a completely shaded panel can produce (not much).

make them a good choice in all conditions

Even if rapid shutdown isn't required such as for a ground mount?

[ButchDeal]

That depends on how the shadow falls. If only one cell group is shaded then they would still producing even with a string inverter. Most panels are divided into 3 cell groups.

thats the point, shadows move. There are few installs with static shadows throughout the day. cover part of a cell group and the diod is not likely to kick in, you need to cover a lot of the cell group.
And further diodes are not for optimization but to protect the cells. They do result is a very rudimentary shadow mitigation but just that VERY RUDIMENTARY shadow mitigation.
Optimizers and in particularly module level MPPT which you get with micros and SolarEdge optimizers, gives as much as each PV module can give in almost all conditions.

As for worth the cost, well your cost estimates are slanted but that is a different assessment. I believe module level monitoring and efficiency as well as NEC2017 requirements make them a good choice in all conditions ( I am not alone but also not unanimous in this). Few if any believe that string optimization is better in shadow conditions or even close.

[nwdiver]

NO No, it isn't See you are doing STRING level MPPT NOT module level MPPT. SolarEdge does Module level MPPT thus every module is operating at its MPP, not the combined string MPP..

Except Tigo doesn't do Module level MPPT and string MPPT doesn't match module MPPT when there are shadows and shadows tend to move around..

??? So what exactly is a TS4-O/L doing that the TS4-S isn't?

What system is NOT using MPPT at string or module? Why do you think string level MPPT is better than module level for shadows?

.... I don't think that string level is better than module level. I agree that you get more energy yield with module level MPP. My point is that the gain is no where near worth the cost. Instead of adding $1k to the cost of a system with module level MPP.... just add ~500w in panels. You'll get >4x more energy than if you'd spent the $1k on optimizers.


Adding to the reduced benefit of optimizers is that due to lower panel cost it makes more and more economic sense to over-size PV systems. My last install was limited by the service size. My current install is also limited by the service size. Optimizers don't help a saturated inverter...

One of these days I'd love to see a true side-by-side of string level vs module level optimization. What's the annual yield increase with module level with an oversize ratio of 1.4? Would it even be 1%? Optimizers would have added ~15% to my cost for this 22.77kW Ground Mount. Is 15% worth a ~1% increase in annual production? IMO.... no.

Screen Shot 2019-06-21 at 2.48.28 PM.png

The two in shadow would not be able to keep up at string level MPPT but with module level they are putting out 66w and 95w still.

That depends on how the shadow falls. If only one cell group is shaded then they would still producing even with a string inverter. Most panels are divided into 3 cell groups.


Here's an example of shaded panels still producing with a string inverter because the shade is only affecting one cell group. That cell group would still be 'deleted' with an optimizer since those sections are causing a drop not a rise in voltage.

Screen Shot 2019-06-21 at 3.04.08 PM.png

[ButchDeal]

Tigo has both. The 'L' is used if you want longer strings (buck) and the 'O' is used for all other optimization (boost). The string MPP is irrelevant.

NO No, it isn't See you are doing STRING level MPPT NOT module level MPPT. SolarEdge does Module level MPPT thus every module is operating at its MPP, not the combined string MPP..

The purpose of the Tigo optimizer is to allow an individual panel to operate at its MPP. The point here is that ~99% of the time panel MPP IS string MPP or close enough to not matter.

Except Tigo doesn't do Module level MPPT and string MPPT doesn't match module MPPT when there are shadows and shadows tend to move around..

The only efficiency benefit to optimizers is allowing current to vary in a series circuit. But each panel will usually have multiple MPPs for the same reason strings have multiple MPPs. Sure, only one is the highest at any given point but the difference is a rounding error and likely too small to even be distinguishable by the tracker. Voltage can still vary panel to panel and that 'naturally' follows the MPP for the string current because physics. If string current is 8A but one the MPP of a panel is closer to 8.2A then the panel voltage will be slightly higher with that panel running at 8A.

modules tend to have only ONE MPP with a few others that are close but by definition there can be only ONE Maximum Power Point. Further a string with multiple shadows is going to have a MPP but that will not be the same as every PV module in the strings MPP.

Shading can cause issues but that would be almost completely eliminated with string level MPPT.

What system is NOT using MPPT at string or module? Why do you think string level MPPT is better than module level for shadows?



It seems that you can't get your head out of the string marketting BS. If there are shadows then there will be more power with optimizers than with a string system, you must be the only person claiming that strings can keep up with optimized systems in shadow conditions.

see here is a good example, all one string two modules have shadow from chimney at the instant:
Two PV modules have MPP of 25.88V while two near by have MPP of 31.38V and 31.63V (higher V are in shadow)
The two not in shadow are producing 8.07a and 8.14a while the two IN shadow are producing just 2.09a and 3.03

The two in shadow would not be able to keep up at string level MPPT but with module level they are putting out 66w and 95w still.





Layout.jpg

[nwdiver]

well here is the thing. Tigo optimizers are NOT even close to SolarEdge Optimizers. SolarEdge optimizers have MPPT & Buck/Boost, Tigo does not have MPPT or boost, just buck.

So the ONE that you put a tigo on is not working at ITs best MPP but the strings MPP since there is just one MPPT for the string.

Tigo has both. The 'L' is used if you want longer strings (buck) and the 'O' is used for all other optimization (boost). The string MPP is irrelevant. The purpose of the Tigo optimizer is to allow an individual panel to operate at its MPP. The point here is that ~99% of the time panel MPP IS string MPP or close enough to not matter.

The only efficiency benefit to optimizers is allowing current to vary in a series circuit. But each panel will usually have multiple MPPs for the same reason strings have multiple MPPs. Sure, only one is the highest at any given point but the difference is a rounding error and likely too small to even be distinguishable by the tracker. Voltage can still vary panel to panel and that 'naturally' follows the MPP for the string current because physics. If string current is 8A but one the MPP of a panel is closer to 8.2A then the panel voltage will be slightly higher with that panel running at 8A.

If you watch the current and voltage of a string inverter it's interesting how consistent power output is even with fluctuations in voltage and current.

This is two 11.4 kW arrays located next to each other on an old Fronius string inverter. 4 strings, 1 MPPT. Shading can cause issues but that would be almost completely eliminated with string level MPPT.

Screen Shot 2019-06-21 at 12.15.32 PM.png

[ButchDeal]

I did; The last SMA system I installed I optimized one panel since Tigo allows a mixture; 14 panels with 'shutdowns' and 1 with an optimizer. So that one panel is allowed to operate at its peak MPP. The difference between the 'optimized' panel and the others that just operate with the string is negligible.

well here is the thing. Tigo optimizers are NOT even close to SolarEdge Optimizers. SolarEdge optimizers have MPPT & Buck/Boost, Tigo does not have MPPT or boost, just buck.

So the ONE that you put a tigo on is not working at ITs best MPP but the strings MPP since there is just one MPPT for the string.