SolarEdge Overheating/Throttling

[soby]

My 10.075kW system with a StorEdge SE7600A Inverter and LG Chem RESU10H battery has been up and running for a few weeks now. I was able to get admin rights from my installer and setup a custom TOU charging profile that charges from clipped energy only. This has been wonderful because my system would routinely clip at 8kW AC before the battery was functional and now it can harvest well over that from the panels. I've seen it as high as 12.5kW which I didn't think would be possible.

I've noticed that on recent sunny days, it hasn't been able to charge the battery as much as it did a month ago. I believe the reason is due to overheating of the inverter. Today when I checked the inverter at peak production, the internal temperature was 142degF. The operating conditions specify up to 140degF so it would seem that the inverter hit a limit and was throttling itself to keep that temperature from going any higher.

The inverter is located in my relatively cool (65degF) basement so I am thinking about adding temperature controlled cooling fans to pull air over the fins on the rear of the inverter against the wall. This is what I have in mind: https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1. I've previously used it on my computer desk to cool a desktop PC inside a cabinet. It comes with a configurable controller and temperature probe so I can set it to only come on when the inverter heats up.

Has anyone else had issues with inverter overheating at max capacity?

[emartin00]

It could simply be that the heat of the sun is heating the panels more, and decreasing their output. Solar panel efficiency decreases as the temperature increases.
It wouldn't hurt anything to add a fan, but I can't really say how much it will help.

[J.P.M.]

I'd start with the installer and ask them. Perhaps there is something about the installation that is amiss and is causing the overheating. Or, it may be something as simple as a bad temp. sensor or monitoring system quirks/setup. Besides - still under warranty, right ? Has the inverter shut down due to heat ?

If I had to SWAG, I'd look to poor air circulation around the inverter which sounds like it's located in an area that doesn't get much or necessary air flow. Before you call the installer, read/check the inverter installation instructions, particularly with respect to clearance requirements and location.

As for adding a cooling fan, if it's called for, a fan needs to be sized for the amount of heat that needs to be removed. Bottom line: Big loads = big fans.That fan you show is most likely not capable of removing more than a small fraction of the waste heat generated by an inverter.

Also, adding a fan can sometimes impair cooling efficiency over time if the increased forced convection causes things like dust/stuff to get into places, coat/muck up surfaces and reduce cooling efficiency more or most likely sooner than if natural convection were left to its own devices.

Depending on how it's done, adding aux. cooling may also void a warranty if an installer/mfg. is looking for way out.

I'd start with the installer, but do my homework with the inverter manual. If nothing else, reading/studying the manual may help reveal to you what the installer does/doesn't know or may be trying to B.S. you about. I learn a lot by playing dumb, but trying to know the answers to my questions before I ask - 1.) I learn technical stuff, 2.) I sometimes learn a vendor is honest or is trying to B.S. me. Both are useful.

[Ampster]

I am not sure I understand why the inverter could put out more power because the charging load was internal. That load and the other loads (grid or house) are all on the output side of the inverter. That is a selling point for that integrated inverter/EVSE. I spent half a day hooking up a CT, relay and programmable switch to get my EVSE to charge when solar is producing.

[J.P.M.]

I am not sure I understand why the inverter could put out more power because the charging load was internal. That load and the other loads (grid or house) are all on the output side of the inverter. That is a selling point for that integrated inverter/EVSE. I spent half a day hooking up a CT, relay and programmable switch to get my EVSE to charge when solar is producing.

I was wondering that as well, but not being there, and with what's probably incomplete information presented so far, and the system probably still being under warranty, I'd think - after educating/reading up in the manual - the installer would be the best place to start.

[soby]

I'm essentially on my own. I don't have an installer labor warranty. Long story short: my installer sucked and never finished the job so I had a second installer finish. The second installer did not provide a labor warranty and their quote reflected that.

I setup a simple tower fan in front of the inverter to see if cooling it down made a difference today. It did not. The temperature dropped from 140degF down to 125degF but there was no change to production. I still plan to install a permanent temperature activated fan because lowering the temperature will extend the inverter's life.

IMG_0234.JPEG

When I refer to "charge from clipped energy" I mean that the DC coupled battery is being charged only by the energy that would be clipped due to the 8kW AC output limitation. If the solar production is 10kW DC, the inverter will export 8kW AC and charge the battery with the remaining 2kW DC. The chart below illustrates the clipping. Solar Production is not clipped and reaches ~9kW while actual System Production is clipped. The difference between the two green areas is the energy used to charge the battery and this only gets counted later toward System Production when the battery discharges (see the dark green area early in the morning). Note that I have a battery backup reserve of 50% so the battery only discharges down to 50%.

Plot.JPG

[Ampster]

Makes sense. I thought you were talking about charging the car. You were talking about the DC coupled battery. Thanks for the explanation.

[pete4]

No matter what, you're running at the inverter power limits and if I was you and I was on my own as you said I would definitely put fan on it to keep it as cool as possible. I have standard 7600 in a garage and right now I still have freaking winter here, temp are in 60's day, 40s at night, but in few weeks it will get hot so I'm putting fan in a garage to keep it cooler and fan on the fins to cool inverter directly. Nothing kills electronics faster than overheating and some cheap fan is much better than headache with broken inverter even under warranty.

[J.P.M.]

Soby:

To get the most effect from forced convection it's necessary to get as much air movement over the heat transfer surfaces that cool the heat sink/inverter as possible. From your photo, it looks like the fan is probably not forcing much air through the inverter or over any cooling fins. Circulating air in/around the area probably won't do much besides make some noise.

FWIW, to gauge the effectiveness of any cooling scheme, rather than rely only on an inverter's temp., consider the difference in temp. between the inverter and the surrounding air rather than the inverter temp. by itself gives more/better information.

A comment: The downside of changing a design to get a lot more air past cooling surfaces than the original design may have intended - as I mentioned in a prior post - is that the heat transfer surfaces/fins and other parts of the inverter may collect dust/gunk/stuff quicker and in places where it might not otherwise collect under original design conditions. That will mean more care/attention to cleaning such surfaces with a likelihood of some (inaccessible) areas/parts getting fouled and never cleaned. Too much fouling from too much forced convection may actually be counterproductive to the desired result.

More FWIW, I've got a rebadged 5 kW ABB Power One inverter that looks a lot like yours but not identical with vertical cooling fins on the front face behind a removeable cover. That scheme makes the fins/heat transfer surface accessible for cleaning although I'm pretty sure the designers had other reasons for doing it that way. Because I like to measure stuff, I've fitted a Rube Goldberg system of a 20" box fan below the inverter blowing air up/over/through the fins and find that doing so reduces [(the monitor's reported temp.) - (ambient garage air temp.)] by about 30-35 % or so. So, under full sun at near normal solar incidence angle on the array and array output ~ 4.6 kW or so to the inverter, the inverter to garage temp. diff. with the fan running is usually ~ 18 -19 C. (inverter heat sink temp. ~ 44 C. with garage amb. 25 - 26 C. or so).
Under similar conditions with no fan running, the inverter heat sink to amb. temp. diff will run something like 27 to 28 C., with the inverter temp. being something like maybe 52-54 C. when the amb. air is ~ the same 25 - 26 C. The fan draws ~ 80 running W. I'm pretty sure the inverter' efficiency is not affected in any way due to lowered temps., so that fan draw of 80 W is purely parasitic. I do suspect (but not much more than that) it may add some longevity to the inverter life, but I'd not begin to hazard a guess as to how much. Anyway, I chalk that off as the cost of acquiring information.

Take what you want of the above. Scrap the rest.

[solardreamer]

Can you take the inverter cover off just for testing purposes?

[soby]

Can you take the inverter cover off just for testing purposes?

I wouldn't dare have the thing running with the cover off after reading the warnings in the manual. Those high voltage capacitors will kill you quick. There is a tiny fan inside the cover to circulate the air internally but it's a sealed enclosure and the only way for heat to escape is via convection through the aluminum body and fins on back.

Soby:

To get the most effect from forced convection it's necessary to get as much air movement over the heat transfer surfaces that cool the heat sink/inverter as possible. From your photo, it looks like the fan is probably not forcing much air through the inverter or over any cooling fins. Circulating air in/around the area probably won't do much besides make some noise.

FWIW, to gauge the effectiveness of any cooling scheme, rather than rely only on an inverter's temp., consider the difference in temp. between the inverter and the surrounding air rather than the inverter temp. by itself gives more/better information.

A comment: The downside of changing a design to get a lot more air past cooling surfaces than the original design may have intended - as I mentioned in a prior post - is that the heat transfer surfaces/fins and other parts of the inverter may collect dust/gunk/stuff quicker and in places where it might not otherwise collect under original design conditions. That will mean more care/attention to cleaning such surfaces with a likelihood of some (inaccessible) areas/parts getting fouled and never cleaned. Too much fouling from too much forced convection may actually be counterproductive to the desired result.

More FWIW, I've got a rebadged 5 kW ABB Power One inverter that looks a lot like yours but not identical with vertical cooling fins on the front face behind a removeable cover. That scheme makes the fins/heat transfer surface accessible for cleaning although I'm pretty sure the designers had other reasons for doing it that way. Because I like to measure stuff, I've fitted a Rube Goldberg system of a 20" box fan below the inverter blowing air up/over/through the fins and find that doing so reduces [(the monitor's reported temp.) - (ambient garage air temp.)] by about 30-35 % or so. So, under full sun at near normal solar incidence angle on the array and array output ~ 4.6 kW or so to the inverter, the inverter to garage temp. diff. with the fan running is usually ~ 18 -19 C. (inverter heat sink temp. ~ 44 C. with garage amb. 25 - 26 C. or so).
Under similar conditions with no fan running, the inverter heat sink to amb. temp. diff will run something like 27 to 28 C., with the inverter temp. being something like maybe 52-54 C. when the amb. air is ~ the same 25 - 26 C. The fan draws ~ 80 running W. I'm pretty sure the inverter' efficiency is not affected in any way due to lowered temps., so that fan draw of 80 W is purely parasitic. I do suspect (but not much more than that) it may add some longevity to the inverter life, but I'd not begin to hazard a guess as to how much. Anyway, I chalk that off as the cost of acquiring information.

Take what you want of the above. Scrap the rest.

I have a temperature monitor in the basement and I know it fluctuates very little ~58F winter to ~65F summer. I didn't think the inverter would have any issues in such an environment but the complete lack of air circulation in the basement is allowing the inverter to just heat up the air around it. The air immediately above the inverter is quite warm and has nowhere to go. If the inverter were outside, there would be nothing overhead and, even on a hot summer day, the breeze would remove heated air and supply fresh air.

I think that is why my simple tower fan test made such a difference by simply moving the air around in the basement. By blowing the warm air away and introducing cool air into the mix, the natural convection was more effective.

I'm going to build a fan housing out of some plywood to sit on top of the inverter and suck air up through the fins on the back. The fan assembly I intend to use only draws 3W so there wont be a significant parasitic load and it can be programmed to only activate when the air in the fins hits a certain temp. J.P.M is right about dust and I will have to set a reminder to run a Swiffer through the fins once a year when I change HVAC air filters.

I'll post pics whenever I get the time to make the baffle.

[J.P.M.]

Can you take the inverter cover off just for testing purposes?

If that question is directed at me, the cover I'm referring to is the one that only covers the cooling fins. It is held on with 4 easily removed screws. Removing it allows access to the heat sink cooling fins. It can be easily removed for inspection/access to the fin surfaces. I've done so several times, ~ 1X/yr. for the first couple of years and infrequently thereafter as the cooling scheme I use doesn't seem to cause much fouling of the fin surfaces, and being in a semi-controlled environment of the garage may be a bit less dusty than being outside.

[soby]

Over a 40degF difference with the new fan box running at peak production (down from the operating max at 140degF to about 100degF). It drops to just under 70degF when idle at night.

I ended up just getting the version of the fan box without a temperature controller. The added complexity (and cost) wasn't worth the energy savings of having it shutoff when off-peak. Cooler is better...

2020-05-25 21.46.58.jpg

[J.P.M.]

Over a 40degF difference with the new fan box running at peak production (down from the operating max at 140degF to about 100degF). It drops to just under 70degF when idle at night.

I ended up just getting the version of the fan box without a temperature controller. The added complexity (and cost) wasn't worth the energy savings of having it shutoff when off-peak. Cooler is better...

2020-05-25 21.46.58.jpg

Thanx for the info. Have the lower inverter temps. affected performance in any way(s) ?

[soby]

Thanx for the info. Have the lower inverter temps. affected performance in any way(s) ?

It’s impossible to tell given the thick layer of pollen over each panel. I doubt it boosts performance but I’m content just increasing the inverter’s useful life.