SolarEdge users: What's your inverter temp during high production?

[TAZ427]

If the inverter is still pumping out a constant 7.6 kW under high temps, I do think it is pretty safe to say the it is not de-rating much or at all. It is probably putting a huge stress on my inverter, but I don't care. I have the 25 year warranty and I have two companies standing behind the install.

Not adding a larger inverter was done for a few reasons. The first is I would have to spend a few thousand to modify my panel for the larger breaker needed. I also have three separate arrays. The clipping is bad, but it probably is not as much as bad as it looks. I am probably averaging a little over 8 kW during the clipping time, which means that I am missing out on about 4 kWh max on a good solar day. The two thousand dollars to modify my panel at this point wouldn't be a smart financial decision.

When my panels degrade I will not see the same losses as most people do because of the clipping. Plus my output is higher so far than what was expected, even with the crazy clipping. I produced 2,180 kWh in May with the 7600 inverter.

That makes a bit more sense when you're not peaking out on all the panels at the same time. Looks like my two roofs peak out about just over any hour apart, but yours I with the 3 roofs probably has a flatter profile mid day than mine does, and in your circumstances the $2000 cost to upgrade the panel makes the decision that much easier to make. I only had a double breaker spot left in my panel. That's one thing that I have to take into consideration if I get an EV, as I don't have another spot for a breaker and no 240v line in the garage. I'm thinking I'll relocate the laundry rooms 240v plug to the garage side, luckily it's on a wall shared with the garage. If I'm not allowed to do that, then I'll put some conduit in place have the plug plugged in the inside, and the charger sitting on the garage side wall.

[discodanman45]

If the inverter is still pumping out a constant 7.6 kW under high temps, I do think it is pretty safe to say the it is not de-rating much or at all. It is probably putting a huge stress on my inverter, but I don't care. I have the 25 year warranty and I have two companies standing behind the install.

Not adding a larger inverter was done for a few reasons. The first is I would have to spend a few thousand to modify my panel for the larger breaker needed. I also have three separate arrays. The clipping is bad, but it probably is not as much as bad as it looks. I am probably averaging a little over 8 kW during the clipping time, which means that I am missing out on about 4 kWh max on a good solar day. The two thousand dollars to modify my panel at this point wouldn't be a smart financial decision.

When my panels degrade I will not see the same losses as most people do because of the clipping. Plus my output is higher so far than what was expected, even with the crazy clipping. I produced 2,180 kWh in May with the 7600 inverter.

[TAZ427]

I'm not sure if you can draw that conclusion from his production graph. Since he is clipping so much, there's no telling if some derating is or isn't going on. The inverter may indeed be reducing current, but even the derated production may still be beyond the AC limit of his inverter. Any change in production due to temperature might only be noticeable when the inverter is not clipping.

What I find puzzling is that Solaredge provides that 120F/50C ambient temperature for when derating begins, but doesn't state what the internal inverter temperature at which derating occurs is. And it doesn't seem likely that the inverter itself has two separate temp sensors, one for "ambient" and one for internal electronics, so the derating algorithm should be based on the actual internal temperature reading anyway, so why isn't that value given?

I agree that, I don't think one could tell based on that if it's de-rating or not. That said, they (SolarEdge) should be able to calculate what the internal component temps will reach based on the thermal dissipation of the heat sink at a given ambient temp, within reason, based on heat radiation from the heat sink with static air (no air flow.) And when you have a 'canned' solution, this is the preferred way to provide the information to the end user, so that the user doesn't need to do any extra calculation when they have limited knowledge in the field, and also will not be live monitoring the 'temp' provided by the equipment (which I've yet to be able to determine where that temp is measured.) But I would also assume, that this wouldn't take into consideration extra heat disippation during clipping situations, but the derating would effectively become a moot point do to the clipping.

It would be nice to have both the ambient and some internal source point temp value for derating, for those with more knowledge could monitor it, especially if one starts to generate excess heat during stress situations such as clipping.

[library]

I read "derating" as directly affecting the inverter AC output. If he's pumping out the max his inverter can do while sitting at 187F it's not derating at the inverter which is what we're discussing. And yes, if Solaredge would just tell us the magic number the inverter actually derates at we wouldn't be on page 8 of this thread

[Kendalf]

That's great anecdotal evidence that at least at an internal inverter temp of 187 degrees it's not derating.

I'm not sure if you can draw that conclusion from his production graph. Since he is clipping so much, there's no telling if some derating is or isn't going on. The inverter may indeed be reducing current, but even the derated production may still be beyond the AC limit of his inverter. Any change in production due to temperature might only be noticeable when the inverter is not clipping.

What I find puzzling is that Solaredge provides that 120F/50C ambient temperature for when derating begins, but doesn't state what the internal inverter temperature at which derating occurs is. And it doesn't seem likely that the inverter itself has two separate temp sensors, one for "ambient" and one for internal electronics, so the derating algorithm should be based on the actual internal temperature reading anyway, so why isn't that value given?

[TAZ427]

Here is a picture from a day when my inverter got to 187 degrees from the heatsink reading. The inverter was maxed from 9:45 AM until 2:30 PM. As the weather heats up I will see what happens. This was on a day which reached 101 degrees. It will get at least 10 degrees hotter later this summer. Once the inverter does reach over 180 degrees there is a decent hum that comes from the unit.

Going back through posts, it looks like you've got an 11.25kW system, and I'm guessing you've got a something like a SolarEdge SE7600H, based on the clipping. That's looks to be > 4hrs a day through the summer months of clipping. A lot of wasted power, and some apparent stress on the inverter.

Based on comments in another thread, you're on PG&E EV-A plan, and I'd think that's a lot of money down the drain to save < $300 on an inverter (About the retail difference of what I've seen between SE7600H and SE10000H.)

Was this self inflicted or installer suggested? I'd look into getting that inverter swapped out, personally.

[library]

That's great anecdotal evidence that at least at an internal inverter temp of 187 degrees it's not derating. The hum is probably the little tiny fan (assuming you have an HD Wave SE7600). It's a dinky little fan but I guess it's moving some air.

I think I'm done obsessing over it. My panels get too hot so I never hit the inverter max during summer anyway so it's harder for me to check with a nice max graph like you provided.

[discodanman45]

Here is a picture from a day when my inverter got to 187 degrees from the heatsink reading. The inverter was maxed from 9:45 AM until 2:30 PM. As the weather heats up I will see what happens. This was on a day which reached 101 degrees. It will get at least 10 degrees hotter later this summer. Once the inverter does reach over 180 degrees there is a decent hum that comes from the unit.

[Kendalf]

Personally, I'm not worried about longevity at those temps. But if there was a percent boost to power production, by dropping the temp a bit, then I'd grab one of my arduino boards, one of my temp senors, a relay, a couple fans, may even design and 3D print a ducted housing, and have a nice little project. But if it effectively did nothing, well then I'll spend my time on something else, like trying to get the SolarEdge API to work.

I think I have a pretty good setup right now, with the ability to quickly drop the inverter temp or allow it to heat back up again and then track the results. I'm planning on running some tests when we have completely clear skies to see if there is any measurable difference to production if the inverter is cooled with fans versus when it isn't. My plan is to let the system run without the fans until I get to the peak production point of the day, and when the power starts to trend down I will turn the fans on and see if that will cause power production to level off or even go back up slightly. I don't imagine the difference will be much, if any, so it may be hard to catch. But as I said in an earlier post, the effort involved in putting together my setup was minimal, and even if I get just 50 Wh of power extra from having the fans on, that will be more than sufficient to make up for the energy used by the fans each day.

[TAZ427]

My inverter is in the garage, which is much cooler than the outside. However, the air doesn't flow in my garage. Yesterday my inverter got to 187 degrees and it was still pumping out the max 7.6 kW limit of the inverter. I will probably find out what temps affect the inverter soon and will let everyone know. If I can hit 187 degrees now, 200 plus will happen soon as the 110 constant outdoor temps hit my area.

Their Ambient temps for de-rating would have to be done at static air flow. Any air movement would be do to natural convection of the radiant heat from the heat sink heating the air causing it to rise and pull in the cooler ambient from below (yes this is very slow moving, but w/o a forced air system they would have to calculate on this radiant transfer.)

[TAZ427]

That's what we're all trying to determine. Taz427 and discodanman45. Solaredge says at 122 F ambient they start derating. But 122F ambient on the south side of a wall is a lot different from 122F ambient inside your garage. And since they're never going to tell/show us when they're derating the inverter we're all sort of in the dark. Especially since the only temp we can see is the inverter heatsink temp.

My wild guess is that anything below 180F on inverter temps is probably fine. I base that only on my basic knowledge of integrated circuits and their common operating temperatures. Most will run up to 85-90C just fine. And just to reassure discodanman45 the temp you saw is in no way a fire hazard. No one freaks out when their computer hits that temperature which it probably does multiple times a day.

From my reading of the thread so far people have certainly reduced their temps with a fan but it's made no discernible difference in inverter output. Which is really the answer we're seeking. What inverter temp does it start derating at.

Yeah, while I've seen the question somewhat implied, I hadn't seen it directly asked, and there's a bunch of ho/hum and nothing definitive being stated beyond what they're able to get the temps (either inside their Inverters or at the heat sink) down to.

I was reading the de-rating of the temps (which the SE7600H has a steeper current de-rating but starts at 50C/122F than most of their other Inverters.) But that as you mentioned is the Ambient temp. Over the past 2.5yrs+ of living where I'm at, the garage hasn't exceed 88F (even w/ 110F days.) It helps that there's a floor above it, interior wall to the South (and East), Tree shading during most of the day, and only gets direct sun during the evening hours. It get's more heat from the cars being pulled into the garage of the evening than anything. It may increase a tiny bit w/ the heat radiated from the Inverter now, but still the ambient temps will remain well below the derating point in my case. Also, FWIW, this derating of current delivery is not the same as efficiency drop, you can have an efficiency drop due to temps and never derate the current output.

FWIW, I work in the Semiconductor industry, and yeah the temps are really low from an electronics stand point. They're not even using Electrolytic Caps inside (not sure what exactly they're using) which is more of the concern for lifespan being reduced at those temps than just about anything you'd typically find in such a system.

Personally, I'm not worried about longevity at those temps. But if there was a percent boost to power production, by dropping the temp a bit, then I'd grab one of my arduino boards, one of my temp senors, a relay, a couple fans, may even design and 3D print a ducted housing, and have a nice little project. But if it effectively did nothing, well then I'll spend my time on something else, like trying to get the SolarEdge API to work.

[discodanman45]

My inverter is in the garage, which is much cooler than the outside. However, the air doesn't flow in my garage. Yesterday my inverter got to 187 degrees and it was still pumping out the max 7.6 kW limit of the inverter. I will probably find out what temps affect the inverter soon and will let everyone know. If I can hit 187 degrees now, 200 plus will happen soon as the 110 constant outdoor temps hit my area.

[library]

That's what we're all trying to determine. Taz427 and discodanman45. Solaredge says at 122 F ambient they start derating. But 122F ambient on the south side of a wall is a lot different from 122F ambient inside your garage. And since they're never going to tell/show us when they're derating the inverter we're all sort of in the dark. Especially since the only temp we can see is the inverter heatsink temp.

My wild guess is that anything below 180F on inverter temps is probably fine. I base that only on my basic knowledge of integrated circuits and their common operating temperatures. Most will run up to 85-90C just fine. And just to reassure discodanman45 the temp you saw is in no way a fire hazard. No one freaks out when their computer hits that temperature which it probably does multiple times a day.

From my reading of the thread so far people have certainly reduced their temps with a fan but it's made no discernible difference in inverter output. Which is really the answer we're seeking. What inverter temp does it start derating at.

[TAZ427]

Has anyone determine how much savings if any is achieved by lowering the temp of the inverter? I pulled some temps today during it's peak operations and it's gets to about 163F. I could rig up a temp sensor, to trigger a relay and enable a 12v supply and drive some fans, but I'm wondering if it's worth it (well beyond the enjoy of building it?) Note this is on a SE7600H (so a lot smaller footprint than the OPs inverter.)

[discodanman45]

People are worried about 150 degree temps? I reached 185 degrees today... My system is maxing my inverter from 10 am until 3 pm every day and I think it will probably die by the end of the summer if not dealt with. I am contacting my contractor about this and seeing what they can do. My panel will have to be modified to get another inverter or larger inverter, but I am not paying for that. I also don't want a fire hazard in my garage!