Micro Inverters on a hot day

[agarg]

Folks:

I have a grid tied solar operational with NEP BD800 micro inverters. On July 14, and 15th, I noticed that they were producing zero electricity in the middle of the afternoon when it was too hot. After a little diagnosis, I started to suspect if it was because of overheating. I have a charcoal roof with composite shingles and there is about 4 inch of space between the modules beneath the modules and the shingles.

To test this out, I took a hose and sprayed a lot of water to soak the modules and hoped that it would start producing again. AND it did.

Is this normal? Any suggestions?

Anil

 

[timselectric]

They are supposed to be designed for the heat of being on the roof. I would probably be trying to make a warranty claim.

 

[agarg]

They are supposed to be designed for the heat of being on the roof. I would probably be trying to make a warranty claim.

OK.

 

[SpongeboB Sinewave]

We never would have thought that putting power electronics on a HOT roof was a great idea...

Hot and cold and hot and cold etc... But somehow, it works OK.... Typically.

And I understand that there are actually some batteries along with microinverter that are located on the roof. Just doesn't sound logical to me

boB

 

[timselectric]

No way I would put batteries on a roof.
Or even outside, for that matter.

 

[agarg]

No way I would put batteries on a roof.
Or even outside, for that matter.

I had to jump many hoops with the Fire Department and AHJ before they allowed me to put the battery inside the garage. And, even after the approval, I had a more rigorous two stage inspection for the fire risk from thermal runaway for Li-ion battery represent.

 

[ArtDeco]

No way I would put batteries on a roof.
Or even outside, for that matter.

I am going nuts trying to figure out a safe place to store batteries. I was planning to put them underground but there is too much condensation, frost heave, etc. here. The house and finished firetrap of a basement have exactly the same footprint so no where without a "habitable area" above it, and I don't want them in the living area. I'm leaning towards an insulated shed on a brick wall underneath the panels here.

 

[timselectric]

I had to jump many hoops with the Fire Department and AHJ before they allowed me to put the battery inside the garage. And, even after the approval, I had a more rigorous two stage inspection for the fire risk from thermal runaway for Li-ion battery represent.

Lithium ion can be a problem. I wouldn't put them in, or anywhere near my house.
I have LiFePO4's. No worries at all.

 

[timselectric]

I am going nuts trying to figure out a safe place to store batteries. I was planning to put them underground but there is too much condensation, frost heave, etc. here. The house and finished firetrap of a basement have exactly the same footprint so no where without a "habitable area" above it, and I don't want them in the living area. I'm leaning towards an insulated shed on a brick wall underneath the panels here. View attachment 157806View attachment 157806

That's why I went with lifepo4.
They're in the house, 20' from where I sleep.
They like the same environment as humans.

 

[Brett V]

How many panels? What kind of power are we talking about? Spraying water on the roof cooled something, but I have a hard time believing you hit all of them! Maybe a connector problem?

 

[agarg]

Lithium ion can be a problem. I wouldn't put them in, or anywhere near my house.
I have LiFePO4's. No worries at all.

All Tesla and LG are Li-ion.
My battery is also LFP from BYD.

 

[agarg]

How many panels? What kind of power are we talking about? Spraying water on the roof cooled something, but I have a hard time believing you hit all of them! Maybe a connector problem?

Going to check that out on a cooler day!

 

[Rob@SaguaroEnt.]

Allot of interesting thoughts here. And now my$.02, that has cost allot more to learn.
At 4pm as I write it is 110 degrees outside. I had replaced a magnum inverter that died during Covid with a renogy. Last week I noticed it was shutting down in the evening hours. As this is in a rv utilizing factory wiring I have it supporting a 1/2 ton ac and the refrigerator. Pretty obvious it’s gonna take the rest or July and August off. IT’s performance specs from the data sheet show a max of 104 ambient. Next time maybe I’ll read the data sheet before I buy a seasonal inverter!
The aio s (6548s) seem to be hanging in there better however, once the battery bank is full and ac load isn’t consuming
Most of the pv it seems like as the charge controls start to shut it off there comes an overheat code.
Cleaning the filters seemed to help as did opening the door to their room and allowing a bit more ventilation.
Point being all solar power equipment doesn’t like the heat and if your climate sees over 100* your gonna have issues.

With regard to batteries and storage my plans are a small shed that is well insulated with sheet metal or aluminum on inside walls. A microwave oven is designed so if something goes wrong the fire can’t spread. While I’ve heard stories of batteries and inverters that have “thermal runaway”, just like you have I wanna be able to stand back with a cold one and watch the show without concern for my residence. Or, no matter what ul says, I can replace a shed for a small portion of what the home costs.
If you didn’t follow the insulated part, I’ll be able to add a small ac to keep the electronics cool on the real bad days as needed.
if your chasing off grid ac with solar you’re going to have to have 20-30 % cushion in your numbers.

 

[agarg]

While I’ve heard stories of batteries and inverters that have “thermal runaway”,

Only batteries have this tendency to overheat and not the inverters. Also LFP have significantly less (in not near zero) propensity for such overheating.

 

[agarg]

Going to check that out on a cooler day!

I checked all connections and they look great. I also connected exposed bare copper on one of the modules that wasn't producing and then took the voltage measurements and it did show 48 DC. I then connected a heat gun to the exposed copper and it started the fan and making observable heated air. So the module is surely working. And, therefore the mystery remains. I will keep testing till I isolate the root cause.

The big learning from all this is that if trip to roof top is less desired then string inverters do well.

 

[fafrd]

I checked all connections and they look great. I also connected exposed bare copper on one of the modules that wasn't producing and then took the voltage measurements and it did show 48 DC. I then connected a heat gun to the exposed copper and it started the fan and making observable heated air. So the module is surely working. And, therefore the mystery remains. I will keep testing till I isolate the root cause.

The big learning from all this is that if trip to roof top is less desired then string inverters do well.

Did you ever get this sorted out?

Are your BDM800s working reliably now?

 

[agarg]

Did you ever get this sorted out?

Are your BDM800s working reliably now?

Some are and some are not. Waiting for free time on a cooler day to take a look. Earlier, I did check out the DC connection on the Kitchen side panels and once I re-seated the DC connectors the non-functioning also started working.

I am producing a lot of surplus and so there is slight lethargy to the sense of urgency! A friend in the Southern LA area is using Hoymiles and he is generally happy except on the hot day, around the peak temperature his micro shut off.

I will surely post my progress. Thanks for checking.

 

[fafrd]

Some are and some are not. Waiting for free time on a cooler day to take a look. Earlier, I did check out the DC connection on the Kitchen side panels and once I re-seated the DC connectors the non-functioning also started working.

OK, so bad MC4 connection is at least one of the issues you were facing. Be careful about that - high-resistance MC4 connections can melt MC4 connectors very quickly.

I am producing a lot of surplus and so there is slight lethargy to the sense of urgency!

I don’t know how easy it is for you to access your MC4 connectors, but if easy, I’d strongly advise checking temps of the connector housing during the peak part of the day with an IR gun or alternative.

A friend in the Southern LA area is using Hoymiles and he is generally happy except on the hot day, around the peak temperature his micro shut off.

That makes me happy I avoided Hoymiles. The NEP inverters handle overheating much more elegantly than shutting down. They will start throttling output as they approach maximum temperature to stay under it and avoid shutdown.

On the peak part of a hot day here in the Bay Area, I’ve seen output drop by 15% or even 20% over a 50% duty cycle that I’ve seen increase to as much as 70%.

So output is throttled by 7.5% to as much as 14% but that is still far preferable to inverter shutdown which loses 100% until the inverter has cooled down…

Next time we have a very hot day, go watch your meter for a few minutes if you want to confirm the same behavior. In my case the cycle seems to be about 18 seconds, so just record 10-20 consecutive readings off of your meter and you should be able to see whether your BDM800s have a similar cycle…

I will surely post my progress. Thanks for checking.

Keep us posted - your are the BDM800 pioneer here in the Forum!