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

[bcroe]

You are getting close by changing the size of the fan. Next step is a variable speed fan, after
that replace the manual speed control with closed loop electronics. Bruce Roe

[Kendalf]

Update on my experiments with external cooling fans for a SolarEdge SE6000H inverter. I switched to a pair of lower power computer case fans (1000 rpm, 12V, 0.1A) to see if it would result in less temperature ripples as the initial fans that I used cooled a bit too quickly, bringing the temperature down to the shut-off temp for my thermostat. This still gave a lot of ripples in the temperature, as the heatsink would warm and cool quite rapidly.
Temp Chart with Fan - Heatsink temp sensor.png
I then moved the thermostat temp probe from the inverter heatsink to the interior of the inverter, using the gland openings for the communication antenna. After a bit of tweaking of the thermostat on/off points, this gave me much smoother results. The initial crest and dip at 10:15AM is when the fans first turned on, and they remained on for most of the rest of the day, keeping the peak temperature of the inverter at 138.6F, despite the ambient temperatures for the day reaching the low 90s.
Temp Chart with Fan.png

In contrast, here's the temperature chart for a sunny day before I added the fans, with peak inverter temperature hitting 163F, despite the ambient temperature being nearly 15F cooler on this day.
Temp Chart without Fan.png
My last tweaks may be to relocate the temperature probe to another location inside the inverter. The temperature readings at the current location don't quite match the actual inverter temp readout, making it more difficult to adjust the on/off temp thresholds for the fan thermostat.

[Mike90250]

Just let the fans run, once they come on. I use a timer on my gear, 11AM fans come on, 5:30pm, go off. The 8 deg swing, will be a bit harmful, warmer/cooler/warmer/cooler and thermal cycles - even small ones, count against you.

[Kendalf]

Yes, the ripples are from the cycling fans. I had the thermostat set to a 5 degree range between on and off points. I'm testing an 8 degree range today.

[bcroe]

That appears effective, are the ripples from fan on/off cycles? I would not expect much effect on
efficiency, the main purpose to be extending equipment life. Generally a larger fan will be more
efficient and less noisy than several small ones, but has no redundancy. Bruce Roe

[Kendalf]

Keep us updated on this as i just ordered 2 fans and was going to do the same as mine with a fan blowing across the heatsink reduces the temps significantly. I have seen 178 and seen it reduce to 135 with the fans on while at high peak of 7600 watts.

First sunny day we've had in awhile, and first opportunity to see if my fan setup is effective. Here's a picture of the fans sitting on top of the inverter (SE6000h). They are extra 120mm CPU heatsink fans that I had laying around (12V, 0.25A, 1350rpm max, dual ball-bearing): Cooling Fans for Inverter_cr.jpg



I originally purchased L-brackets to mount the fans horizontally over the inverter cooling fins, but thought I would just mount them temporarily as pictured at a slight angle just to test things out before I drilled any more holes in my garage. The fans are connected to a 12V power adapter, which is in turn plugged in to a digital thermostat outlet controller. The thermostat has a temperature probe that I simply poked down between the inverter heatsink fins, and once the temperature reaches a certain level, the outlet will receive power, and the fans start spinning.

Turns out even this jury-rigged fan arrangement is effective enough, perhaps too effective. Here's my PVOutput chart from today showing PV power in green and inverter temperature in orange. Temperature Chart with Fans 2.PNG


I allowed the inverter temp to get close to its peak for the day (148.3F) and then plugged in the fans. The temp quickly dropped 13F within 10 minutes and then the temperature probe hit the lower turn-off temp. The rest of the afternoon, the inverter temps oscillated within a 4 degree F range as the fans cycled on and off, gradually decreasing as production decreased.

For tomorrow, I've increased a bit the temperature at which the thermostat powers up the fans, to see how that affects the temperature chart. I was wondering if my jury-rigged fan arrangement wouldn't be able to draw enough air through the inverter cooling fins to do much good, but it looks like they are just fine, and I'll probably just leave the fans as is.

I don't have a Watt meter to measure the draw of the fans and the thermostat when running, but it should be minimal. The fans themselves will only draw a maximum of 6 Watts, so the main question is how much does the thermostat controller draw.

While there doesn't seem to be any noticeable effect on production, today's conditions didn't come close to when I see inverter temps in the 160-169F range, so perhaps there may be a bit of benefit then. That said, this is more an exercise in reducing the temperature of the inverter electronics and possible extending the inverter life. But for the minimal cost and work involved in setting this up, I'm happy with the results, even if there is no more benefit than knowing that the inverter won't get near those previous peak temperatures anymore.

[Mike90250]

The internal fan is useless. It Never turns on.

Then I would examine to find out why. Thermo-disc switch bad or fan bad. That's the first thing to check.

[bcroe]

Excessive temp is bad, but constantly running the temp up and down isn't good either. Perhaps
a good fan system will try to keep the temp roughly constant. Doing this with a PID loop, only
the Integral element seemed to be needed. DC fan speed can be controlled by the supply voltage,
or by a duty cycle input the the fan driver circuit. Bruce Roe

[TundraLu]

I also have a 7600 HD and track my inverter temps every minute over modbus tcp. It's on the north exterior wall always shaded and I hit 170F today right about the time we hit our daily high of 87 degrees. I remember seeing a small 30-40mm fan inside the inverter when I opened it up but it's obviously not cooling using anything but the heatsink on the back for the HD Wave series.

I'm aware of the SolarEdge de-rating document but it seems fishy to me to specify 122 degrees ambient as their derating limit. But I can't really see any difference in production. I suppose to test I could put a fan on it like others in this thread have done but those results showed no difference in output either. I think the HD Wave's just run hot. But we'll hit 100 ambient here in the next month and I'll be keeping track.

The internal fan is useless. It Never turns on.

[TundraLu]

library I have an SE6000H and my peak recorded temperature has been 169 degrees F. It usually goes above 160 when production hits 6kW and the inverter starts to clip.

This morning I just rigged up two 120mm computer case fans attached to a temperature controlled power source, to see if that will help with reducing these peak temperatures. I have the fans placed at the top of the inverter, over the metal cooling fins. Today is cloudy so inverter temperatures are low, but I'll report when we get some more sunny days if it makes any kind of difference.

Keep us updated on this as i just ordered 2 fans and was going to do the same as mine with a fan blowing across the heatsink reduces the temps significantly. I have seen 178 and seen it reduce to 135 with the fans on while at high peak of 7600 watts.

[Kendalf]

library I have an SE6000H and my peak recorded temperature has been 169 degrees F. It usually goes above 160 when production hits 6kW and the inverter starts to clip.

This morning I just rigged up two 120mm computer case fans attached to a temperature controlled power source, to see if that will help with reducing these peak temperatures. I have the fans placed at the top of the inverter, over the metal cooling fins. Today is cloudy so inverter temperatures are low, but I'll report when we get some more sunny days if it makes any kind of difference.

[library]

Ok I have a 7600 HD Inverter and i am seeing temps in the 180 degrees. Does this seem right? The weird thing about it is that in my garage it is 80 degrees Ambient right now.

I also have a 7600 HD and track my inverter temps every minute over modbus tcp. It's on the north exterior wall always shaded and I hit 170F today right about the time we hit our daily high of 87 degrees. I remember seeing a small 30-40mm fan inside the inverter when I opened it up but it's obviously not cooling using anything but the heatsink on the back for the HD Wave series.

I'm aware of the SolarEdge de-rating document but it seems fishy to me to specify 122 degrees ambient as their derating limit. But I can't really see any difference in production. I suppose to test I could put a fan on it like others in this thread have done but those results showed no difference in output either. I think the HD Wave's just run hot. But we'll hit 100 ambient here in the next month and I'll be keeping track.

[TundraLu]

Thanks! I'm fine (thanks for asking), but am not going to be a CA resident much longer. A new opportunity came out of blue at the beginning of the year, and I've started working in the Pacific Northwest this month. With so much transition, not as much time for the things I enjoy.

Once I finally get settled, the rain gauge on my Davis will have a lot more work to do. And yes, count this as yet another example of why the long term "savings" of a PV system isn't always realized. At least the next owner of my house will have a good 18 years left of grandfathering. (And I'm totally fine with it... The experience of diy installing the system was worth the cost, and I'll be testing first hand if it adds anything the the market value of the house)

Ohh Man, Wish you good luck. Moving is not easy but the next homeowner will be happy with your Solar System .

[TundraLu]

Ok I have a 7600 HD Inverter and i am seeing temps in the 180 degrees. Does this seem right? The weird thing about it is that in my garage it is 80 degrees Ambient right now.

[J.P.M.]

Thanks! I'm fine (thanks for asking), but am not going to be a CA resident much longer. A new opportunity came out of blue at the beginning of the year, and I've started working in the Pacific Northwest this month. With so much transition, not as much time for the things I enjoy.

Once I finally get settled, the rain gauge on my Davis will have a lot more work to do. And yes, count this as yet another example of why the long term "savings" of a PV system isn't always realized. At least the next owner of my house will have a good 18 years left of grandfathering. (And I'm totally fine with it... The experience of diy installing the system was worth the cost, and I'll be testing first hand if it adds anything the the market value of the house)

Congrads. How far north and how far west ? East of the cascades in WA and OR can be pretty sunny. Not much out there though. At least winters west of the Cascades are warmer than New England winters.

After all this time, and now that you'll be out of town, how would you like to meet up sometime for a coke & a chat ? No pitch and I've nothing to sell, including real estate. I bet you can get my e-mail, etc. If not, N.B.D.

Whatever happens, Best Regards and Good Luck,

J.P.M.