Double the lifetime of an inverter?

[DaveDE2]

While installing the new inverters here, got to thinking more about lifetime.

Lifetime of all electronics, barring any catostrophic failures follows a bathtub curve; infant motality, long and expected life in the middle, then high failure rates at the end.

Then got to thinking about what I've heard about most inverter failures - electrolytic capacitors. Aluminum electrolytic capacitors are the best technology we have for large storage and is needed for large power supplies but they're also the main point of failure. Their failure rate basically doubles for every 10C temperature rise (as do semiconductors). They are being "worked" as hard as the big FETs or IGBTs in the inverter but they have much less ability to dissipate heat because of their physical properties, less thermal conductivity to anything they can be attached to.

So I'm thinking of simply installing a small fan below the inverter, blowing air up across the cooling fins in an effort to keep it cooler by 10C, doubling the lifetime from ~10 years to ~20. I hope!

[sensij]

Inverters that would be expected to meaningfully benefit from a fan have one designed in. Aside from the logical fallacy that cooling the fins by 10 deg (if you achieve it) would also mean the components are operating 10 deg cooler, other forum members have rigged up something similar. It probably falls into the category of things that cost more than they save, but since the cost is so small, and the possibility of savings is so much higher... might as well try.

[J.P.M.]

FWIW: I've got a window fan blowing up below my inverter (rebadged power one, 5kW). It does indeed seem to knock about 10 deg. C. out of the heat sink temp. if the inverter display and S.P. monitor #'s are to be believed. I probably could get more delta T with some ducting, but I doubt it would be much more than 5 deg. C. more or so out of the inverter heat sink.

Typical temp. on recent fouling test at the time of minimum array incidence angle: Date:03/31/2016.Time: 1215 P.D.T. Fan turned on. Heat sink temp. without fan: 53.3 deg. C. Garage amb.. T. (inverter coolant inlet T.) = 22.6 C. Min. incidence angle time = 1306 P.D.T. Heat sink temp. at min. incidence angle time = 42.1 deg. C. Inverter coolant outlet T. = 28.0 C. Heat sink delta T (12:15-13:06 hrs.) = 53.3 - 42.1 = 10.9 C. Coolant delta T = 22.6 - 28.0 = 5.4 C.

I'm probably pulling about 1/3 or so of the waste heat (energy) out of the inverter (~~ 10.9/(53.3 -.22.6) ~ .36.). As for cost effectiveness of my Rube Goldberg system - the fan draws about 80 Watts. The inverter operates ~ 3,800, say 4,000 hrs./yr. 80 X 4,000 = 320 kWh/yr. At, say, $0.25/kWh, -->> 320 X $0.25 = $80/yr. back of envelope cost for a start dart throw # for cost analysis.

More back of envelope stuff: The inverter's about 97% eff. That leaves 3% as waste heat. Pulling out 1/3 of that is 1%. The inverter input rate for that test was 4,745 W. 1% of that is 47.5 W. Switching to customary units: = 162 BTU/hr. The air (coolant) in/out delta T was 22.6 - 28.0 = - 5.4 C. = - 9.7 F. That means the coolant (air) flowrate through the inverter was 162/((.24)*(.0762)* (9.7)) ~ 913 ft.^3/hr. ~ 15.2 ft.^3/min. I'm SWAGing the free space past the fins is ~~ .25ft^2, giving the coolant something called a face velocity of ~ 1ft./sec.

Ducting might improve that face velocity, but I kind of doubt most coolant fans for electronics which seem similar to bathroom vent fans could cut it. Converting from free stream flow rate as listed in fan specs to ducted flow rate subject to a pressure drop can extract a heavy penalty.

So, Dave DE2, if you do add a cooling fan, I'd plan on something that can put ~ 3 CFM of air past the cooling fins/kW of inverter capacity, and plan on reducing rated free air flow on any air mover by a factor of ~~ 3 or 4 or more.

Separate from the cooling fan games, for those interested, the array P.O.A. irradiance on the panels for that test at 1306 P.D.T. 03/31/2016 was estimated at 25,127 Watts (963 W/m^2). The ave. array temp. was estimated at ~ 48.0 C., using inst. voltages. The roof amb. air temp. at the array at 1306 hrs. was 17.8 C., and the 6 min. est. ave wind velocity over the array was ~ 1.8-1.9 m/sec. The inverter output was 4,603 W, giving a system instantaneous eff., including all estimated/measured wiring and inverter losses of (4,603/25,127) ~ = 18.3 %. The array was pretty clean that day, having been washed @ ~ 0700 hrs..

[DaveDE2]

After taking the covers off and looking at this SB 6.0, I don't think it's a logical falicy at all that cooling the fins by 10C will cool all the internal components by that much once thermal equalibrium is reached (say 30mins-1 hour max). I'm fanning mine, your's can bake

[DaveDE2]

@JMP
Great stuff, thanks, have to re-read but have to run right now. - What I'm mainly going for is longevity, not efficiency as far as the fan goes, I'm willing to give up those watts (Ithink), ... think Arrhenius equations, etc.

[SunEagle]

After taking the covers off and looking at this SB 6.0, I don't think it's a logical falicy at all that cooling the fins by 10C will cool all the internal components by that much once thermal equalibrium is reached (say 30mins-1 hour max). I'm fanning mine, your's can bake

Moving air across the radiator fins may help increase the delta T but you might be able to find other types of heat exchangers or another way to get the heat away from the electronics. Something as simple as running a fluid in metal tubing that is in contact with the fins can pull the heat away faster then using forced air.

[J.P.M.]

After taking the covers off and looking at this SB 6.0, I don't think it's a logical falicy at all that cooling the fins by 10C will cool all the internal components by that much once thermal equalibrium is reached (say 30mins-1 hour max). I'm fanning mine, your's can bake

Depending on where the heat sink temp. is taken, a 10 deg. C. drop in heat sink temp. will probably not result in an equal temp. change (decrease) in the temp. of every part of the inverter. I'd be pretty confident of some cooling of the entire inverter as a result of the laws of thermodynamics, but probably not 10 deg. C. throughout, and certainly not uniformly.

The benefits of adding a cooling fan to an inverter designed for natural convection cooling or adding more forced cooling to a design may or may not be cost effective or increase the life of an inverter, and those are probably the two big reasons for design considerations with respect to forced cooling.

The common thinking is that cooler running electronics tend to last longer. In applications such as residential PV inverters, the heat transfer designer's job is to get the safest, most fit for purpose (including most cost effective) cooling design that accomplishes the mission. Designing out the need for forced cooling is usually preferable to assuming it will be a better way and throwing forced ventilation at a design. Sometimes nat. convection may actually add to the initial cost, "forcing" a fan or forced cooling design. Sometimes vice versa. Sometimes both, or neither, or something else.

Adding forced ventilation to a nat. convection design in a bolt on, after market fashion may enable an inverter to last longer. How much longer and, assuming the additions are safe, and what those additions may mean to the cost from added material, labor, maint. and energy costs, not to mention aesthetics are other questions.

I did some DIY cooling stuff as I described above, and my gut tells me it's probably not, or maybe borderline cost effective, but I'm having fun, satisfying my curiosity, and maybe slowing down the rate of brain cell death. Not practical perhaps, but priceless.

[SunEagle]

Depending on where the heat sink temp. is taken, a 10 deg. C. drop in heat sink temp. will probably not result in an equal temp. change (decrease) in the temp. of every part of the inverter. I'd be pretty confident of some cooling of the entire inverter as a result of the laws of thermodynamics, but probably not 10 deg. C. throughout, and certainly not uniformly.

The benefits of adding a cooling fan to an inverter designed for natural convection cooling or adding more forced cooling to a design may or may not be cost effective or increase the life of an inverter, and those are probably the two big reasons for design considerations with respect to forced cooling.

The common thinking is that cooler running electronics tend to last longer. In applications such as residential PV inverters, the heat transfer designer's job is to get the safest, most fit for purpose (including most cost effective) cooling design that accomplishes the mission. Designing out the need for forced cooling is usually preferable to assuming it will be a better way and throwing forced ventilation at a design. Sometimes nat. convection may actually add to the initial cost, "forcing" a fan or forced cooling design. Sometimes vice versa. Sometimes both, or neither, or something else.

Adding forced ventilation to a nat. convection design in a bolt on, after market fashion may enable an inverter to last longer. How much longer and, assuming the additions are safe, and what those additions may mean to the cost from added material, labor, maint. and energy costs, not to mention aesthetics are other questions.

I did some DIY cooling stuff as I described above, and my gut tells me it's probably not, or maybe borderline cost effective, but I'm having fun, satisfying my curiosity, and maybe slowing down the rate of brain cell death. Not practical perhaps, but priceless.

If you look at the "cooling fan system" on the high end gaming computers they use some type of "closed fluid" system in conjunction with the fan to keep the main processor cooler allowing it to run very fast. I wonder if something similar couldn't be used on a solar inverter components.

[Mike90250]

i'd used a small desktop osc fan on a simple timer, aimed at my inverter, (GT3.8) which brought my logged heatsink temps down about 15F.

Can't hurt UNLESS the fan adds vibration to the unit and lead free solder joints start failing

[J.P.M.]

If you look at the "cooling fan system" on the high end gaming computers they use some type of "closed fluid" system in conjunction with the fan to keep the main processor cooler allowing it to run very fast. I wonder if something similar couldn't be used on a solar inverter components.

Yea, that's probably a closed system not unlike how a fridge or an air conditioner works. Advantages include perhaps less dust/crap exposure to the electronics and probably lower pumping costs. Reliability is probably similar to that of a refrigerator. The circulating fluid may be a refrigerant or not. Bigger inverters may work that way. Probably pretty expensive for residential inverters both in $$ and space requirements. I'd guess the $$ curve is different for gambling machines where down time can cost real money.

[DaveDE2]

All I care about is prolonging the life of my inverter electronics per Arrhenius etc. I'm going to install a few small ducted fans and measure the with and without fan temps of the heatsink (FETs) and inside ambient air temp (Caps) and report back. Thanks.

[J.P.M.]

All I care about is prolonging the life of my inverter electronics per Arrhenius etc. I'm going to install a few small ducted fans and measure the with and without fan temps of the heatsink (FETs) and inside ambient air temp (Caps) and report back. Thanks.

+1 on Mike's comment about vibration.

[Willaby]

All I care about is prolonging the life of my inverter electronics per Arrhenius etc. I'm going to install a few small ducted fans and measure the with and without fan temps of the heatsink (FETs) and inside ambient air temp (Caps) and report back. Thanks.

I think its a good idea. I have a SE5000 (no fan) and was going to run a small computer fan via a small 5w solar panel, but I never got around to it. I figured the solar output would automatically ramp up the fan speed in conjunction with inverter load on a hot day at noon. IIRC, the inverter has never exceeded 110F, so I kind of lost interest thinking it really doesn't get that hot, but I still eventually want to do it.

[peakbagger]

IMHO, air flow is important but air temperature and loading of the inverter are major factors in inverter life. I have my inverters attached to a concrete wall in a below grade basement. On the hottest day in the summer the incoming air temp is no more than 60 deg F. I also try to oversize inverters slightly with the goal of keeping the output below the maximum rating of the unit (no clipping). I find that the various solar installers all tend to oversize their arrays for the inverter and rarely do I see a microinverter sized to match up with a panel rating. Enphase even came out with technical paper insisting that a microinverter should intentionally be undersized so that it clips as that is best economics. I prefer long term reliability and probably would not even install a microinverter due to the environment it is located but I sure wouldn't overpanel it.

Knock on wood I have a 14 year old Advanced Energy inverter that is still running. These units were an early grid tied only inverter and most installers over paneled them. They had lot of warranty failures and it drove the company out of business. At least one former employee blamed a lot of the issues on thermal issues. It is passively cooled (no fan).

I see PR shots of PV installations and see inverters mounted on exterior walls siting right in the sun. Might be good for a PR shot but lousy for the electronics. The best place is indoors preferably in air conditioned space or in a cool basement.

[Mike90250]

..........I see PR shots of PV installations and see inverters mounted on exterior walls siting right in the sun. Might be good for a PR shot but lousy for the electronics. The best place is indoors preferably in air conditioned space or in a cool basement.

But that's an extra 15 feet of conduit !!