Pv solar install the inverter in loft or garage

don0301

let me see...

a MTBF of 50,000 is average in the Industry?

so, 50,000/24/365= 5.7 years?

so, according to zeupater a 10C difference means a life time of 2.85 years?

oh wait, he Backtracked what he said before...

best he explained it...:D

Dave_Fowler

don0301 wrote: »

let me see...

a MTBF of 50,000 is average in the Industry?

so, 50,000/24/365= 5.7 years?

so, according to zeupater a 10C difference means a life time of 2.85 years?

oh wait, he Backtracked what he said before...

best he explained it...:D

Generally 50,000 hours is the MTBF at the maximum rated working temperature for a relatively stable component, but we should not be considering working the components at temperatures above maximum which is implied by asking for a working life at 10C higher temperatures.

I've said before that there are other components which should be considered as more likely to fail than these stable parts.

This is not the place to be swapping spec sheets from different manufacturers so I've picked one for a standard part I would consider as a suitable component in an inverter. I'm reading the spec sheet for a 105C rated electrolytic capacitor - one which typically would be used in an inverter. This says that at a temperature of 105C the life is 5,000 hours and at 40C it is 500,000 hours. The life is defined as a 3% failure rate for the component. A failure is further defined as being when the component falls outside its specification, not when there is a catastrophic failure. So design in a little safety factor and the working life would be further extended.

Remarkably if you take 500,000 hours at 40C and halve the time for each 10C rise, by the time you get to 105C you can expect a lifetime of 5,555 hours which is fairly close to the 5000 hours at 105C in the specification.

If an inverter manufacturer does his design correctly he will place the most unreliable components nearest to the air in-take, so that they run at the lowest possible temperature. More reliable components are placed further from the air in-take and so the lifetime of the entire design is optimized. This is standard design practice. So a component whose 40C life expectancy is 500,000 hours can be expected to last 125,000 hours at 60C. (over 14 years).

Taking a mix of many different components, most of which are more reliable than the capacitors, it is not unreasonable to expect a lifetime of 10 years or more.

Dave F

don0301

my post was mocking zeupater

but thanks for confirming my point.

Dave_Fowler

don0301 wrote: »

my post was mocking zeupater

but thanks for confirming my point.

Was it your point? Your post appeared to be quoting an MTBF of 50,000 hours and suggested that Z's halving life expectancy every 10C would produce a ridiculously low life-span. Whereas your MTBF was for a maximum working temperature and you should have been doubling the life expectancy for every 10C drop in temperature - exactly the opposite of what you posted.

let me see...

a MTBF of 50,000 is average in the Industry?

so, 50,000/24/365= 5.7 years?

so, according to zeupater a 10C difference means a life time of 2.85 years?

oh wait, he Backtracked what he said before...

best he explained it...:D

The one (totally and truly randomly selected) component specification I used gave absolute credence to his figures showing a deviation of only 10% over a 60C+ temperature range and a very realistic lifespan of 14 years at a temperature of 60C.

Dave F

don0301

Dave_Fowler wrote: »

Was it your point? Your post appeared to be quoting an MTBF of 50,000 hours and suggested that Z's halving life expectancy every 10C would produce a ridiculously low life-span. Whereas your MTBF was for a maximum working temperature and you should have been doubling the life expectancy for every 10C drop in temperature - exactly the opposite of what you posted.

The one (totally and truly randomly selected) component specification I used gave absolute credence to his figures showing a deviation of only 10% over a 60C+ temperature range and a very realistic lifespan of 14 years at a temperature of 60C.

Dave F

my post was making fun

sadly, you took it seriously

you really must be special

Dave_Fowler

don0301 wrote: »

my post was making fun

sadly, you took it seriously

you really must be special

I don't think it was sad that I took you seriously, it was just unfortunate for you that it made you look so foolish.

I don't think I'm anything special, just someone who hopes to disseminate what knowledge I have to help people who follow this thread. It does not help when people start mocking others within what is a serious discussion with statements which, taken at their face value, are misleading.

Dave F

kolofarthos

Don 0301,
You appear to be in a minority of one in the technical argument. That does not necessary make you wrong but the fact remains you are.

don0301

arrhenius equation applies just as equally in winter as it does in the summer...

i did agree with Dave F that the most likely temperature related failure in an inverter will be from temperature cycle, not ambient temperature issues.

stating lofts can get hot, garages rarely do doesnt add much credit to your argument. its just an assumption. taking it further you could argue the opposite in the winter.

as ive stated before inverters are designed to operate in a temperature window. they also have design features to control internal temperature.

if your right, why dont manufacturers state to avoid loft installation at all cost?

you think zeupater is right. good for you.

larkim

Don - sometimes I wonder why you bother to carry on arguing. Then again, sometimes I do argue continually just for the sake of it, so I suppose I can understand it!

I would have thought that the situation, without the technicalities, is fairly readily summed up:-
- inverters work better cooler than hotter
- inverters will generally last longer in cooler rather than hotter conditions
- inverters (it seems) automatically de-rate themselves when they detect it getting "too" hot (i.e. >40degC ambient it seems)
- lofts, in general as an area directly exposed to solar radiation more than the rest of the house, may well get hotter than most other areas of the house in the summer; potentially exceeding 40degC
- garages (or any other room in a house, perhaps with an exception of a conservatory) are likely, on average, to have a cooler ambient temperature than the loft
- therefore garages make more sense than lofts to install inverters purely when weighed up on an ambient temperature basis
- there are other issues such as cable run lengths which might need to be considered, but they probably don't outweigh the de-rating effect of an inverter running in high ambient temperatures

I'm not entirely sure what there is to argue about?

Matt

larkim

btw, my inverter is in my loft and I'm happy with it there as I don't have a garage nor anywhere else in my house that would have been "nice" to have had an inverter in.

And the people that will lose in the main out of this (if the inverter starts de-rating) are the RaR company that installed it.

Matt