Sea-level power supplies..?

[daveza]

Curious -

A client has 10 or so new pc's using the low energy boards and 200W power supplies.

Over the past month 5 of these have failed - pc turns on but on all of them nothing is fed to the lcd monitor.

When the first failed I replaced the power supply with a 450W and all was well.
When the next 4 failed within a week I queried this with the supplier.

Apparently there is a component in the power supply which fails and they simply replace this and not the whole power supply.

Here's the thing - of all the the pc's that have been returned with this problem every single one has been from a coastal site. None from inland - ie failures in CPT, PE, DBN, none from the Free State or Gauteng..

Coincidence, possible ?

 

[Conradl]

The air is dryer up here. I'm surprised anything works in Durban....

 

[Drake2007]

No coincidence, our PC's rust just like our cars.

 

[daveza]

These are new power supplies - less than 3 months old.

I've had power supplies on the edge of the sea which are 4 years old and still working ?

 

[ToxicBunny]

Cheap stuff just doesn't cope with the climates down here...

I can't speak for PE or CPT, but DBN is incredibly rough on PC equipment if its not looked after.

 

[LazyLion]

These are new power supplies - less than 3 months old.

I've had power supplies on the edge of the sea which are 4 years old and still working ?

Higher Atmospheric Pressure, greater humidity. It's bad enough that you can feel it.

 

[Drake2007]

Higher Atmospheric Pressure, greater humidity. It's bad enough that you can feel it.

Add to that it's rainy season, power lines get nailed more with the inevitable power surge or spike that will cause, possibly those small PSU's can't handle it.

Suggest the OP sell them some UPS's

 

[bekdik]

Sounds like cheap badly built kit.

 

[daveza]

Possibly, but I've checked with my Joburg guys - they have dozens on sites and all working fine.

Why would the same component in the power supply fail yet only at the coast .... just doesn't make sense to me.

 

[ponder]

Why would the same component in the power supply fail yet only at the coast .... just doesn't make sense to me.

Must be influenced by humidity somehow.

At least we don't have problems with static here

 

[EchoZA]

The air is dryer up here. I'm surprised anything works in Durban....

hey I still work

 

[analogsa]

Actually there is nothing in a PS which is sensitive to humidity. Heat - certainly, but for humidity to have any effect it needs to be extreme for mould to start growing on the board. And mobos will be the first to go. Probably simply a bad batch.

 

[kayvee]

Actually there is nothing in a PS which is sensitive to humidity. Heat - certainly, but for humidity to have any effect it needs to be extreme for mould to start growing on the board. And mobos will be the first to go. Probably simply a bad batch.

I don't entirely agree. Sure there is, high voltage switching circuits which under extreme humidity environments can cause arcing across a poorly designed board or inferior components. This can contribute to breakdown in mosfets and/or other semiconductors in the PSU leading to it's failure. No high voltage stuff on the mobo.

 

[czc]

How do countries like Malaysia cope then?

 

[kayvee]

How do countries like Malaysia cope then?

I would say it relates to the quality of the PSU. If that batch was shipped there they would have had the same issues.

Edit: And that is precisely why environmental (humidity and temperature) testing is included under the Safety Testing heading for CE Certification (>67V input).

 

[Drake2007]

From Intel ATX12V Power Supply Design Guide

5.1. Temperature
Operating ambient +10 °C to +50 °C
(At full load, with a maximum temperature rate of change of
5 °C/10 minutes, but no more than 10 °C/hr.)
Non-operating ambient -40 °C to +70 °C
(Maximum temperature rate of change of 20 °C/hr.)

5.2. Thermal Shock (Shipping)
Non-operating -40 °C to +70 °C
15 °C/min ≤ dT/dt ≤ 30 °C/min
Tested for 50 cycles; Duration of exposure to temperature
extremes for each half cycle shall be 30 minutes.

5.3. Humidity
Operating To 85% relative humidity (non-condensing)
Non-operating To 95% relative humidity (non-condensing)
Note: 95% RH is achieved with a dry bulb temperature of
55 °C and a wet bulb temperature of 54 °C.

Humidity would be a factor if those PSU's don't follow the ATX spec.

 

[ponder]

From Intel ATX12V Power Supply Design Guide



Humidity would be a factor if those PSU's don't follow the ATX spec.

On monday night the humidity was 94% here at 1am in the morning, Durbs goes over 85% a lot in the summer, not the winter though.

If you stick your tongue out you can lap up the water in the air