Reply to post: Re: How about a nice long hot summer?

I heard somebody say: Burn baby, burn – server inferno!

SImon Hobson Bronze badge

Re: How about a nice long hot summer?

I think we've probably all met at least one of the A?C cowboys - or at least, one of their installations.

A favourite of mine was for a client who arranged a dedicated little server room int heir new building, and had A/C installed in it. To be fair, the client asked me (me by name) to look at the A/C proposal they'd had since I knew a little about it. Having checked that the capacity was about right (with a bit of spare), and especially that it could handle dry air[1] I said it should be OK. And to be fair, it was for a year or two - even though they put in in the wrong place in the room.

After said year or two, the room got hot, and the A/C wouldn't run - so it was down to propping open the doors and directing a fan into it. A (so called) engineer was dispatched to look at it and ... declared that it was "the wrong sort of room", and there was no way that unit could have kept it cool (despite the empirical evidence to the contrary for the last year or two). He insisted that it was gaining too much heat through the ceiling, but needless to say, adding an extra layer of insulation did nothing.

After a few days of this finger pointing, the client asked for my opinion, I spoke with the (so called) engineer and he was not be being persuaded that there was anything wrong with the system - in spite of it tripping out a minute or two after it was reset.

So I rang the engineers employer, spoke to the engineering manager there and explained the situation - to get a "that doesn't sound right" response. An hour or two later it was magically discovered that there was a fault (stuck reversing valve in the outdoor unit) and it got fixed.

[1] That being a favourite problem when systems designed for comfort cooling of homes and offices are put into a server room without all that sweat and breath moisture us meat bags put into the environment. Without the extra energy removal needed to overcome the latent heat of fusion in condensing that moisture, the evaporator coil runs too cold, what moisture there is condenses and freezes on it, and after a short time the system stops working with the evaporator coil embedded in a solid lump of air-flow blocking ice.

Some better systems can detect the onset of this and go through a defrost cycle - most don't.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon