[Tfug] *Way* OT: Swamp coolers, ACbrrs, etc.

Adrian choprboy at dakotacom.net
Mon Jun 18 19:22:57 MST 2007


On Monday 18 June 2007 15:53, Bexley Hall wrote:
[snip]
> the ACbrrr...  I've yet to undertake an empirical
> study of which is most *economical* to operate
> (if I get time, perhaps I will do that this week)
> though I have definite preferences about which I
> *like* best, and when...
> 
> Now, I'd like to try to codify rules for when to
> use each of the above (we have dual cooling).  It
> would be nice to build an expert system to tackle
> this instead of relying on the "expert system" (ha!)
> in my *head*!
> 
> But, this is a non-trivial algorithm!  (hence the
> beauty of an expert system approach)

> For example, we can typically get a 25 - 30 degree
> delta from outdoor temperature (based, of course, on
> RH, condition of the pads, etc.).  So, a simple
> rule of thumb for the cooler is:
> 
> if (outdoor_temp > indoor_temp + 25)
>    dont_use_cooler;
> 


Well now.... I would go out on a limb and say that, until the humidity exceeds 
~75% (WAG), a swamp cooler will ALWAYS be more efficient than an AC unit. Now 
notice... I said more efficient, not more comfortable...

The power used by the circulation fan in both cases is approximately the same. 
The power used by the water pump is negligible, while the AC compressor power 
usage is very significant. Typical power usage is going to very depending on 
AC SEER rating, size, ducting, etc. as well as the required duty cycle to 
maintain temperature. The more humid it is, the better the AC unit works 
(more heat capacity per volume of air over the condensor coils), but the 
worse the swamp cooler is (reduced evaporation rate). A guess for high temp, 
low humidity summer is something like:

Swamp -
  Fan 120V*2.5A = 300W
  Pump 120V*0.25A = 30W
  40% duty cycle == 3.2kWh/day

AC -
  Fan 120V*2.5A = 300W
  Compressor 240*18A = 4300W
  40% duty cycle == 44kWh/day

So... basically the only case a swamp cooler is inefficient is when humidity 
rises such that it is run continuously to make a small temperature 
difference. I think you really need to plot 3 axises to find the convergence 
points of swamp/AC efficiency: temperature delta, humidity, duty cycle.

Now... comfort wise, that is left up to the user. And I would also guess users 
very greatly. But again, a WAG, I would propose that a rule of thumb would 
be:

if (make_me_cold ){
  delta_T = max(0,min((outdoor_T - indoor_T),30));

  if ( dew_point < (indoor_T - delta_T) ){
    use_swamp_cooler;
  }else{
    use_ac;
  }
}

Adrian
  




More information about the tfug mailing list