A couple of posted topics in the past month or so motivated me to perform an experiment with extreme low power personal cooling. One post which I have not been able to relocate discussed methods of reducing overall household electricity usage. The other post by Teresa from Hershey, linked below, was chock full of useful tips on reducing energy usage for household and personal cooling.
http://greenwizards.com/node/1320
While all of her tips are good ones, I believe that the most comfort enhancement per buck (or Whr) can be had from fans directed at the individual. Teresa's post in fact features her sitting at her desk next to her favorite personal use fan, a 41" Vornado tower fan. I was not familiar with this particular product and was curious about its power usage. I visited the Vornado website and located Teresa's particular model of the many they make. Unfortunately neither the specifications page nor the user manual provide info on power usage which I find to be a curious omission. Based upon some prior experimentation with very low power whole house ventilation I was convinced that very low power fans for immediate area personal cooling were worth a look. I have mentioned on several occasions in these forums that I have a particular interest in the best deployment of limited electrical energy as represented by very low capacity off grid scenarios. For me personally the most critical comfort need is adequate cooling to get a decent night's sleep. If I am not sleeping well my physical and mental performance is impaired. For testing purposes I decided to find a ceiling fan with a diameter sized to provide airflow over the surface of a double bed with little wasted elsewhere. I had never looked very far into the technical details of ceiling fans so that was the first action. A brief synopsis of my research follows. The electrical versions date from the 1880's though there are cases of earlier deployment powered by waterwheel or steam engine. While there were some instances of use of "universal" motors which could be powered by either AC or DC, almost all made in the US from the 1880's to 1970's used AC only motors of the "shaded pole" design. This choice undoubtedly driven by their low cost of manufacture relative to other motor designs. This motor design is very inefficient in converting electrical power to mechanical power. In ceiling fan sized motors they offered % efficiencies in the low single digits (!). In designing for efficiency, in the case of motors, size matters. Once above a few 100 Hp efficiencies of above 95% are attainable and above a few 1000 Hp up to 98% is attainable. At the other end of the spectrum at the level of very small fractions of a Hp efficiency does not come easily. A shaded pole motor on a ceiling fan may consume up to 100 W and produce a mechanical power of only a few milli-Hp (0.001Hp). One Hp expressed in W is 746W. For most to the 20th century while US manufactured fans were almost exclusively the shaded pole design in other parts of the world where incomes were lower and electricity more expensive more efficient motor designs were deployed, specifically the permanent split capacitor (PSC) motor. In the case of a ceiling fan sized motor efficiencies in the high single digits became achievable. This was the limit of advancement in ceiling fan technology until the beginning of the 21st century at which time advances in electronics technology enabled a more efficient motor option. Specifically, the enabling electronic technologies were low cost power MOSFETs (a type of transistor) and low cost Hall effect sensors (senses magnetic fields). These two items enabled the development of low cost Brushless DC (BLDC) motors which can double or triple the efficiencies attainable by the PSC design. Efficiencies in the low double digits are now attainable in a ceiling fan sized motor. This is currently the state of the art for ceiling fan motor efficiency. Other advancements are remote controls (questionable whether an advancement IMHO) and better design of the fan blades. Most fans still use the flat paddles as blades which are not the most efficient at converting mechanical power to airflow. That is why you never see an airplane propeller of this design. More efficient fans utilize blades which are engineered airfoils rather than flat paddles. If you do a search on fans with the highest "Energy Star" ratings you will see that those have fan blades rather like airplane propellers.
For the purposes of my intended experimentation I wanted the ability to vary the fan power to enable finding the minimum value to be useful. From the research discussed above I knew I wanted a fan with a BLDC design and blades that at least suggested some effort was made at airfoil design. What I ended up getting was a 20" fan rated for use on 24 VDC with a nominal power of 6W. There is an identical version of the same fan available rated for use at 12 VDC. I went with the 24 VDC version as I suspected that it might continue to operate down to a lower fraction of it's design voltage than the 12 VDC version. Ideally the 24 VDC version could be made to do something useful all the way down to 12 VDC. Also favoring the 24 VDC version was it's availability from several US based vendors vs the 12 VDC version which would entail the wait of shipment from China. Frankly, judging from the pictures these looked like cheap plastic toys. A weight of just under 1/2 lb did nothing to dispel that impression. Keeping an open mind I clicked on "Buy". Well these things can move an impressive amount of air for very little power. The particular unit received was capable of starting down to a voltage of 11.2V and run down to a voltage of 9.5V. Plotting power vs input voltage suggested that at the lower end of the range it was running into limits on the control circuitry before limits of the electromechanical factors. Time to open it up and have a look at the circuitry which proved to be easy to reverse engineer. This is a BLDC single phase two coil 20 pole motor design with the two coils driven by a half-bridge driver. About as simple a drive circuit as would be possible. By changing the value of a single resistor I was able to eliminate the low voltage being the limiting factor for the control circuitry as well as reducing it's power consumption a bit. The fan would now start down to a voltage of 7.5V and run down to just slightly lower. At 8V it draws only 0.99W but was not judged to be a useful airflow at this value. At 10 V it draws 1.46 W which seems to be as low as it can go and produce useful airflow. I decided to do my testing at 12.5 V which corresponds to a power of 2.14 W. I suspended the fan over my bed as close to the center side to side as permitted by the placement of the ceiling joists and about 2/3 of the way from the foot to the head of the bed. Distance from fan blades to bed measured at 54". At the tested power level of 2.14 W airflow was very sensible and fan noise much less than the normal HVAC noise level. My testing went down hill from here. My plan was to run the heat up incrementally to find the maximum tolerable for sleeping on top of the sheet with no bedclothes without the use of the fan and then do the same with the use of the fan. Unknown to me until I tried it, my thermostat while permitting the AC to be set as high as 99F, only permits the heat to be set as high as 80F. I can sleep on top of the sheet with no bedclothes at 80F. It now became a somewhat subjective judgement of how much benefit the fan provided. All I had to go on was the amount of my body I needed to cover with the sheet to be *warm enough* at 80F with the fan running. My judgement, admittedly subjective, is that the benefit is in the range of 5 to 10 degrees F. At a higher voltage of 16 V the fan draws 3.22 W at which level I would judge to benefit to be on the order of 10 to 15 degrees F. That is a lot of potential comfort benefit for very little power. At under $20 I would judge this to be a very high cost/benefit emergency preparedness item. I suspect that you could be selling at these at huge markup out of the back of a pickup in Louisiana at the moment where 100s of thousands are still without power after the recent hurricane. A car battery could power one 24/7 for at least a week. How durable are these? If your parakeet flew into the blades it would be more likely to survive than the blades. Barring that type of mishap, expected runtime to failure is an unknown. Analyzing the electronics I see no problems there. The limitation likely will be the bearing arrangement which supports the rotating components which total under 5 oz in weight. I currently have had one running mostly 24/7 for 3 weeks now with no evidence of degradation.
I may edit this post at a later date with more details about the technical aspects of it's design as well as more detail on my testing methods and resulting data for those here with a greater engineering/technical orientation.
Photo below as installed over bed for testing.