Estimating cfm and dba levels for undervolted Panaflo fans

DuckTape · Post by **DuckTape** » Thu Mar 13, 2003 3:52 pm

Hello folks.

I obtained the following numbers from http://fans.cuzuco.com/fans.html

PANAFLO FANS

80mm:
L1A (24 cfm, 21 dba)
M1A (32 cfm, 28 dba)
H1A (40 cfm, 32 dba)

92mm:
L1A (42 cfm, 27 dba)
M1A (48 cfm, 30 dba)
H1A (57 cfm, 35 dba)

120mm:
L1A (69 cfm, 30 dba)
M1A (86 cfm, 35 dba)
H1A (104 cfm, 42 dba)

The cfm and dba ratings provided by Panasonic would be for all of these fans operating at the default 12 volts, correct?

(My best guess would be that the above numbers would be the ones provided by Panasonic.)

Is there a way to calculate (given the above numbers) at least a rough estimate of the theoretical cfm and dba levels for these fans at the undervolted voltages of 5 volts and 7 volts?

Thanks much!
DuckTape

ez2remember · Post by **ez2remember** » Thu Mar 13, 2003 4:50 pm

Yes those specs are for 12 volts, it is pretty much linear so you can calculate both cfm and theorectical dBA.

For 7 volts
7/12 * 100 = 58.3333333333%

For 5 volts = 41.6666666667%
5/12 * 100

Example of Panaflo 80mm L1A @7v and 5v respectively
58.33% of 24cfm is 14cfm
58.33% of 21dBA is 12.25dBA

41.67% of 24cfm is 10cfm
41.67% of 21dBA is 8.75dBA

BUT TO SIMPLY MATTERS JUST USE THE FORMULA BELOW:

For 7 volts
7 / 12 * cfm or dBA

For 5 volts
5 / 12 * cfm or dBA

Update: The dBA rating are purely theorectical calculations.

I am not sure if the calcuations for noise level is correct, since it is logorithmic!

DuckTape · Post by **DuckTape** » Thu Mar 13, 2003 9:38 pm

Hello.

Thank you very much for your reply.

Well, I looked in a sound physics book, and as far as I can tell, it explains that for every rise in 3 dba, there is a doubling of sound pressure (loudness).

So, 4 dba is twice as loud as 1 dba, and 7 dba is twice as loud as 4 dba, and so on.

If indeed a decrease in voltage by half results in: a decrease in cfm by half, as well as a decrease in loudness by half -- then one could calculate the resulting dba level for 6 volts (half of 12 volts) as 3 dba less than the dba level for 12 volts.

I suppose calculating the theoretical dba for 6 volts could at least give an idea of the corresponding dba levels for the 5 volt and 7 volt levels.

Can anyone confirm that cfm and dba levels are, theoretically and for the most part, "directly proportional" to voltage levels in these fans?

Thanks again!
DuckTape

Post by **MikeC** » Thu Mar 13, 2003 10:13 pm

Have a look at the Scientific Fan Selection article. It's a little confusing; basically the second XL file is more accurate, IMO.

Franz · Post by **Franz** » Fri Mar 14, 2003 12:16 am

3db is about twice as much energy. Our ears need about ten times as much energy to think something sounds twice as loud. Also at low sound levels our ears are much less sensitive to low and very high frequencies. So somethings that have equivalent sound pressure measurements will sound louder or quieter to us depending on what frequency bands the energy lies in. Also if two sound sources, one near(say a monitor), one distant(say a fridge), are measured to have the same energy level/distibution at, say the computer chair, the closer source will sound louder. Also, also, some frequency bands are harder to ignore than others.

Franz

Politik · Post by **Politik** » Mon Mar 17, 2003 8:24 pm

From looking at the volts vs. CFM vs. air pressure chart, I dont think the relation between volts and CFM is linear.

Lower voltages will drop CFM more quickly at the same air pressures.

http://www.panasonic.com/industrial/app ... fba08a.pdf

dukla2000 · Post by **dukla2000** » Tue Mar 18, 2003 1:17 am

Politik wrote:From looking at the volts vs. CFM vs. air pressure chart, I dont think the relation between volts and CFM is linear.

Lower voltages will drop CFM more quickly at the same air pressures.

http://www.panasonic.com/industrial/app ... fba08a.pdf

I think the theory (rpm directly proportional to volts, and free airflow cfm directly proportional to rpm) is OK. The main problem with those charts is the printing of the curves. Looking at the U series, my ruler has the x-axis scale from 0.5 to 0.75 much longer than from 0.75 to 1.0 (approx 1.16 times). Also it seems the 'offset printing' has offset the blue curves to the right - there is no way I could read 1.33m3/min at 12V (which is the spec) from the curves.

The key to the theory is 'in free air' - once you have static pressure (i.e. the real world) then aerofoil design also plays a part. But as nobody has a clue what the static pressure of their case (or psu or hsf) is, we all happily manipulate and quote free air numbers!

BTW - the dB(A) calculation is theoretically
dB2=dB1-50*log(rpm1/rpm2,10) {in Excel format}
In the article Mike links to, SPCR Forum member gmJamez has built the main table calculating dB(A) by extrapolating manufacturers' data for a fan range, and has drawn curves to highlight this approach versus the log formula. I have no evidence to directly prove the log formula is valid, but am equally enough of an amateur to figure I have to accept it as I dont know better.