It suddenly occured to me (after coming across a number of speakers without a label) is there a home way of testing a speakers potential power rating or is that company only stuff?
Sort of, for woofers usually you can monitor the amps output voltage and the excursion of the woofer and determine the excursion limited power handling in box for a box woofer and on just a baffle for IB/dipole woofer. This only applies to drivers that will be driven to low frequencies where excurision requirements are high ~<60-80hz. Try different frequencies and you can find out how little power is required to excurse a driver as you lower frequency.
For testing the thermal limits I would suggest reading up on the resistance change of copper magnet wire as a function of temp and see if there is a transition point. With this info you could apply an increasing signal for short durations, remove, and measure Re (DC resistance) and plot the change in Re. This will enable you to find the max short duration thermal limit of the coil. To find the rms rating you are going to have to reduce the max power just found by about half and then conduct long term tests to analyze the cooling capability of the motor structure, by long term I mean when you reach steady state (Re remains constant) at this point resistive losses of the coil are equal to the heat loss to the environment from the motor. Make sure to use either pink noise or sine freq. in the passband of the driver.
If Re keeps climbing on the long term tests then you have exceeded the cooling ability of the motor.
Sounds fun doesnt it!
Oh yeah some people use the "sniff" method, which is apply increasing amounts of power until you smell the enamel insulation burning and then stop and let everything cool. Noting the power that caused the odor, you can lower the input and try again, if no odor arrives you found the rms rating.
This has the potential though for shorting the coil or starting a small cozy fire.
Manufacturers know the coil gauge, length, insulation properties and any pertinent geometric factors and therefore know the powerhandling of the coil. All they need to do is assemble the driver and run a standard test to make sure that no damage results, this is where rms, peak and program ratings come from and are conducted usually with pink noise.
This is how I would do it if I actually cared. My philosophy is if you dont know dont push. Turn it up until you are happy with the output, if they fail they were insufficient for the job and need replacing.
For testing the thermal limits I would suggest reading up on the resistance change of copper magnet wire as a function of temp and see if there is a transition point. With this info you could apply an increasing signal for short durations, remove, and measure Re (DC resistance) and plot the change in Re. This will enable you to find the max short duration thermal limit of the coil. To find the rms rating you are going to have to reduce the max power just found by about half and then conduct long term tests to analyze the cooling capability of the motor structure, by long term I mean when you reach steady state (Re remains constant) at this point resistive losses of the coil are equal to the heat loss to the environment from the motor. Make sure to use either pink noise or sine freq. in the passband of the driver.
If Re keeps climbing on the long term tests then you have exceeded the cooling ability of the motor.
Sounds fun doesnt it!
Oh yeah some people use the "sniff" method, which is apply increasing amounts of power until you smell the enamel insulation burning and then stop and let everything cool. Noting the power that caused the odor, you can lower the input and try again, if no odor arrives you found the rms rating.
This has the potential though for shorting the coil or starting a small cozy fire.
Manufacturers know the coil gauge, length, insulation properties and any pertinent geometric factors and therefore know the powerhandling of the coil. All they need to do is assemble the driver and run a standard test to make sure that no damage results, this is where rms, peak and program ratings come from and are conducted usually with pink noise.
This is how I would do it if I actually cared. My philosophy is if you dont know dont push. Turn it up until you are happy with the output, if they fail they were insufficient for the job and need replacing.
There are standardized methods to test a driver's power handling capacity, IEC has one I think it states that the driver should stand a noise signal for x hours with a well-defined spectrum without permanent damage. I think it is hard to test the max power without actually damaging the driver. The smelly version above definitely damages the driver.
I could imagine that one could feed a signal to the driver and measure the temperature of the coil by noting the voltage and current (and calculating the resistance, which is proportional to absolute temperature). If you decide on a maximum temperature, then you could experimentally determine what power is required to reach that temperature.
I could imagine that one could feed a signal to the driver and measure the temperature of the coil by noting the voltage and current (and calculating the resistance, which is proportional to absolute temperature). If you decide on a maximum temperature, then you could experimentally determine what power is required to reach that temperature.
funny i was thinking about something similar last night
my take on it is that you're only really interested in how much power it takes before power compression occurs anyway (what's the point of adding more power after that)
play pink noise through it and watch an SPL meter whilst slowly turning it up
it'll reach a peak and then start to drop off as the power compression takes effect
my take on it is that you're only really interested in how much power it takes before power compression occurs anyway (what's the point of adding more power after that)
play pink noise through it and watch an SPL meter whilst slowly turning it up
it'll reach a peak and then start to drop off as the power compression takes effect
synergy said:
That doesn't sound right to me. I though power compression was when you receive less output for a given increase in input, causing the driver to be less efficient at higher power levels.
For example, when you double the power to the driver from 1 to 2 watts you get a 3db increase in acoustic output. When you double from 2 to 4 you get 2.99. From 4 to 8 you get 2.98. Eventually this becomes significant, where doubling power only gets you 1-2db more output. This is why dynamics become compressed when speakers are driven close to their limits.
....
Thinking about this some more, I see how it's possible for output to start dropping off if you're increasing voltage instead of power. As the voice coil heats up, and the impedance increases, less current flows. Eventually the increase in impedance could be large enough that power drops despite the increase in voltage, causing acoustic output to fall. I don't think the driver would survive this test.
Dan
- Status
- Not open for further replies.