Originally Posted by LTD02
i would *really* like to agree with you. drivers that minimize flux modulation such as those manufactured by bms have very low distortion numbers. i want to believe your power compression claims as well.
HOWEVER, with no measurements, how are we to know? jbl has measurements for the 2226h for frequency response, distortion, and power compression. where are the measurements for the ae drivers???
I'm aware of the JBL paper as I have seen it referenced many times. What I find interesting is that I've seen papers from EV and others that show their own drivers to be better in terms of power compression and distortion than the JBL2226H. It is quite easy as a manufacturer to come up with a scenario to make your drivers measure superior. That scenario may mean nothing for someone else's application of the same drivers. We've been asked before for measurements to compare to other manufacturer data and without knowing the exact measurement methods and conditions it's pretty much impossible to do.
In terms of power compression you need to take into account the enclosure or lack of enclosure used for the test. If it is open baffle, the drivers with softer suspension have an advantage as the coil moves farther at low frequencies with the same input power and in doing so cools better. Stiffer suspension keeps excursion down more at lower frequencies and there is less airflow for cooling. In an enclosure the system Q also affects excursion. The JBL 50-500hz signal will tell a different story than a 20-200hz signal or a 100-1000hz signal as well as excursions are different. You need to know the crest factor, the slope of the filters to cut the frequency range, etc. As you go up in frequency, drivers that rely on air flow for cooling like the 2226H will have less advantage than at the lower frequencies.
Distortion on the other hand changes a lot in large drivers based on the enclosure, baffle size and mic position as well. Up close you get some additional issues caused by reflections off of the surround that would not be there at 2m distances. Distortion measurements between manufacturers like JBL and EV even are not directly comparable. One measures distortion on a large 10x10ft baffle with a large sealed enclosure behind it at 2m with a particular AES standard input signal. The other measures with a totally different method. While they both provide data that "looks good" and gives you some nice THD percentages, you can't say which one looks better as they aren't comparable. THD also doesnt' mean a whole lot different orders of distortion are more or less audible. In reality the only optimal way to compare 2 drivers is in the alignment you intend them to be used in. Our OEM customers who have done this have validated for their own needs many times the superiority of the Lambda motor in both distortion and power compression.
The vented cooling in the JBL does help compared to a standard design. It needs airflow to work though meaning high excursions. Showing measurements factoring in higher excursion levels at 50hz doesn't show much if you plan to crossover to a subwoofer at 100hz as you now have only 1/4 the travel on the driver and your air cooling factor is greatly decreased. Air however is not a good medium for cooling anyway. It requires large surface area to be effective. The area in the 2226h is quite minimal. Compare air cooling to other methods of cooling. Only small gasoline engines can be air cooled and larger engines need liquid cooling. Amplifiers, CPU's, etc need heat sinks first to pull the heat away quickly, and then a fan to aid in the cooling of the larger heatsink. A VC is similar.
The first thing you need to do is get the heat away from the coil. To do this it needs to be transferred to the top plate or the pole. These steel pieces have tremendous thermal mass in comparison to the VC and have huge surface area to dissipate the heat as well. The larger the air gap, the harder it is to transfer heat to the steel. The other main key is in the surface area for heat sinking, not just the diameter of the coil. Drivers with thicker top plates have more outer heat sinking area close to the coil and can pull heat from the coil better on the OD. Drivers with a kapton former or similar are insulating the coil on the ID which drastically reduces heat transfer to the pole. Drivers with an aluminum former pull heat from the coil on the ID and get it close to the pole. As the alum former has even larger surface area than the coil windings alone, it is quite effective at transferring heat to the pole. What we do to greatly increase this heat transfer is sleeve the whole pole with copper. This is a thick copper sleeve, not just a thin layer of plating. While the steel pole can absorb huge amounts of heat, it doesn't do it very quickly. The copper is much more thermally conductive. It can absorb heat much quicker from the air gap and then as it is rigidly affixed to the steel pole there is direct transfer of heat to the steel. The same concept is used on the bottom of frying pans with copper to help them heat faster.
If you look at the drivers with more surface area (in this case larger diameter coils) they all perform better in terms of power compression. The TAD with alnico slug has to have a former that insulates the coil from the alnico, as alnico begins to demagnetize more with heat. As a result it suffers in terms of power handling. It is not typically a driver that is ever used with very high power though so I'm not sure why it is in the comparison. Alnico is also less prone to flux modulation as it has higher flux density. This means lower distortion and reading JBLs claims shows this about the TAD "Generally, the performance is good, with distortion components in the mid-band uniformly 35 dB below the fundamental, similar to the JBL 2225H.
" At lower than their 115dB test levels, the TAD will have lower distortion than the JBL2226H(their best) even due to lower flux modulation, but of course the JBL paper won't tell you that.
That all said, we have a massive 19mm thick top plate around the OD of the coil. This combined with very tight tolerances and the addition of the full copper sleeve on the pole makes them very efficient at transferring heat away from the pole. We then go a step farther and have a huge solid billet aluminum phase plug on the front of the pole. This further pulls heat out from the pole. Now we have a huge heatsink and the large cone is blowing air back and forth over this phase plug to help cool it. In addition to that, by being round, you have the convection cooling of air going around the phase plug to remove heat as well.