Testing BG Neo10 at high power

[cvjoint]

Some interesting discoveries. Here is a plot of the harmonic distortion at vol 50 on my source unit:



Here is the same speaker just 1 db louder, at vol 51 on the source (my source increases output 1db/step):



Notice that huge spike in distortion around 200hz. I actually heard it through the ear plugs before I saw it on the screen. It seems the diapharagm shrieks audibly. It does so everytime I get past that volume level, and gets worse with more power. That is one defining characteristic of planars vs. cones, when they reach the max, they shriek, and you hear it clearly.

This was a design using the BG Neo10 as a midrange in a car, EQ. was applied to flatten the FR from the listening position, so the HD plots are not relevant in an absolute sense. Relatively, I just increase output 1db to show how the planar has a very sharp increase in distortion at some point. I didn't find this behavior with any cone type speakers.


The right one at vol 49:



At vol 50 it shrieks again, but this time around 1.3khz. 3rd and 5th order spike at that point. Why are they hitting the brakes at different frequencies? Because these plots are post EQ. I equalize +-5db, which means some frequencies ask up to 10db more output from the driver to get flat response. Such is the car environment.

Note the smoothing is 1/92. That's one step away from no smoothing at all. It's the least smoothing I can apply so that the results are readable, but this way all the distortion shows, no matter how narrow the peaks are. This is my favorite setting.

At this point I think I can also safely say that planars don't require a lot of power. What is the point of 200w to these if they shriek at 60w? The best thing I can do is add another BGNeo10, not double the power.

So from this exercise I would say there is some evidence that BG planars have hard limits. When you hit them odd order 3,5th harmonics spike. For that matter running a lot of power to a planar doesn't seem to matter. They are efficient and hit the brakes hard. Best thing to do with this technology is to use multiples, or larger membranes. At least that's my interpretation.

[studiotech]

Are these the same units from your other post of them in the garage or in the car pillars? When they are in the pillars, is there any absorptive material behind them?

Greg

[cvjoint]

Yeah, same pair I've been testing. I'm using enclosures that are 2" or more deep and have 1" open cell foam glued on the back of the enclosure. There is about 1" of open space between the foam and the back of the driver.

[twest820]

Originally Posted by cvjoint

That is one defining characteristic of planars vs. cones, when they reach the max, they shriek, and you hear it clearly.

The only time I've measured behavior like this was due to mechanical resonance in a Neo3's mounting. Never hit it in a Neo10.

[cvjoint]

Linkwitz tested the Neo3 and he reported shrieking as well:

"The microphone signal amplitude decreases over the 100 ms duration of the burst. It is probably due to the driver heating up and leads to a 2.3 dB signal amplitude change within this short time when about 5 W were applied. The tweeter shrieks audibly during the test."


Spatial distortion


Linkwitz thinks it's a thermal breakdown. Personally, I don't think so, at least not in the Neo10 case. The diaphragm probably cringes, it's too sudden to be thermal. I move the volume knob 1db up and down and it is predictably always there on the sine sweep 1db over the limit.

I never hit this limit either when I used them in the nearfiled, I moved them a bit farther away which increase the output requirements and every now and then I hear them flip out. Output density in technologies like ribbons, AMT, or planars seems to be very low. Very large elements are needed to rival conventional cones and domes. Still my favorite midrange ever, but it requires a lot more surface area than I originally planned out for.