Car Stereo Forum banner

1 - 5 of 5 Posts

1,808 Posts
Discussion Starter #1 (Edited)

I've notice that few if any drivers that you can buy are actually stress tested to certain decibels of SPL for audible distortion/motor noise. So, I wanted to think up a way to manually stress test drivers as safe as possible but with some risk in a cost-no-option method. Certain drivers that are unable to produce the output might fail or blow, but if I'm willing to chuck them and buy/try another set until I get what I want, maybe it will work. Crazy? Maybe. The outcome when the right drivers are found should be exciting.

One thing I noticed while playing test tones is that even with a highpass that seems 'safe', when a 0db tone is played well below that highpass frequency, an amazing amount of bass can still reach the component speaker or midwoofer. For instance, in my tests on my comps I found that with an 18db/octave and even a 24db/octave filter were insuffient to keep low tones from generating enough bass at high outputs to prevent the speaker from creating noise unless of course the filter frequency is increased.

My comps when running 120Hz at 104db SPL at volume 25 produced no audible speaker distortion. But at the same volume and while set at 110Hz highpass, the speaker also produced 94db SPL at a 70Hz 0db tone which created audible speaker distortion and motor noise. Interesting results. When I moved the highpass frequency up to 150Hz it dropped that same 70Hz 0db tone to 87db and the distortion was inaudible. So, I decided to try it on a couple of other speakers and found the same thing.

I concluded that if someone wants to have a completely clean yet highly dynamic output, the transient peaks in the bass frequencies must be free of audible distortion. So, they better make sure the midbass/comp they are using can truly handle said transients without distortion at the crossover points they are selecting.

In this case, test tones aren't used to identify flat response, but rather test the boundaries and limitations of their drivers so they can properly set their crossovers to avoid introducing distortion that doesn't belong in the sound.

So, here is a summary of my procedure. First, the stress test based on 0db bass test tones from 20Hz to 110Hz. (Frequencies above 110Hz I found in my car produced no audible resonances at SPL outputs as high as 104db at 120Hz and as high as 110db at 200Hz. 110db at 500Hz. I became convinced through my tests that the critical range that drives most comps/woofers to audible stress distortion and rattles plastic car panels/metal are from 20Hz to 110Hz. The most offensive being plastic buzzing from 60Hz to 90Hz. These stress distortions are what I hear in most people's systems and they are either the speakers themselves or the materials of the car.

The 'stress test' standard for the equipment AND acoustic environment:

The amps for each driver must be capale of driving CLEAN RMS output as set with a DMM to:

Push the entire speaker system to 115db at 0db test tones from 20Hz to 100Hz in the car enviroment for a 3-second burst to simulate a transient peak as played on a music CD.


1. Defeat the subs & midbases then run the comps with the highpass still engaged on each of the test tones at the same volume then listen for distortion. If the speakers can't pass and you are hellbent on keeping the speakers you must either raise the crossover frequency or make the slope of the filter more steep to remove the audible distortion.

2. Repeat step 1 for each set of remaining speakers testing the subs and midbasses separately at the reference peak output of each frequency and listen for distortion/speaker noises.

Now if you don't hear the speakers making noises from being driven too hard, the speakers are verified as able to handle the peak output of any transient without distortion up to 115db. That's pretty loud.

Repeat the test again but this time locate audible acoustic vibrations and damp them or replace the panel(s) with alternate materials that don't resonate at the peak output of the system.

Now the vehicle's acoustics and it's sound system are verified to handle to the loudest dynamic output of any recorded material with no distortion or added noises that don't belong in the sound up to 115db.

This means from the headunit to the amps to the speakers and to the car itself, distortion is inaudible to the listener.

Finally, add an equalizer in line and REDUCE (not increase) all frequencies necessary to reach the desired response curve. DONE. That is what I would like to do, but it would cost money to go through failed speakers until I found the ones that could handle it. I know that after testing my comps, a highpass of 250Hz at 18db/octave should be sufficient to handle all transient bass peaks at what would be a system reference peak output of 110db. Just good enough for me to keep them. To go up to 115db I also could just make the filter more steep by buying a simple in-line RCA fmod if needed.

Ok, that's my kooky idea. It should work but It's rather ill-advised though. ;)
1 - 5 of 5 Posts