Ask an Acoustic Engineer (me)

[Justin Zazzi]

Am I able to reply before geolemon?

So thats a super interesting graph. If im reading that right. We have very little reverberation in a car until basically below the Schrodinger frequency( i think im using that right). If everything below 100hz is already in the green, Whats the point of a bass trap? Is it too make the bass frequencies more flat and around the 0.1, 0.2 level?

If his idea is to mimic the reverb of a large listening area, how does one increase reverb in a car?

and what is reverb?

Cars tend to have high decay times in the lower frequencies and low decay time in the mid and high frequencies. Bass traps reduce the decay time of the lower frequencies and adding diffusors increase decay time in the mid and high frequencies. The result is a more even decay time across the frequency range. Look at the RT60 of my car when I measured it to see what a non-treated car looks like.

Reverb is the effect when you talk or sing or play an instrument, the sound will bounce around for a while causing echos before eventually dying off. Reverb is a very specific kind of echos that are pleasant. Stella does a great job explaining:

also straight from Rew.
"RT60 is a measure of how long sound takes to decay by 60 dB in a space that has a diffuse soundfield, meaning a room large enough that reflections from the source reach the mic from all directions at the same level. Domestic rooms are usually too small to have anything approaching a diffuse field at low frequencies as their behaviour in that region is dominated by modal resonances. As a result RT60 is typically not meaningful in such rooms below a few hundred Hz. Use the waterfall, spectrogram and decay plots to examine the decay of low frequencies in domestically-sized rooms."

So maybe rt60 wont work for cars?

My understanding of waterfall graphs is basically that sound is sticking around longer at certain frequencies. What that means exactly or how that sounds. I dont actually know.

Like all audio things, we rarely work in an ideal environment so we make the best with what we have. We can use RT60 in a car but it won't be textbook reliable. Waterfall plots are a way to view sound decay vs time, yes, and RT60 is a way of describing the shape of a waterfall plot or how steeply waterfall shape "falls" down into the bottom of the graph.

[jtrosky]

Ok Justin - I have one for you that I hope you can answer with 100% certainty...

Is it possible for a speakers "Voice Coil DC Resistance" to be *higher* that it's stated "Nominal Impedance" spec?

I have some speakers that measure 3.4 ohms with a multimeter, but the manufacturer is saying that they are absolutely 2-ohm speakers. Is this possible?

Thank you!

[jtrosky]

Impedance is frequency dependent and nominal impedance is a general average of a speaker’s impedance curve. So it makes sense that a multimeter taking a single measurement can’t measure the minimum average impedance over a frequency range.

See - it was my understanding (and most others that I talk to) that a "Nominal" impedance spec is the impedance at it's lowest point (aka "nominal") - not an "average" of the impedance curve. If you see an impedance curve diagram for a speaker, it's usually the point where the impedance is at it's lowest (from what I've seen). I've never heard of "nominal impedance" as being an "average" impedance?

What I am trying to understand is if the "Voice Coil DC Resistance" value is, in fact, directly related to the "nominal impedance" spec or not - and if the "DC Resistance" value is *always* equal to, or lower than, the "Nominal Impedance" spec or not.

Again, this is my understanding - I'm not saying I'm absolutely correct - which is why I'm here looking for clarification. :-)

In my experience, the value measured with a multimeter is always lower than the nominal impedance spec - but that is just form *my* experience - it doesn't mean that is *always* the case.

[ckirocz28]

See - it was my understanding (and most others that I talk to) that a "Nominal" impedance spec is the impedance at it's lowest point (aka "nominal") - not an "average" of the impedance curve. If you see an impedance curve diagram for a speaker, it's usually the point where the impedance is at it's lowest (from what I've seen). I've never heard of "nominal impedance" as being an "average" impedance?

What I am trying to understand is if the "Voice Coil DC Resistance" value is, in fact, directly related to the "nominal impedance" spec or not - and if the "DC Resistance" value is *always* equal to, or lower than, the "Nominal Impedance" spec or not.

Again, this is my understanding - I'm not saying I'm absolutely correct - which is why I'm here looking for clarification. :-)

In my experience, the value measured with a multimeter is always lower than the nominal impedance spec - but that is just form *my* experience - it doesn't mean that is *always* the case.

If I'm correct, a moving voice coil will have a higher impedance than a stationary voice coil.

[Jdunk54nl]

Isn't the Re spec what you should measure for dc resistance?

Usually Re and impedance are close, but impedance is frequency dependent as mentioned.

[jtrosky]

Isn't the Re spec what you should measure for dc resistance?

Usually Re and impedance are close, but impedance is frequency dependent as mentioned.

Yes - I believe so. I believe the Re spec is "Voice Coil DC Resistance".

In my specific case, the "2-ohm" speaker is reading "3.4 ohms" DC Resistance, which doesn't make sense to me - but maybe it is possible? Dunno...

I completely understand that impedance changes with frequency, but my understanding is that the "Nominal Impedance" is the absolutely lowest the impedance will go - and that the DC Resistance would always be equal to, or lower than the "Nominal Impedance" (not higher like I'm seeing).

As a "for example" - if a speaker measures 3.4 ohms with a multimeter, that *usually* means that it's a 4-ohm speaker (nominal impedance). I'm trying to understand if that is *always* the case or not... I've never seen a 4-ohm speaker measure as 4.6 ohms, for example (measure higher than the nominal impedance rating).

[Justin Zazzi]

Nominal Impedance is based on the lowest impedance which is usually the DC resistance and not an "average" of the frequency-dependent impedance.

There is a industry standard that states the minimum impedance must be no less than 0.75x (or 0.8x?) of the nominal impedance. So a Re of 2.6Ω couldn't be listed having a nominal of 4Ω because it is too low. I don't remember the spec off the top of my head, but I have it at the office ... it's one of those EIA or IEC or AES standards.

I have not seen any rules about the Re being higher than the nominal, and I have seen it a few times such as Nominal=4Ω and Re=4.3Ω. The few times I have seen this, the Re is very close to the nominal and not almost double like the example you give.

I have not seen one so different like you mention. My first thought is a dual 2Ω nominal voice coil might measure 3.4Ω when wired in series because each coil would have an Re of 1.7Ω which is perfectly reasonable for a 2Ω nominal.

The consequence of something like yours might be getting less output than expected and most people wouldn't notice. The consequence of marking a low impedance woofer with a higher nominal number might be amplifier instability or triggering a protect mode which is much more noticeable.

[jtrosky]

Thanks Justin. In this case, the speaker is just a regular 6x9 with only one voice coil. So do you think it's even possible for a 2-ohm speaker to read 3.4 ohms with a multimeter? Or is it 100% NOT a 2-ohm speaker? I guess I'm just trying to understand if it's even possible - regardless of how "un-common" or 'unlikely" it may be.

Another unrelated question for you.... :-)

When setting up time alignment, is there any advantage in measuring from left speakers to your left ear and right speakers to your right ear? I mean measuring both to the "center" of your head would just seem like it would be off by a few inches? Or is that just not how it works (where the sound from the left speakers go to your left ears first and right speakers go to your right ears first)?

Thanks again - I really do appreciate the feedback!!

[cueball981]

Thanks Justin. In this case, the speaker is just a regular 6x9 with only one voice coil. So do you think it's even possible for a 2-ohm speaker to read 3.4 ohms with a multimeter? Or is it 100% NOT a 2-ohm speaker? I guess I'm just trying to understand if it's even possible - regardless of how "un-common" or 'unlikely" it may be.

Another unrelated question for you.... :-)

When setting up time alignment, is there any advantage in measuring from left speakers to your left ear and right speakers to your right ear? I mean measuring both to the "center" of your head would just seem like it would be off by a few inches? Or is that just not how it works (where the sound from the left speakers go to your left ears first and right speakers go to your right ears first)?

Thanks again - I really do appreciate the feedback!!

Jtrosky,

When Justin was taking me through the ropes (...and I have a ton more rope to go through...trust me!), he had me hold my mic in my hand and then sweep it from ear-to-ear for several sweeps to find the average response from each speaker.

[jtrosky]

Jtrosky,

When Justin was taking me through the ropes (...and I have a ton more rope to go through...trust me!), he had me hold my mic in my hand and then sweep it from ear-to-ear for several sweeps to find the average response from each speaker.

Sorry - let me clarify - when I say "measurements" here, I'm talking about tape-measure *distance* measurements - not frequency response measurements (I also do ear to ear measurements for frequency response measurements).

For example... When obtaining time-alignment distance measurements, should I measure the distance from the left speaker dust cap to the left ear and the right speaker dust cap to the right ear? Or should I just measure from each dust cap to the tip of my nose, for example? To me, the results could be a few inches different depending on how you do it (I've seen both ways mentioned). Not to mention that if your measurements are also a little off, the difference could be even greater.

I'm just curious how much it matters and which way is technically the "best" way. Obviously, you can always "fine tune" the time alignment distances later with band-limited pink noise, but if you want to get as close as possible from the get-go, which way is the "best"?

Thanks!