Mazda3 Forums banner

1 - 20 of 49 Posts

·
Registered
Joined
·
1,201 Posts
Discussion Starter #1
I usually laugh at my mom when she tries to understand computers. "Megabytes, megahurtz, RAM, ROM, harddrive, what's does all that crap mean!?" But now looking at car audio, I feel just the same way, and I'm not having much luck on finding any literature on the topic (although I haven't hit the library), so I figured I'd just ask you guys (and hope you don't treat me like I treat my mom).

So, I guess I'll start with the basics? What's the basic design of an audio system? As I understand it, there's the head unit which outputs to the amp which outputs to the speakers, or one can add an intermediate equalizer to separate and control audio channels? Guess I need some clarification on the options one has in this area.

Also, speaker terminology. What’s the difference between 2/3 way speakers, coaxial, and components?

Lastly, if nothing else, does anyone have any links to information about this stuff?

Thanks.
 

·
Registered
Joined
·
4,849 Posts
Great topic!!!!!!

I'm a newbie to audio (car and home) too and know VERY little on the fine points of it.

So our resident audiophiles are invited to take the floor.
 

·
Registered
Joined
·
302 Posts
Ok just to start off you have the basics down, and ot answer some of your questions...the difference between 2/3/4 way speakers is the amount of drivers the speaker has, a standard 2-way speaker has 2 drivers one for the lows(a.k.a. subwoofer) and a driver for the highs(a.k.a. tweeters). The 3-way speakers add an additional driver for mid-range sounds. And as far as four way speakers it varies from manufacturer but it is usually just a super-tweeter for the very high freq. sounds. A coaxial connection is just a type of cabling which can be used to carry digital or analog information, this is mainly used for RCA cables to connect amplifiers. As for component speakers these are a set of speakers which basically just seperate the 2 drivers form a 2-way speaker into a small subwoofer and a tweeter which are generally much higher quality than just a regualar 2-way due to the fact that they are not competing for air space etc. As for as adding an equalizer after the head unit in my personal opinion unless you are a stickler for sound quality or enter competitions they are unnecessary but then again that's just my .02.
-Phil
 

·
Registered
Joined
·
17 Posts
So how do you guys feel about RMS and WATTS. from what i'v been told the "watts" is just marketing at it's best, that what under theroy it's peak, RMS is the dediacted amount per channel, heince 500Watt amp may never reach that much but a 100rms amp gurentys 100watts per channel.
is that correct or not?
 

·
Registered
Joined
·
2,756 Posts
RMS = Root Means Square, or power that works at a general pace. Typically this is the number you want to know. Its the constant wattage the amp can put out. Peak usually is double this number, or also known as MAX.

Almost all amps show a RMS rating. If not, MAX can safely be said its 2x the amount of RMS. IE: Amp says 500w, it is safe to assume the amp is 250w RMS. If it was a 4 channel amp, it would be 62.5w x 4 RMS.
 

·
Registered
Joined
·
1,201 Posts
Discussion Starter #10
Okay, here's another question-

When you're looking at speaker specs, what specs do you want to focus on to gauge the quality of them?

Also, can you really tell how a speaker will sound in a car wihout actually installing them? Wouldn't the acoustic environment of the car change the sound of the speakers dramatically?
 

·
Registered
Joined
·
2,756 Posts
Some big factors would be: frequency response, power handling, impedence and sensitivity. Heres a quick glossary of what they mean:

Frequency is the measurement of sound, in Hertz (Hz). Its a spectrum of sound between what the ear can hear, 30Hz to about 20k Hz. Frequency response means what the speaker can reproduce. Typical is 20Hz-20k Hz.

Power handling is what the voice coil inside the speaker can physically handle, measured in watts. Higher the number, more power you can give to it.

Impedence is the resistance given off by the voice coil. Measured in ohms (omega sign). This is important as you dont what a super low impedence going to an amplifier that cant handle it. Most 4 channel amps are 2 ohm stable. Most single voice coils are 4 ohms.

Sensitivity is the measurement of sound output for a power input. This number usually is in db/Wm (decibel per watt at one meter). IE: a speaker says its sensitivity says 88db, that means it will produce 88db of sound at 1 meter.

Speakers always sound different in different environments. I tell my customers to listen for the highs, because the higher frequency the more directional sound becomes. Dont worry too much how a speakers' midrange sounds. If the higher frequencies sound good to you, chances are they will be close in the car too.
 

·
Registered
Joined
·
272 Posts
The Impedance spec is important for one more aspect:

when driving a passive compenent (speaker) with a source (amp), the impedance should BE MATCHED, otherwise there is a reflexion in the transmission line (wires) of the power and the system lost its efficiency.

This Impedance-Matching phenomenon is well known, and one can find the explanation in all electronic text-books.

PS: Impedance is not exactly a resistance (Impedance is a complex number, and only the real part is the resistance, the imaginary part is either from the Inductance/Capacitance of the circuitry, the resistance - when driven - produces heat, the inductance/capacitance stores the loss-less electromagnatic energy)
 

·
Registered
Joined
·
315 Posts
bluong1 said:
The Impedance spec is important for one more aspect:

when driving a passive compenent (speaker) with a source (amp), the impedance should BE MATCHED, otherwise there is a reflexion in the transmission line (wires) of the power and the system lost its efficiency.

This Impedance-Matching phenomenon is well known, and one can find the explanation in all electronic text-books.

PS: Impedance is not exactly a resistance (Impedance is a complex number, and only the real part is the resistance, the imaginary part is either from the Inductance/Capacitance of the circuitry, the resistance - when driven - produces heat, the inductance/capacitance stores the loss-less electromagnatic energy)
Eh, no. Considering a couple of things:
1. Unlike the electromagnetic fields simple model for a TL, a component set (along with crossover) is MUCH different. Why would it be mandatory to match impedances of speakers (which is impossible considering each driver is very different) when you've got opamps, caps, inductors, and resistors that would be different for each driver to create the crossover effect? The only way to "impedance match" is to use two of the same drivers. Even then, they aren't perfectly matched. Its just like caps generally having the +/- 20% tolerance... they're not always going to come out exactly the same.

2. I would worry about losses in the components (caps, inductors, resistors) in the crossovers before worrying about any TL losses. Switch to active crossovers and you lose a lot of those losses since you're eliminating the inductors.

If anyone needs a little more explination on the mumbo-jumbo from above, I'll try to convert it to english :) This stuff is 3rd and 4th year stuff for EE students, so I don't expect everyone to know this stuff!
 

·
Registered
Joined
·
272 Posts
Hummm,

I do read few articles from high-end stereo companies that insist on the matching between amp/speaker (may be not necessary perfect, but nevertheless one has to pay minimum attention). In those references, they don't quantify the loss, so I admit I'm a little blind with their explanation.

Why you think the EM field inside the copper wire is different than inside the TL? Afterall, aren't Maxwell equations universal? Or the loss is negligible at the sonic frequency (20-20000 Hz)?

PS: It's not a retort question but I want to learn, and It's great if you can convinced me.
 

·
Registered
Joined
·
315 Posts
If you could post links to those articles, that would be great. I don't want to speculate, but sometimes companies try to use terminology to justify cost and things of that nature. You might have seen companies selling interconnects for hundreds of dollars claiming things such as it stops skin effect (electrons go toward the outside of the wire and don't flow uniformly for people that don't know about skin effect). If you do the math, you find that skin effect doesn't have a decent effect until you hit frequencies MUCH higher than what any human can hear... MHz and such.

I'm not saying that copper should not be included in the transmission line equations/theories. I'm saying the parallel loads (assuming a tweeter and woofer in the component set) are going to be completley different from each other based solely on the crossover components if we assume the speakers are ideally exactly the same. Without using equations and just looking at a simple S-domain schematic of a basic first order crossover (and assume that the drivers are an ideal resistor), you see the two loads will be completely different in terms of numbers. I hope this explains what I was trying to get at before. Give me a heads up if any of that did not make sense.

From what I have read (http://www.audioholics.com/techtips/audioprinciples/interconnects/SpeakerCableFaceoff.htm), since the resistane of wires is a handful of milliohms, the power loss is negligable.

audioholics.com also has more articles that are interesting to read. If nothing else, they use logic and math to show you how to save tons of money on buying cables by debunking myths. Here's link to the articles: http://www.audioholics.com/techtips/audioprinciples/cables.htm
 

·
Registered
Joined
·
272 Posts
Here is one link http://www.veritasaudio.com/accuart.htm

Let's me think about the problem a little more.

But here is a though from me:

- In case of a parallel load for one source, I'm sure one can't match all the loads at the same time as you said. But if there is the match between the impedance of the source with the sum of all sub-impedances, then it will match automatically for all for each parallel load. Right or Wrong? The hard part now is find the spec of equivalent parallel/serie L,C,R for speakers.

- Cable: Yes, I agree the skin depth of regular copper at 20 kHz is still large. But now this large skin depth is an obstacle for any efficient electromagnetic isolation. It's will be harder to prevent cross talk between cables. IMO good cables should avoid cross talk. Cable electric resistance is not an issue as you said.

Thanks for the link, I'll take time to read over those.
 

·
Registered
Joined
·
2,756 Posts
The suspension of the driver also makes a difference. If the coil was in the positive field the resistance changes. Yes, matching drivers is near impossible. But the tolerences are always well within spec of the listed impedence.

Bluong, your right about impedence not being 100% resistance. It is also a measurement of electromagnetic transfer. Easiest form to measure is resistance.

In car audio, expensive cabling has almost no use for typical speaker installs. Ive built systems (few have won awards) and used 16 gauge standard copper. I do recommend a good pair of shielded RCA cables.
 

·
Registered
Joined
·
315 Posts
bluong1 said:
But here is a though from me:

- In case of a parallel load for one source, I'm sure one can't match all the loads at the same time as you said. But if there is the match between the impedance of the source with the sum of all sub-impedances, then it will match automatically for all for each parallel load. Right or Wrong? The hard part now is find the spec of equivalent parallel/serie L,C,R for speakers.
Hmm, let me look that stuff up. My "specialty" is not EM fields, so I don't want to give a response. The only thing I can think of at this point is that amps and loads generally completely opposite. One is very high while the other one is very low. I'll look at the Veritas link and post a response tonight (it might be a big "i don't know").
 

·
Registered
Joined
·
2,344 Posts
- In case of a parallel load for one source, I'm sure one can't match all the loads at the same time as you said. But if there is the match between the impedance of the source with the sum of all sub-impedances, then it will match automatically for all for each parallel load. Right or Wrong? The hard part now is find the spec of equivalent parallel/serie L,C,R for speakers.
No. This is not true. Running in parallel, you would have two speakers double the imp. For example, if I wanted to run two 8" subs off a 4ohm amp, I would need two 8ohm speakers to match it. If I wanted three subs, I would need three 6ohm speakers.

The good news is that most mid-high grade amps will do an auto switch for you. At a minimum you will switch from 8 or 4 ohm, but if you are lucky (like my Phoenix Gold), it will actually drop to 2 or 1 also.

I may be a little off in the speaker matching, but I do know that you do NOT want to match speakers in parallel with the ohm on the amp. This will cause to heavy of a load and start a fire. (btw, that's what fuses are for - use them!)
 

·
Registered
Joined
·
272 Posts
Running in parallel, you would have two speakers double the imp. For example, if I wanted to run two 8" subs off a 4ohm amp, I would need two 8ohm speakers to match it. If I wanted three subs, I would need three 6ohm speakers.
I think you mean three subs of 12 Ohm.

- 1 sub -> 4 Ohm
- 2 sub -> 8 Ohm
- 3 sub -> 12 Ohm

And here is the general formula for n parallel Impedance

1/Z = 1/Z1 + 1/Z2 + ... + 1/Zn

you will find the above numbers work just fine. (This assume the Impedance is co-linear, i.e., the ratio between real and imaginary parts are the same, if not one have to do a complex algebra, that's why I said one need to know R, L, C of each the speaker. The module of the impedance alone is not enough).
 
1 - 20 of 49 Posts
Top