Audio Education/FAQs/Myths

[Chad]

Model specific car forums are a hotbed for mis-information about electronics and audio, it's not the user's fault, they are here to learn about the CAR not about a specific facet of another hobby. This place ran rampant with myths and lore, "Stanzor" started a thread long ago and it got lost, I'm going to try to revive it.

Feel free to post up links to FACTUAL information or learnings you can back up with numbers and scientific evidence. Lets keep everyone safe and sounding good!

Chad

[Chad]

An older one from the "old EPHatch"
------------------------------------

So instead of doing this Faq style I decided to just submit little articles to contribute to this thread. I am heavily involved in pro audio production and system design. I did a lot of consultation work and workshop speaking, one of my attributes was speaking on system optimization and my favorite and most well received topic was on “free power” because everyone likes free stuff right?

Many of my topics do not apply to car audio and some do, the ones that don’t apply to low end optimization and coupling and line array steering and optimization. We can’t put a line array in a car and low end coupling in free-field does not apply in a car with a transfer function (especially a TF like in the EP!)

The first thing I’d like to present is the spheal on “The first watt is the one that counts, buy drivers wisely!”

All drivers have a sensitivity specification, usually stated in 1W/1M This is the sound pressure level that a driver or system will produce with one watt of input at one meter distance. It is a VERY important spec for system optimization and purchase considerations. I’ll explain why, but first I’d like to share a dirty trick that seems to be popular now. Sometimes you will see the sensitivity rating at 1W/1M and other times 2.83V/1M, these two figures are the same for 8 ohm drivers but for 4 ohm drivers 2.83V is actually 2 watts! The marketing department is playing with your mind by swapping these numbers around freely and creatively, don’t be fooled.

Lets say Brand X specs out at 90dB 2.83V/1M and brand Y specs out at 88dB 1W/1M. We know that by doubling your power you gain 3dB so brand X is rated at 2 watts for the sensitivity rating whereas brand Y is rated at 1 watt. If all things were equal brand Y would make 91dB with 2 watts and brand Y would make 87dB with one watt. Brand Y is more sensitive after doing a little math and extrapolating the “truth in numbers”

In the audio world the fact that air is a very poor medium of sound transfer plays against us, big time! We deal with things such as inverse square law and the fact that actual power transfer thru air is far less than optimum. We won’t discuss inverse square law in a car because we are dealing with such close quarters but the efficiency matter is important. To gain 3dB you must DOUBLE your power, 3dB is not a lot, few people notice a 1dB increase, usually 2dB is noticeable, this is a lot to bank on in power doubling at high levels for the average listener. The way to calculate the SPL of a driver given the sensitivity and power delivered is as follows : Sens (driver sensitivity) + 10LogP (P=Watts delivered). For example a driver with a sensitivity of 88dB1W/1M driven with 50 watts will develop 104.99dB. The same power presented to a driver with a 91 dB1W/1M sensitivity would produce 107.99dB, the same as driver #1 with twice as much power. Big whoop eh? 100 watt amps are pretty common and power is cheap right?….. Wrong! It’s right for a while but once you start getting into subwoofers and higher levels things take a turn for the worse.

In the quest for smaller sub enclosures and more trunk space Kicker introduced the solobaric sub years ago, everyone followed suit. These drivers are awesome, now I can have low end and golf clubs in my trunk! Nothing comes for free, placing an over-damped driver in an over damped enclosure comes at the price of sensitivity, today’s sub systems are grossly inefficient, but power amps have come a long way! So lets take a couple real world subs that can be had today, this is no testimony to quality, just sensitivity ratings. These are drivers that would be common to fit into an EP, first the JL Audio 10W3V2-4 with a sensitivity figure of 84.87dB 1W/1M and the JBL P1024 with a sensitivity of 92dB 2.83V/1m (OOPS! JBL played the number game on us so we will edit this to 89dB), both are about the same quality IMHO and readily available and fit in the same size enclosure. (although the JBL is cheaper)

So now we go out and buy a 300W amp, it’s of good quality and has true RMS ratings (not an ISBL (if struck by lighting) figure). At full power the JL audio driver will make (rounded up) 85+10Log300 or 109.77dB. With the same amount of power the JBL will produce 113.77dB. As you can see the difference of the numbers do not change from the difference of the sensitivity rating. We know the JBL is 4dB more efficient therefore the JL will require OVER twice as much power to play at the same level. In the single digits of tens of watts that’s no big deal, but when we start talking about the price of 500 watt amplifiers versus 1KW amplifiers the price can become staggering! This is where it becomes important to select drivers that will meet your needs, if you are going for SPL go for as an efficient of a driver as possible that will shill have the frequency response that you desire.

So you decide you want the less sensitive speaker and will apply more power, that’s cool. BUT two more things play an important factor here. As you increase your power output you also increase the demands of your electrical system, it has a cap too. If your electrical system can only provide 750 watts to the boot then your 500 watt amplifier will work just fine, the 750 watt amplifier will starve for power and the voltage will sag. Battery capacity, cable loss under load, and the ability for the alternator to keep up play a role in this. You can count on an amplifier being 60-80 percent efficient with class AB being the least efficient and clad D (PWM) amplifiers being the most efficient. So a class AB amplifier providing 750 watts will require a minimum of (750/*.6/12)A to operate at the bare minimum at full output which equates to 104.167Amps, that’s just for the sub and not calculating line voltage loss but figured at 12V nominal. That’s a lot of current folks! Especially a lot of current for a system to maintain 12V at! You can see where speaker efficiency not only gets louder but at sane levels can cause less stress on your electrical system, thus costing you less money in auxiliary parts. A cap will not help, it does not make power but STORES power, the overall load on the electrical system will be the same plus the ESR loss of the cap (which is almost nil.)

Another factor to consider with applying more power is power compression in the driver. Voice coils get hot, heat causes the impedance of any circuit to rise, this rise in impedance is represented at the amplifier outputs and the amplifier puts out less power, combine this with the human interaction to turn up thus putting MORE power at the coil, getting it HOTTER and you can see where this effect is going. So you have the JL and JBL driver. Both drivers are putting out X and X-4dB, you up the power to the JL to mathematically raise the SPL to that of the JBL, this will only happen in the real world until the voice coil heats on the JL from the increased input and then the output reduces AGAIN! So in essence the sensitivity goes more and more out the window with the more power you apply to the driver. This, of course, is assuming that both drivers have the same ability to cool the cone. I KNOW that JBL spends a lot of time on R&D to reduce power compression, I expect no less out of JL Audio. Physics bites you again!

So we can conclude that for many reasons selective shopping for drivers will net you a more efficient and overall possibly less expensive system, even if the initial cost of the driver may be higher. Look carefully into these specs and choose wisely.

In conclusion I might add that as you have noticed the SPL numbers with 500 watt input are low for realistic numbers in a car. These measurements are taken free-space (anechoic). A car has what is called a transfer function. Anytime a driver plays in an environment with any dimension that is smaller than the wavelength played the waveform becomes Modal. Our cars have NO dimension that is longer than the wavelengths played by a true subwoofer. At this point the driver is simply pressurizing the cabin and there is a significant gain in output. The larger the cabin the lower this pressurization point becomes. So this is why the numbers are low for the subs. Transfer function brings them up quite nicely.

That’s enough for this installment, I’ll figure out the next session soon :)

[Chad]

How amplifiers bridge.....

Here's how it works......

All car amplifiers use 2 positives (driven) for bridging, all (most) amplifiers have a positive (driven) and a ground for the speaker. The negative of the second channel is actually the INTERNAL driven lug and the positive is the GROUND.

Here's why.

Bridging works by inverting the phase on one channel, you can still use two speakers wired correctly, they will be out of phase, wire the second backwards and they will be In phase and IN absolute phase. By going to the two positives you are doubling the voltage at the outputs (one channel is swinging up while the other is swinging down) in this app both speaker terminals are "live" and there is no ground reference.

Pro amps do this with a switch, the switch simply comes off the preamp stage of channel A, inverts it and drives channel B with that, this ensures the same gain on both channels, it's mono, driven with only one channel of preamp, you then go off the two positives for a voltage double, Chb's preamp just pisses in the wind, often thru a resistor to prevent internal oscillation.

Car amps are "pre bridged” although they are labeled +-+- they are wired internally +--+. The phase of channel B is flipped from the get go and chB's output labeling is backwards. This was hinted at in the original post of this amp when I suggested driving headphones then finding out I had to invert one channel. So when you wire + chA and -ChB you are actually wiring + and + internally and bridging happens by simply monoing the inputs and matching gains. Some amps have a bridge switch which is simply a summing network.

There is a couple reasons this is done and has proven in the pro world to make a difference. The technical reason is that most LF information is mono anyway. Let’s say it is hooked up "conventionally" and you get a kick drum hit. Since amplifiers are bipolar and half of a channel amplifies the positive pulse and the other half does the negative side, both speakers pull off of the positive and negative section of the power supply at the same time. By bridging the amp or flipping the phase of one side, during the exact same impulse each speaker is pulling from different sides of the bipolar supply. So on a positive swing chA is pulling from the positive side of the PS but chB is pulling from the negative side, flip the phase of the speaker and both cones move in the same direction but the power supply is effectively balanced. This has been proven to increase headroom.

Here's another explanation of the power supply thing :)

http://www.prosoundweb.com/live/articles/daverat/amptricks.shtml

Enjoy, let me know if I confused you, I'll assist, it's important to know this stuff!

Chad

[Chad]

And one I wrote today :msmooch: Some can probably guess why :mangel:

If I had a dollar for every time I have heard a certain type of enclosure recommended for a genre of music I would not be driving a Civic ;)

There is no enclosure designed or even attributed to certain styles of music, but different enclosure designs CAN have undesirable attributes if improperly implemented.

A sealed enclosure alignment is the easiest of all enclosed systems to build and is also the most forgiving, it has a predictable 12dB/Oct roll-off and said roll-off and response is normally rather smooth, for this reason people assume as a knee jerk reaction that a sealed alignment is the end-all-be-all for SQ and music “with no boom” This assumption is further reinforced by the multitudes of premade vented enclosures out there that give you absolutely no specifications other than the fact that it’s vented! Before going into visuals it’s important to understand how a vented (ported) enclosure works, it’s actually simpler than you would expect, the vent is simply a resonant chamber that when excited via the back-wave of the driver in the enclosure resonates in phase with the output of the front of the driver, this resonation causes a gain or bump in that frequency region due to acoustic summing. Now what you do with this tuning is up to you, you can tune a box for “SPL use” by placing the tuning frequency “bump” in a region of the sealed response that will yield a large bump in the frequency response.

Lets get into some graphs…..

For reference I’ll use the Dayton Audio RS265 (Dayton 10HO) it’s a very well-rounded driver that has remarkable distortion characteristics, decent X-Max and good power handling (real world power handling.) Even better is the fact that it is flexible in its enclosure requirements performing well in both sealed and vented alignments. It is also affordable.

It’s this one:
http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=295-462

All graphs were done in a .7CuFt enclosure.

Here is the enclosure sealed:

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS265sealed.jpg

Certainly do-able for a smallish sealed enclosure, for reference the enclosure tuned at 5Hz looks exactly the same remember the closer you tune an enclosure to 0Hz (sealed) the more it will represent the sealed variety,

Now, lets play with an “SPL” enclosure, we will place the “bump: at 50 cycles, right around the -3 point of the sealed enclosure, this resultant bump makes the enclosure far from ”sq” of flat but it will play very loud for the given input power at 56Hz, if you had an SPL geared rig you would want to “Burp” it in this area. It is 6dB above 0dB at this point which is the equiv of quadrupling your power above flat, it’s 9dB Above sealed! That’s a HUGE rise in efficiency, but at a very limited bandwidth.

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS265sealedvs50Hz.jpg

It’s common for premade vented enclosures to be tuned around 40 cycles, lets see what that looks like:

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS265sealedvs40Hz.jpg

Ooh not pretty, THIS is why we hear about a vented enclosure being “one notey” or “boomy.” Slap this quality driver in a premade box tuned to the general realization of 40 cycles and you have made a bad sounding sub.

Now, as you can see, as we lower the tuning frequency the peak lowers in amplitude, we sacrifice this narrow banded efficiency for less efficiency but lower extension, we are getting there. Lets lower the tuning frequency again to 29 cycles ;)

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS265sealedvs29Hz.jpg

What have we here? A flat response graph, much lower -3 point than the sealed alignment and a great efficiency gain from 22 cycles to well over 100 cycles, the roll-off is predictable and there is no hump, do you see how a vented enclosure can be used in a sound quality application now and not be “genre specific?”

Now, is it safe to assume that ALL .7CuFt boxes tuned to 29 cycles will do the same thing? NO! Vented enclosures are way more sensitive to driver parameters than sealed alignments, lets take the sealed plot out and model up the same box for 2 different drivers, say a JL 10W3 (common) and we will leave the Dayton pink and the JL blue….

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS26510W3vs29Hz.jpg

So, In conclusion, you can’t toss any-old driver in any-old vented enclosure and expect predictable results, you really have to plan a vented alignment AROUND the woofer used, and above all the tuning of an enclosure is simply acoustic assistance at a certain frequency at a certain bandwidth, if the volume of the enclosure and the tuning of the enclosure are chosen correctly there is absolutely no reason that a peak-free response cannot be achieved with extension way lower than that of a sealed design!

One more graph, here’s the Dayton; sealed, tuned at 50, tuned at 40, Tuned at 30, and tuned at 20, all in a .7 Cu Ft box.

http://i6.photobucket.com/albums/y231/cwahls/Tech%20stuff/EPhatch%20Tech/RS265differenttunings.jpg

I recommend downloading WinISD Beta, http://www.linearteam.dk/default.aspx?pageid=winisd this was used to calculate all these graphs and it’s easy to use, they have a Pro version for free too but it’s tougher to use, I encourage you to download and play with it, make your sub-system sing!


Chad