posted
Q "o" da day : A line driver is what? Nothing more then a pre amp, right? And a preamp is incredibly easy and cheap to build (what..? A couple of Caps, a couple resisters, and an IC, right?). Sooo...Could I build a line driver and actually make it usable? I should be able to get a fixed output voltage fairly easily (right? hehe), but is there any thing else I should know? (like is this more complicated then what I think it is? ). Thanks!
posted
A line driver is a pretty simple device to do. But you need good circuit design to make sure you aren't adding more noise than you are going to gid rid of. In college I did a quick experiment. Ran the output from a CD player (<1V rms) through a pretty noisy enviroment to an IC based amplifier and measured the amount of induced noise. Then I added a very simple line driver that I built with the knowledge I had at the time. The amount of gain I added was roughly 6dB, the amount of induced noise was much less than before but I had more noise due to the line driver so overall it didn't help too much. Most HUs now have a pretty high signal voltage, and you don't gain any amp power by adding a line driver so in most cases now they aren't necessary. Plus you have the problem of getting the +/- 15V supply for any circuit you build for a car which in many cases will add mucho noise. I don't want to stop you from experimenting, you will learn more from doing projects like this than you will in your classes. I would try to use this to learn about circuit design, component choices, and tolerances and how they affect distortion and noise because you don't learn that in school. I would go ahead and build it as a balanced line transmitter while you are doing it.
posted
you are correct, a line driver is a few caps, few resistors, an ic(op amp), a dc/dc converter, remote turn on circuit, fuse, gain control, and a few other odds and ends.. but yes you can build it yourself, if you find the right power supply (for something you wanna do I'd look for a 12v to +/- 15v dc/dc converter a quality audio op amp (burr brown.. I love those things) that'll get you started, if you have some knowledge of op amp circuit design you may just want a non inverting gain stage with 10db+ of gain (all depends on what your deck puts out, and your amps like at the inputs) you can get away with building it on a "perf" board but I found noise/reliablity issues doing that, so ideally you'd wanna etch your own circuit board (very easy once you've done it a few times and have the right equiptment) if you need a little more help let me know. Robin robin@rwaudio.com
------------------ Digital Designs 9515b Memphis 1000D Pioneer source MB Quart fronts ADS power plate (soon Adcom 4404) RW Audio processors LD4.12MBR Member of Team Hertz Canada
Posts: 735 | From: Calgary AB | Registered: Oct 1999
| IP: Logged |
posted
You're only going to need a couple op-amps so you might as well use some good ones like BB. The only other thing I can think of is to have a very good power supply that is tightly regulated and stable.
posted
Do you suggest putting some resistors on the output as when you get he power supply stuff ti sometimes reccomends this, although it does make nonsense the efficiency. About Rload/10 Colin
posted
I wouldn't care too much about efficiency on a 1 watt device. Adding resistors to the outputs of the power supply depends on the supply. I once used a switching supply that would get strange oscillations if there wasn't a 100mA load, so there I needed some resistors. Most of the things I do for car audio use PWM control that are tightly regulated, so they can do without.
posted
I thought the idea behind a line driver was to split each channel into a pair of signals, Invert one so its 180 degress out of phase, send the signals to the place you want them to get. Invert the other side that wasnt inverted, then add the two. the signals will add, the noise will be cancled. The idea is that the noise on both lines is the same, so if you take two lines with identical noise signals, invert one, then add it to the other the noise signal will be cancelled out. Is this what a line driver does?
------------------ 2-DD 9512's 2-2000X's (for now) 8.0ft^3 box in a civic trunk One big ass port ?????dB's
It's not the size of your woofer, It's the size of the waves your woofer makes!
Posts: 1259 | From: Fullerton. CA ,USA | Registered: Oct 1999
| IP: Logged |
posted
Interesting isea, i could add it on as a module at a later date, not difficult to design that, 3 op amps, couple of resistors, should work quite well. Colin
posted
If anyone whould like to go into some more detail about PWM devices, I'd love to hear it. In all honestly, besides knowing the acronym, I have no idea what a pulse width modulated device is.
posted
PWM - pulse width modulation It can be used for any system that varies the pulse width of a square wave. In a class D amp the pulse train is sent through a low pass filter to reconstruct an audio signal. In PWM power supplies a control IC varies the pulse width of the switching FETs. In a regulated supply the output voltage is fed back to the IC, if the voltage is low the pulse width is increased, if it is too high the pulse width is decreased. What else do you want to know?
posted
To add to what Jc2 said incase your not an engineer, If you have a square wave that goes from 0v to say +10v, and half the time its at 0V, the other half it is at ten, this is called a 50% duty cycle. This square wave will have a 5V-rms value. Now change it to 75% duty cycle its value is now 10V 75% of the period and 0v the other 25%. It’s now equal to 7.5V rms. So take a square wave at say 100khz. Give it a peak voltage of like 100V. By changing the duty cycle you can make it anywhere from 0V to 100v rms. Now think of music. It is just a continuously changing voltage. Now change the pulse width (duty cycle) so the voltage is an multiple of the music source rms voltage and you have an amplifier. One more thing needs to be done. You have super high frequencies you don’t want to send to your speakers, put a lowpass filter set at 20khz. Your speakers will only see a changing rms voltage and not the squarewaves.
------------------ 2-DD 9512's 2-2000X's (for now) 8.0ft^3 box in a civic trunk One big ass port ?????dB's
It's not the size of your woofer, It's the size of the waves your woofer makes!
Posts: 1259 | From: Fullerton. CA ,USA | Registered: Oct 1999
| IP: Logged |
posted
ouch!!!..my brain hurts!! very interesting topic...and i kinda understand it..I wanna thank you guys for these tech. tidbits so us "electonically impaired" people know somthn' about what we own..
Ryan
------------------ 4 thump 12" + 4 earthquake D2's = 1 big fire!(and 195.3 db but venus exploded when I hit it)
posted
A PWM regulated switching power supply is special because of it's efficiency and it's ability to change one DC voltage to another. PWM is only used to regulate the output voltage. A simple switching power supply works on these principles: A square wave is created; then amplified; this new higher power square wave is then sent through a transformer to step the voltage up or down (or by using multiple taps, to many different voltages). All of the outputs of the transformer are then rectified to DC voltages. The high efficiency comes from the fact that when you amplify a square wave (or any other duty cycle pulse train), the transistors are either turned all the way on (no resistance from collector/drain to emitter/source) or all the way off (infinite resistence from c/d to e/s). This means that there is always either no current flowing through transistor or no voltage dropped across the transistor and thus never any power dissipated in the transistor. One way to regulate the output to a constant DC voltage (without using power consuming linear regulators eg. 78** series)is to vary the duty cycle of the "square wave". When the load on the power supply becomes to heavy, the output voltage drops, the PWM circuitry will then increase the duty cycle to cause the output voltage to rise until it reaches the voltage that it's supposed to be. This regulation happens very quickly and cannot be seen at the output.
------------------
Posts: 76 | From: Quispamsis, New Brunswick, Canada | Registered: Oct 1999
| IP: Logged |
Printer-friendly view of this topic