diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Need to get the highest voltage swing, which output topology? (http://www.diyaudio.com/forums/solid-state/52593-need-get-highest-voltage-swing-output-topology.html)

EWorkshop1708 1st March 2005 05:02 PM

Need to get the highest voltage swing, which output topology?
 
I'm designing a small PA system for under the hood of my truck. It doesn't need to be the best sounding, but it needs to be loud and can drive a 2 ohm load. It will be completely hidden from view, and I'll have a microphone for inside my truck if I need to yell at someone with it, or just whatever, :lol:

I'm wanting to keep it simple and use the 12-14V from the battery directly. I'm going to do a BTL amp. I'm going to use some Biplolar 80W transistors for the outputs. (since I lost my bag of 12V car radio IC chips during moving :mad: :mad: :mad: )

I was thinking that a Darlington is going to reduce voltage swing because of the multiple base-emitter voltage drops.

I imagine a CFP output may have a fairly better swing, but I think it still may not be the best solution.

I was also thinking of just a 2 transistor output with no drivers, but just a strong VAS stage. I'm just not sure how good something like that would drive 2 ohms BTL, but I was thinking it would be easy wtih only 12V or so.

The point is that I want to get the swing as close to the rails as possible.

I'm also curious how the car audio IC's get a high swing, but I'm also searching around.:D

Thanks

Mr Evil 1st March 2005 05:35 PM

CFP with a little bit of gain (maybe 1.1-1.5, depending on the available swing from the VAS). That can get you to within 1V of the rails.

Not having a driver stage for the output is a bad idea, especially for driving a low impedance load, since even if the VAS can supply the large currents necessary, the heavy load will cause lots of distortion.

K-amps 1st March 2005 05:46 PM

Noobie Question:

Is there an easy way to convert a emitter follower into a CFP output stage. i.e. Flip the PNP/NPN op devices being driven or do the drivers need to be modified too?

How is the gain of the CFP stage set? Degeneration resistors?

Sorry for strying from the thread ....

EWorkshop1708 1st March 2005 05:55 PM

You make a good point, but I'm not sure how I'd achieve the voltage gain you talk about for CFP.

Instead of connecting the emitters of the driver to the collector or output, would it make more sense to use some kind of resistor network to ground?


Also, I've taken a look at an IC chip, the TDA7560. It has 4 BTL channels that can drive 2 or 4 ohms each :eek: depending on heatsinking. I imagine it may have a better SOA than a few 80W discrete transistors, but I just imagine such a chip is expensive, especially compared to my collection of free transistors. I'll keep looking around.

The chip seems easy, but building it myself discrte would seem to be more enjoyable, but after looking at the specs of that IC, I'm blown away.

Decisions, decisions :D

richie00boy 1st March 2005 06:16 PM

To limit the gain of the CFP you simply add a feedback network between output and where the drivers emitters connect. Like this:

http://www.richie00boy.pwp.blueyonde...c/cfp_gain.png

You should make up two identical amps and bridge them to get the highest power with the available voltage, but watch it as each amp will see a 1 ohm load with a 2 ohm speaker.

Personally, I would just buy a cheap or used car amp. It will cost you less and probably work better.

K-amps 2nd March 2005 01:08 AM

You mean the larger the value of R35 the more the gain?

EWorkshop1708 2nd March 2005 05:46 AM

I think the two 47 ohm resistors R29 and R30 determine the gain.

I imagine using uneven ratios could yield some more gain, but for this project, I'd probably keep the two resistors the same value. I wonder if 47 ohm would be ok, or if I'd need to go lower since it would drive such a low impedance. Maybe 22 ohm 1 watt resistors? The transistors I have should have enough gain to not need too low of a value, but I'll figure it out.

Notice how one of the feedback resistors goes to ground. In a 12V single power environment, a Capacitor would be needed, and I think 1000uf cap would be enough.

Eva 2nd March 2005 06:56 AM

Use a TDA7560 IC from SGS-Thomson. It features 4 MOSFET BTL rail-to-tail amplifier channels in a single power package, it drives 2 ohms provided enough cooling and here it costs approx. 15 euro [not much more than $20]

Discrete is not an option for 12V operation, common amplifier topogolies don't work with such a low single rail voltage [including the suggested schematics]

In the other hand, if you really want to build your own discrete amplifier be prepared to build also a SMPS to get decent +-30V rails or higher


All times are GMT. The time now is 12:45 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2