diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Lateral MOSFET, How do I calculate power output (http://www.diyaudio.com/forums/solid-state/113718-lateral-mosfet-how-do-i-calculate-power-output.html)

synonymous 16th December 2007 07:55 AM

Lateral MOSFET, How do I calculate power output
 
Hello's

Wondering if I can gain some insight on either rules of thumb calculations or if there are some pretty straight forward ways of calculating the power output of Lateral MOSFET based amplifiers. I realize that there are different losses at different load impedances. Typically it is assumed that the power output can be doubled by simply halving the load impedance. This is untrue however, no?

I guess my question is..
Based on transformer secondary voltage, what kind of power should I expect to see in the real world at either 8,4 or 2 Ohm?

Are these powers affected by more or less paralleled output devices?
Typically, is efficiency lost with more output parallels?

Appreciate any clarification
Scott

AndrewT 16th December 2007 09:32 AM

Hi,
start with the power capability of the output stage devices.
1pair of 100W FET devices can produce 60W into a severe load, or 100W into a moderate load. BJTs tend to give a little less (~80%) due to de-rating for secondary breakdown.

Let's use 2pair of 100W L FETs. and set the target at 100W into a severe load. That's 40Vpk and 5Apk into an 8r0 resistive load and upto 14Apk into a highly reactive 8ohm load.

To get 40Vpk into your load you will need supply rails that hold up above 45Vdc when heavily loaded. You can start with +-50Vdc and this may just give 100W into 8r0 but could be just short of target power.

A 35+35Vac 160VA (mono) or 300VA (stereo) transformer will achieve close to target.

The severe load capability (a real speaker) is a worse case than driving half the load resistance you are designing for. This implies that your amp should be able to deliver at least 180% of it's maximum power into a 4r0 load, but not continuously, since the heatsink will heat up quite rapidly. This 180% power into 4ohms cannot be achieved with the transformers mentioned earlier. I find that putting in a larger transformer than required for the design helps with extending the bass response. I would probably fit a 250VA 35+35Vac for a 100W into 8ohms knowing that this should be able to give about 180W into 4r0 on test. But keep in mind this is a design for driving a severe 8ohm load, not a 4ohm amplifier.

unclejed613 17th December 2007 11:43 PM

also, you have a higher bias voltage for MOSFETS, which means your drivers have to swing another 3-10V for full output close to rail-to-rail. this means you actually need a higher rail voltage for at least the VAS and driver stage, so for 100W output, you will want at least +/- 70v for the VAS/drivers. to simplify things, you can make this your output rail voltage as well, but your transistors need to be 150V devices.

suzyj 17th December 2007 11:55 PM

I go through a bit of a calculation on the webpage describing my lateral MOSFET amp, at http://www.littlefishbicycles.com/poweramp/index.html.

I derived a simple formula to give Rth(max) for the heatsink, but you can easily solve for power, given Rth etc.

To get the dissipated power, I found the easiest way was just to use LTspice. I designed the amp with LTspice, and simply measured the power dissipation in the output transistors.

synonymous 18th December 2007 06:54 AM

Very nice! Thank You

The designs I use are based on Randy Slone's work. I will be using the same Hitachi output devices.

I see you are in the process of making an anti-thump circuit.
Have you considered a circuit that controls the mains power based on an input signals presence?

THX again


All times are GMT. The time now is 01:06 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