how much power?

[smokingmachine]

Hellow!!

Which tranzistors are better for output stage of amp?

I noticed that someone are using mosfets,others are using darlington and again others use bipolar tranzistors?

Could anyone tell me which type is better and why?



Thanks!!!!

[smokingmachine]

i would like to do an amp which has around 600rms at 2 ohm with supply around +/-56V!

Is it posible to get around 800 rms with this amp using a 4 ohm load(need to push up supply voltage to...................how mutch)?

amp has 2x6 darlington tranzistors!

i think this could be done but will the amp heat up too much?

in my oppinion the amp would heat up more when using 2hm load witth supply +/-56V(more current) than using a 4 ohm load at 800rms



Give some answer!!!!!

[pinkmouse]

Hi Smokingmachine. I have merged your two threads together as they clearly refer to the same project.

Have you had any experience in building amplifiers before?

[ocool_15]

P=V^2/R
56.6VDC is needed for each rail at the load but you still have to account for losses in the amp.

[Eva]

Bipolars are the easiest to use and the more linear. MOSFETs may allow to get the same power output from less devices when properly employed, though.

[myhrrhleine]

There is no 'better'
Each has good and bad points to them.
That's why different people use different devices.
It just depends on design philosophies and design skills.

[AndrewT]

Hi,
Ocool there is something wrong with your calculator. 56.6^2/2=1600Wpk =800W into 2 ohm, more than he wants.

49^2/2=1200Wpk =600W into 2ohm.

Taking losses into account then +-55Vdc when loaded may be a target to aim for. 2 times 40Vac to 42Vac 1kVA.

I would go for MJ21193/4. Neither darlingtons nor Lfets can match at 2 ohm loading.

I'm told that Vfets might be able to compete and probably cheaper if you go quasi Nchannel.

[Johan Potgieter]

Smokingmachine,

As said this is indeed a difficult question with no simple unique answer.

But certain topologies are better than others. In an analysis by Prof. Otala a long time ago, a double emitter follower was rated 10 times better than quasi-complementary, and full complimentary topology (the best) 20 times better. The use of "better" is of course very unspecific; it refers to relative distortion but also to other characteristics. The details are somewhat long to post here, and it also depends on the type of semiconductor, etc. as others suggested. (These tests were for transistors but would broadly be valid also for MOSFETs).

This as a very general indication. It would appear that MOSFETs may have the edge over transistors, but I have no experience of them. When I last heard there was a very large spread in characteristics, which could make matching difficult. I wonder if this has improved.

Regards.

[smokingmachine]

well the amp i am trying to do has mj11015(6x) and mj11016(6x)!!this are bipolars!!!

i wonder if this amp will do if I raise voltage to +/-80 or 85 volts and than it could deliver around 800 --- 900 rms at 4 ohm load or the amp will be heating too much(i will use a vent)

i think if amp could deliver 600 rms at 2 ohm load then the amp could deliver around 800 rms at 4ohm load(with proper supply)and still won`t be heating as much as if the amp would run at 2 ohm load with supply around +/- 56 V per rail!!!!
As I calculated the amp with +/-56V at 2 ohm load woult took much ampers than with +/-80........85 V(4 ohm load)!!

Am i right or no?


Ant thanks for all your answers!!!!!

[Workhorse]

Originally posted by AndrewT
I'm told that Vfets might be able to compete and probably cheaper if you go quasi Nchannel.

Hello Andrew,

Yeah Sure,

Quasi-N-channel VFETs always wins the race in 2-ohms loading.....

cheers,
K a n w a r