1000W Simple PA Amplifier

protection

hi apex sir,
can i add this pro to ba1200?
 

Attachments

  • Snap403.jpg
    Snap403.jpg
    61.4 KB · Views: 1,991
the peak output voltage will always be less than the quiescent supply rail voltage.
For an 8ohms load and a good PSU expect the maximum peak output to be 4V to 10V below the Vrail.
Max power is (78-10)²/Load/2 = 289W into 8r0 to (78-4)²/Load/2 = 342W into 8r0

With a poor PSU expect the Vdrop to >15V and for a bad PSU maybe approaching 30V, when trying to drive 2r0.

The voltage drop for 4ohms will be higher, and higher still for 2ohms.

Am I correct?
No.
 
Last edited:
Thanks for the answer AndrewT.

So based on the formula (4 volts drop) you sent me 4 ohms works out to be 685W.
So now how many pairs of output devices should I use? I approximate 100w per pair, so 6 pairs should be fine, correct?

In real life (mid-fi audio like Adcom/Forte etc) you will be lucky to get over 240 watts into 8 ohms and 400 into 4. This assumes a 700va transformer.

Bump it up to 1kva and you can go over 300 watts into 8 ohms. These are real life numbers what you may end up getting as opposed to ideal world calculations. figure 2.5x rated transistors for given wattage you need. so if 800 watt at 2 ohm, built an output stage worth 2000 watts of trannies.
 
figure 2.5x rated transistors for given wattage you need. so if 800 watt at 2 ohm, built an output stage worth 2000 watts of trannies.
I have arrived at 5 to 6times for BJTs and 4 times for FETs.
Cordell has a similar multiplier, but arrives at it using a different method of analysis.

Big amplifiers generally don't use ClassAB and have gone to ClassG, or H, or D to save expense on output devices.
 
if you want 100W amplifier with a BJT output stage then expect the output devices to have around Pmax total ~ 500W to 600W.
If mosFETs then expect around 400W of total Pmax.

The load impedance does not matter.

But there is a secondary effect, due to secondary breakdown.
As power goes up and/or impedance goes up the Vce goes up, a lot.
The increase of Vce brings in the temperature de-rated SOAR, very much sooner, when driving a reactive load.

A 100W into 4ohms amplifier will have a very low Vce and thus very little of the secondary breakdown comes in to limit the peak output current.
Here just 400W of output devices will probably suffice even for an amplifier used for drunken partying duty (probably worse than PA duty).

But 200W into 8ohms will have a very much higher Vce. Thus 6 times may be a more appropriate factor. This would require 1200W of total Pmax.

Compare the terrible 1943/5200 SOA to mj15003/4 SOA (slow), or to MJL4281/4302 SOA (fast).
You will probably need 5pair or 6pair of 1943/5200 to match the reliability of 3pairMJL4281/4302 or 2pair of mj15003/4 in a +-70Vdc amplifier.
 
Last edited:
if the amplifier is rated for 2ohms speakers and it's rated output into that reactive load is 480W then the guidance for the power capability of the output stage would be ~4times for mosFET or 5to6times for BJT.
I.e. 480W into 2ohms would require a ClassAB BJT amplifier to have approximately 2400W to 2880W of total Pmax.

ClassG, H, D would have very much less.
 
if the amplifier is rated for 2ohms speakers and it's rated output into that reactive load is 480W then the guidance for the power capability of the output stage would be ~4times for mosFET or 5to6times for BJT.
I.e. 480W into 2ohms would require a ClassAB BJT amplifier to have approximately 2400W to 2880W of total Pmax.

ClassG, H, D would have very much less.

2400W Pmax=1200W Prms.
so efficiency is (480/1200)*100=40%....hey it is very bad class AB amp..:rolleyes:.