20000 watts per channel


New member
good day sir vedmitra, talking of using the toroid transformer.
i want to know, what is the quick process to make a toroid..
please help me sir because i don't have a exact formula and complete computation.
thanks in advance.


RAEL. Find a mentor whom You can "look over shoulder" as i guess it will be almost impossible to say here or in a mail thread. It is NOT impossible to learn, but nor easy if You do not have easy access to the mentor when qestion raises....... And they WILL :).


Hello Mr. Rael and Mr. Malmir,
The way I do my Torrids is a simple formula , Turns per volt for Torrids are 0.64 turn per volt in comparison to 1 turn per volt of same core area of E I traffo.
Hope this will help.


New member
friends here is a class AB design I made with one of my learned friends, more than 20000 watts per channel,load can be down to 0.25 ohm(in a way can short output wires)distortion from 20hz to 20khz only0.09% on all loads, output voltage more than 70 volts.what else is needed

Hallo sir,..

what software that you aplly for simulating this circuit (20000W PA)..? can i download it..? may you give me the link for download.
thank you. i'm sorry if my english spoke is bad. :eek:

from East Java Indonesia


HelloMr. COPROT,
I designed the circuit with multisim 12, here is the link for it's newest version multisim 13 ,---http://kickass.to/ni-multisim-ultiboard-circuit-design-suite-v13-0-t8168549.html.


new ammendments

20,000 watts circuit now more modified


  • 75 volt output at 1 ohm load 20khz.jpg
    75 volt output at 1 ohm load 20khz.jpg
    427.6 KB · Views: 40
  • 75 volt output at 1 ohm load 1khz.jpg
    75 volt output at 1 ohm load 1khz.jpg
    405.7 KB · Views: 33
  • 75 volt output at 0.5 ohm load.jpg
    75 volt output at 0.5 ohm load.jpg
    422.1 KB · Views: 45
  • 20 hz.jpg
    20 hz.jpg
    368.7 KB · Views: 63
  • New Picture.jpg
    New Picture.jpg
    179.7 KB · Views: 115


New member
guys lets come to the real world...
here you are talking about 20kw class ab amp..
class ab has an efeciency about 50-60%....
so if you want real 20kw at the outpout you have to feed it with about 40kw....
this is a huge power...
where will the amplifier disipate 20kw of heat????????
you need a huuugeeeee heatsink with a lot of fans.....
and something else.....
does this circuit have any kind of protection????
and i don't mean sort circuit protection and speakers delay....
i mean vi-limiter....
you need a very very well designed vi limiter.....
have you ever seen any profesional amplifier wich can outpout about 20kw?????
i don't....
lab groupen has a 14kw amp....
but it's class d...(90-95%efeciency) and not class ab...


Mr. Panos and other friends,
Maybe you have not bothered to see these photos, this is 5.6 kw version on 1 ohm load,not 20 kw,A real practical circuit, 20 kw. version in ab class can also be made but I also know that won't be a practical amplifier, As you have mentioned lots of power supply , which won't be possible with single phase, and tons of output transistors with lots of heat sinks and fans to cool.4 nos. of these 5.6 kw. can be used in practical to give more than 20 kw., about H class and D class i know they are more efficiant , but I am a class AB lover , so I will go only with class ab.To see my works please spare some time and visit ---https://www.facebook.com/vedmitraa


Lets go to the real world Panos, i did 2 with 78% efficientcy or more, so it is surely possible. No no no You do NOT need 40kW to do 20kW in old timers where one did not have almost any other possibilities to reach rail to rail Yes one had to use a 40kW ++ power Supply. But since only one rail at the time is at full (or the anp is defekt). 20kW is enought. Heat is to be removed by some not large but efficient fans. And Yes there will be a huge demand of safety systems on this amp. There are a lot of possibilities here a V-limiter is one thing but not enought, the I-limiter is easy to implement.

Ever seen a amp with 20kW ? No but 16 :). And doing 2 * 5600W bridged could get realy close to 20kW.
Yes it will be a little easier with switched amps, but the beauty disappears a bit :). " have been shown somewhere in the forums here, but looking at connectors my estimate is that something will burn when used fully.

The only realy sad thing about this huge amp is the possibility of explotions due to bad cabling and bad speakers. Distributing 20kW is NOT an easy hobby task :). And remember the fuse of an amp can be changed, the fuse of a person cannot.


New member
can you tell me how you will manage to get 20 kw of sound from an 20kw power supply?
the amp should have 100%efeciency....
the 16kw amp you are talking about is ab class ?
78% efeciency can be obtained with class h amp ..not with class ab...
and what speakers will you power with 20kw???
there is a big diferecne between mulitsim and real world.....


1: Let the supply be partly controlled by output of amp. Not directly but like if the amp output was connected to "zero" of supply. Read i did NOT say GND for avoid confusion.
2: Claiming 100% needed is as insaint as vishing for winning in lotto each draw 4 month in a row. But yes in a perfect world yes.
3: One of the was AB the other switched.
4: Then tell Audio Precision their test stations meassure wrong :).
5: Ahm hm Stop. I did not say i would do this, actually almost opersite. It will be very very dangerous to supply speakers from only one 20kW amp. The voltage and current are well above what most of us finds funny or healthy. Personally I would stop at 2 - 4 kW and then have a couple of them, it will be much much more safe to do. And having 4 *4kW is still nice :).
6: Only when you use bad models. I have seen a company, who did some test on a PC server power supply, 3 phase 2kW. And it was only the rush in current they could not get right in simulation. Taken the picture of a Agilient scope and (OK not MultiSim) pSpice, pSpice showed a slitely higher ripple compared to real world. So in short, it depends on your models... Yes they payed a fantazilliation for some of the models. And they actually do high-end switched amps dayly in simulation without any difference to the real world.


New member
hello vedmitra may i ask u a question regarding on ur photos of u toroid core .that is a perfect cirlce ryt? but how about when i used a not perfect circle core like the core of refregirator machine can i used it even in not perfect cirlce? im new talking about toroids transformer.


New member
Hello...its better than qsc crown and other amp manufacturers i am a fan of yours your speakers look so good what drivers are you using. For such high power wattage amps any schematic of amps you use for 25000 people gathering i am interested.