Subamp with ~ 1000 W RMS at 4 Ohms, built at low cost

[FQR]

Hello guys.

...at first I want to apologize for my bad english. I have no practice at all, and only remember a little of school education in this language. Please don´t think too bad of me, therefore! ^^ ...at second I have to say, i´m practiced in working with (simple constructed) high voltage projects, and know about the hazards, and what to do for security. But --- in audio the only project i did, was a LM 3886 amplifier (as Siegfried Linkwitz showed on his site).

Now I have a problem: I have nearly zero money, but want to build an amp with an output of about 800 - 1200 (best would eventually be 1000) W RMS. From a former project i have some probably usable parts in stock. This is the simplest usable scheme i could find. What do you think about it? Is it good for a subamp? I got (each 24 pcs) IRFPC50A(from IRF) and IRF450(but from Harris, less I.ds and also P.tot). The IRFPC50A seems to have nearly the same parameters like IRFP448, but, unfortunately, only nearly...

Could I use one of this FETs? And when, what do you think, how many pairs should i use for 1000 W RMS at 4 Ohms? And can I still use the showed gate driving BJTs (Q10 and Q11)? Or stronger ones, and when, then which? (Only thing I really know, is, that I´d have to use bigger fuses...) I got two transformers, rated at 2,5 kVA (permanent), 13,3 kVA (short time), with center tap, and by changing primary taps, i can get (+/-) 70.96, or 74.35, 78.05, 81.32 (8 steps of ~ 3.5 V) 102.72 VDC (without diode losses). These are a little too big, but I have no other. (Yes, I know, I only need one... ^^)

Is this schematic acceptable? (No real foults, I mean.) The sound must not be "audiophile", it only should work, without bigger complications. I know, normally in this modern time, if one wants to get so much power, he should do with class D, but this (now) is too complex for me as a more electric (and not electronic) experienced guy. And cannot make PCB, so I´d be glad if I must build the driver stage only once, cause I must use "breadboard" (I think this is the name of this in english), and combine with point-to-point (better silvered copperwire or is varnished enough?), where necessary.

I have one further schematic (which HAS already the possibility to ad enough FETs [the gate-driving BJTs, who drive 2 x 5 IRFP 460, surely would drive a bigger number of smaller FETs, e.g. 2 x 10 of my IRFP 450 [Harris, not to forget, otherwise, if they were "real" IRFP450s from IRF, 2 x 8 would be roughly enough, i think] , cause it is mentioned in the text, that the driving capability is not exhausted) but this needs one voltage 10 volts higher than rail, and i don´t have a transformer with those further tap. (Like I wrote, my voltage changes by changing PRIMARY...) I don´t know, if it would be a good idea, to ad some windings, but... I think, it would not. (Isolation etc.) But, the good thing with the second circuit is, it has a complete description of getting the amp started. And has a second pot, to do this... (Begins with: "Put some 100 Ohm resistors in place of the fuses..." --- really good described, a fine thing.)

I don´t even know what kinds of resistors (evtl. metal film ok?), capacitors (film types like MKS ok? --- except the unipolar electrolytic, of course), and what kind of inductor should be good? (From the inductor in scheme 1 there´s no information at all, and cause I´m not experienced at all, I have no idea. Should it be ferrite or plated iron? Have or have not an air gap, and when, how wide? Only thing I am sure, copper wire should be thick enough.)

Please, help me! Could I make it work?

(P.S.: In a thread I found, one guy said, it would be cheaper, to buy a used PA amp, than build an amp of this power. I hope, he meant the costs of the transformer (my ones I got in former times, when I had money, for 200 Euro each) and the Power-FETs, cause used PA-amps of this power never come under 100 or more Euros --- I simply don´t have, and wouldn´t get, such an amount of money, for now)

First possibility, which fits my transformer:

400 W MOSFET Audio Amplifier Using IRFP448 : Tasel

Second possibility, which needs 10 "extra" Volts:

600 Watt Quasi Amplifier With Mosfet IRFP460

...or do you have a better one?

Greets,

FQR

 

[JMFahey]

Hello guys.

...at first I want to apologize for my bad english.

Don't worry about that

I have nearly zero money, but want to build an amp with an output of about 800 - 1200 (best would eventually be 1000) W RMS.

I got (each 24 pcs) IRFPC50A(from IRF) and IRF450(but from Harris, less I.ds and also P.tot). The IRFPC50A seems to have nearly the same parameters like IRFP448, but, unfortunately, only nearly...

Well, that's something.

Could I use one of this FETs? And when, what do you think, how many pairs should i use for 1000 W RMS at 4 Ohms?

You will *not* get 1000W into 4 ohms but what your PSU rails allow you to.
The first schematis will not work, it's a mess.
The second one looks somewhat better, but the Internet is full of poorly "designed" , untested, noob schematics.
Go for a tried and true schematic such as those published here by ApexAudio.

I got two transformers, rated at 2,5 kVA (permanent), 13,3 kVA (short time), with center tap, and by changing primary taps, i can get (+/-) 70.96, or 74.35, 78.05, 81.32 (8 steps of ~ 3.5 V) 102.72 VDC (without diode losses). These are a little too big, but I have no other. (Yes, I know, I only need one... ^^)

The safest option in primary switched transformers is to use them with the "lowest secondary voltage" , which actually means you are using the full primary (GOOD !!)
Si you have beefy +/-70V rails. Cool.
You have what you have, which is good for 2 350W power amps or so, fed from a single transformer.

Is this schematic acceptable? (No real foults, I mean.) The sound must not be "audiophile", it only should work, without bigger complications. I know, normally in this modern time, if one wants to get so much power, he should do with class D, but this (now) is too complex for me as a more electric (and not electronic) experienced guy.

Go for the ApexAudio designs.

And cannot make PCB, so I´d be glad if I must build the driver stage only once, cause I must use "breadboard" (I think this is the name of this in english), and combine with point-to-point (better silvered copperwire or is varnished enough?), where necessary.

Forget it.
If you can't make (or have it made) the PCB, you can't make "big" power amps.
You'll also need beefy PSU capacitors, big heatsinks, a strong chassis for it all (those must be some *heavy* transformers), etc.

I have one further schematic (which HAS already the possibility to ad enough FETs [the gate-driving BJTs, who drive 2 x 5 IRFP 460, surely would drive a bigger number of smaller FETs, e.g. 2 x 10 of my IRFP 450 [Harris, not to forget, otherwise, if they were "real" IRFP450s from IRF, 2 x 8 would be roughly enough, i think] , cause it is mentioned in the text, that the driving capability is not exhausted) but this needs one voltage 10 volts higher than rail, and i don´t have a transformer with those further tap. (Like I wrote, my voltage changes by changing PRIMARY...) I don´t know, if it would be a good idea, to ad some windings, but... I think, it would not. (Isolation etc.) But, the good thing with the second circuit is, it has a complete description of getting the amp started. And has a second pot, to do this... (Begins with: "Put some 100 Ohm resistors in place of the fuses..." --- really good described, a fine thing.)

Go apexaudio , tried and true.
Plus there are long threads discussing them here (which you should read thoroughly before enbarking in anything) and you can also clear your doubts with other builders.

Please, help me! Could I make it work?

You probably can, but take it easy, go for something tried and built by many.

100 or more Euros --- I simply don´t have, and wouldn´t get, such an amount of money, for now)

Don't think you can build it, you have some expensive parts but will certainly spend more than 100 € on the rest.

First possibility, which fits my transformer:

400 W MOSFET Audio Amplifier Using IRFP448 : Tasel

Second possibility, which needs 10 "extra" Volts:

600 Watt Quasi Amplifier With Mosfet IRFP460

Forget them.

...or do you have a better one?

Already suggested
Good luck.

 

[FQR]

Hello, again!

Don't worry about that

...thank you! But I got problems with...

I have nearly zero money

Unfportunately that´s a fact.

The safest option in primary switched transformers is to use them with the "lowest secondary voltage" , which actually means you are using the full primary (GOOD !!)
Si you have beefy +/-70V rails. Cool.
You have what you have, which is good for 2 350W power amps or so, fed from a single transformer.

I believe you, but what exactly do you mean about this? Only the lower voltage, which is therefore safer, or do you think of better regulation? I don´t know, but want to. Naturally I want to know nearly *anything*... ^^ Two 350 W amps could be enough, if i feed them phase-reversed to bridge them. And with good vent cooling on big heatsinks, enough pairs, and the low rail voltage (in relation to the Vds of the FETs) it should be possible to drive "pseudo"-2 Ohms per channel (which such a config should bring from one 4 Ohm load). But... this thoughts are senseless, it seems, because ... :

You will *not* get 1000W into 4 ohms but what your PSU rails allow you to. The first schematis will not work, it's a mess. The second one looks somewhat better, but the Internet is full of poorly "designed" , untested, noob schematics. Go for a tried and true schematic such as those published here by ApexAudio.

So i want to, and I believe you, that those two designs are not good. That was it, why I decided to ask you guys in this (real famous) forum. But ... I have only N-Channel-FETs.

I not made any powerfull amp with only n-channel fets.
Regards
__________________
http://www.facebook.com/APEXaudio.MileSlakovic

This is, what Mile Slakovic says.

My problem is, that I´m trotting around with only 0 to 150 bits per second (with many pauses. I´m in a really bad radio connection hole, accessing the www with a 3G UMTS Stick, without antenna, and I MEAN bits, not kilobits... ).

So it will take time, but (for now) I only can search for solutions, that fit my given parts, at least as much as possible. Its a pity, I hoped to find a possibility to improve faults on the better one of the designs, if I could find nothing, that fits in. But probably that´s impossible, cause everyone is busy, has his own project, and nobody who is able to do this, is interested in such imperfect and fragmentary work.

Ok, for now, many thanks to you, dear JMFahey!!!