IRAUDAMP boosted to 1KW in 8 Ohms

[ajazz]

Hi Buddies,

I am absolutely a new member in our forum, so please excuse me of any stupid ignorance of our rules.

I have made a duplicate of Iraudamp with slight changes and boosted the output to almost 1KW in 8 Ohms. (for the interested I shall post the schematics & pics 2mrw).

It works directly on AC Mains of 230V converted directly to DC 320V charging 10,000uf / 200V Caps in series (centre tap) with soft start ofcourse and using IXYS Mosfets of 600V-48A.

It works like a Bomb taking the **** out of any speaker in the world. I used Eminence pro 1KW speakers but it couldn,t bear the abuse. (I limited the output modulation to 80%)

Now I have to discuss a small problem ?

It works fine right from 10Hz to 2Khz and then the frequency starts dropping to almost 10% output at 20Khz. I don't know if some part of the circuit is getting saturated on higher frequencies or something else. I tried tweaking around with output filters, feedback pole caps etc but in vain.

Can any of our Guru,s solve my problem.

regards

Ajazz Bukhari.

[Alme]

Hi Ajazz, i'm not guru but trying similar things to yours; what is your 'idling' frequency? Original IRaudamp's is near 400khz. Do you mean that frequency responce is poor? To have lower current ripple in inductor you have to keep L value high enough with 320V supply; but to have wider response you need L as low as possible. What are output LC-filter parameters?
IXYS datasheets confusing me, they don't say about built-in diode specs, which is important. In particular IXKH47N60C type which is similar to your description. Are you satisfied with their mosfets? I have successfully used Fairchild's FQA28N50F (here "F" index is important) in similar amp.
What is amplifier efficiency? How about trying IGBT switches?

[Alme]

Uh finally found mosfet types at IXYS with fast recovery diode specified. They are IXF**** series not IXK****.
Like IXFK44N60. Looking not bad.

[ajazz]

Hi Alme,

Nice to know your equal interest in the field.

The Mosfet that you have pointed out is perfect, but in my prototype I have used IXFK 48N50 (because of easy availability). It has got a Rdson of 100 mili ohms and a Trr of 170ns.
Perfect for our purpose. Since it has got a higher Qgd we have to limit the idle frequency to less than 200Khz to avoid excessive dissipation in the gate. Still it's a good enough resolution for higher frequencies. i.e. 20Khz X 10 time = 200Khz.
Other well know companies recommend frequencies from 100Khz to 175 Khz. (for full spectrum audio 20-20Khz). I will send you the links later. I even tried it at 100Khz with a good response and super cool Mosfets, but the size of the inductor gets larger.
So considering the efficiancy of the Mosfets and a smooth response I selected 175Khz as the best idle frequency. The output coil is Air core DIY type wound on 40mm Dia former and L=40mm with 80 turns of 1.5mmSq litz wire. This gave the best results compared to Ferrite or Iron cored toroids. The Cap is 0.47uf.

By the way my problem is solved as I myself had made a silly mistake on the inductor turns.

Bravo . Now my amp is perfect in all respects and it sounds really great compared to the best in the world. Though it is made with discrete components (not SMD) the size is only 60 X 60mm with all features and auxilary supplies on board.

If any body is interested in my schematics or PCB or Complete assembled board can contact me and share the experience.

Regards

Ajazz Bukhari ajazz@consultant.com

[Tekko]

No pics of this thing ? A amp running directly off the power grid without atleast an isolation transformer seems dangerous to me, is anything fail you send over 300vdc into your speakers.

[theAnonymous1]

How hard would it be to make a 120v version for us Americans? It would make a killer sub amp .

[Tekko]

A 120v --> 230v transformer maybe ? :P

[mzzj]

Originally posted by theAnonymous1
How hard would it be to make a 120v version for us Americans? It would make a killer sub amp .

Full-bridge is the way to go there But beware, this kind of like project can be easily real "killer amp"
Darwin awards, here we go.

[theAnonymous1]

Tekko, a trafo would work but one of the reasons I'm interested is the fact that the amp doesn't need one.

[mzzj]

Originally posted by ajazz
Hi Buddies,

I am absolutely a new member in our forum, so please excuse me of any stupid ignorance of our rules.

I have made a duplicate of Iraudamp with slight changes and boosted the output to almost 1KW in 8 Ohms. (for the interested I shall post the schematics & pics 2mrw).

It works directly on AC Mains of 230V converted directly to DC 320V charging 10,000uf / 200V Caps in series (centre tap) with soft start ofcourse and using IXYS Mosfets of 600V-48A.


Ajazz Bukhari.

Hi there!
What kind of like high-side gate drivers you are using? I was looking for some 600V gate drivers but they are either slow or low-current. Currently i have been playing around with gate-transformer drive but it seems to be limited to subwoofer apps.

(Crown BCA topology, with isolation transformer fed 340V full-bridge, 3kW@16ohms, more with 3-phase feed )