2 kwts discreet class d need help

[norazmi]

lol yah, its for testing and i`ve got neighbor around so better i do with free air or i`ll get some trouble from them. With enclosure of course too loud.

 

[Dimonis]

I meen that i'm pity for it.;ww:

 

[norazmi]

Oo... sory. New coming from me new build ucd discrete + totem pole + 4 pcs mosfet which 2 pairs. hehe . will build after back from vacation with my family maybe 3-4 days after.

 

[norazmi]

propose totem pole driver + double mosfet still half bridge

 

[norazmi]

Ltspice result with totem pole and 4 mosfets which means 2 pairs. Simulate with 4 pcs IRFP250N, rails +/- 75 vdc at 4 ohm load.

Source stepping succeeded in finding the operating point.

Fourier components of V(out)
DC component:0.067617

Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+02 2.896e+01 1.000e+00 -147.97° 0.00°
2 2.000e+02 1.346e-01 4.650e-03 -0.03° 147.94°
3 3.000e+02 8.594e-02 2.968e-03 85.52° 233.49°
4 4.000e+02 2.985e-02 1.031e-03 16.63° 164.60°
5 5.000e+02 1.167e-02 4.030e-04 25.33° 173.30°
6 6.000e+02 1.695e-02 5.854e-04 -3.47° 144.50°
7 7.000e+02 1.010e-02 3.487e-04 -15.14° 132.84°
8 8.000e+02 1.577e-02 5.446e-04 2.86° 150.84°
9 9.000e+02 1.215e-02 4.197e-04 -10.22° 137.75°
Total Harmonic Distortion: 0.570905%


Fourier components of V(in)
DC component:-3.7836e-010

Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+02 2.000e+00 1.000e+00 -0.00° 0.00°
2 2.000e+02 5.407e-10 2.704e-10 105.31° 105.31°
3 3.000e+02 3.796e-10 1.898e-10 22.87° 22.87°
4 4.000e+02 4.277e-10 2.139e-10 -36.38° -36.38°
5 5.000e+02 2.301e-10 1.150e-10 118.43° 118.43°
6 6.000e+02 2.174e-10 1.087e-10 -170.30° -170.30°
7 7.000e+02 5.184e-11 2.592e-11 -112.23° -112.23°
8 8.000e+02 1.538e-10 7.690e-11 24.30° 24.30°
9 9.000e+02 7.886e-11 3.943e-11 -60.41° -60.41°
Total Harmonic Distortion: 0.000000%



Date: Fri Mar 09 06:33:25 2012
Total elapsed time: 1030.363 seconds.

tnom = 27
temp = 27
method = modified trap
totiter = 8172650
traniter = 8169002
tranpoints = 2183417
accept = 1546704
rejected = 636713
matrix size = 122
fillins = 278
solver = Normal

 

[norazmi]

here is the wave form.

 

[irvynejay]

Hi i just want to ask if IRF1307 and irf 2708 can be use on class d amp?

 

[Dimonis]

You foggot about irf3710 ,IRF3808,IRF2807 ?

No to all five.

 

[norazmi]

You foggot about irf3710 ,IRF3808,IRF2807 ?

No to all five.

yes u can Irvy

 

[Dimonis]

yes u can Irvy

Why use so heavy mosfets for the max. supply of +-30V ?
Can't find something more suiteble ? IRF540Z is a good choice for that supply.

 

[stewin]

the power supply needs is +/-75vlts

 

[norazmi]

Dimonis, i`ve test up +/- 50 vdc with IRFP140N, IRF540, IRF540N, IRFB4710, IRF540Z all survive with max volume. The new simulation i`ve posted is for +/- 75 with IRFP250N with LTspice, double the mosfet with totem pole will reduce mosfet heat twice and improve thd and output current, good for low impedance also.

Dimonis wanna build this amp? so we can share something maybe modification or any..

regards.

 

[Dimonis]

IRF2807 , 3808 have breakdown voltage of 75 V , so for safe operation the supply must be about +-30V ,and don't forget bus pumping.

 

[Dimonis]

The new simulation i`ve posted is for +/- 75 with IRFP250N with LTspice, double the mosfet with totem pole will reduce mosfet heat twice and improve thd and output current, good for low impedance also.

Why do you use such slow mosfets ? Try IRFB5620 - their total gate charge is 3 times less! Specially designed by IR for D amps.

 

[Dimonis]

Dimonis wanna build this amp? so we can share something maybe modification or any..

Thanks , but i don't like discreet class D::SD

 

[norazmi]

hehe, you should try it Dimonis, this one is cheap and easy part. Tested with totem pole 4 pcs IRFP140N at +/- 50 vdc and 3.15A 250 vac smps fuse blown at high volume. change it to 5A seems work great.

 

[norazmi]

anyone can build this amp, as you can make it directly with Stewin schema at page 1 and what u need to do is to change R10 440 ohm with 1k trimpot. Before power it up, set the trimpot at 1kohm resistance then slowing turn it down until u get the sound produce from the amp. Thats a minimum duty cycle u get to have ucd amp oscillate , dumy load can be put at the output after LC filter about 2.2kohm ot 3.3kohm connec to the ground. After the amp produce the sound, u can measure trimpot and read the resistance value, and to get better thd u can decrease resistance value of trimpot about 10-25 ohm more to get balance good efficiency and thd. Too low R10 will cause ur mosfet hot and lower the efficiency too but thd will be improve. so make it at the middle.

regards.

 

[Dimonis]

Thanks again!
I know that amp , how it works , know people who built it.
But it was long ago , in 2006 a.c. In our days i prefer more modern stuff.

Maybe this will help you in your reserch

http://www.classd.fromru.com/circuits/ucd5.html
http://www.classd.fromru.com/circuits/ucd4.html

 

[stewin]

hi dimonis have you ever felt the power of this amp when built esp if in bridge mode? please try you'll love it. Do you have a non discreet which can use irfp250/350 (even if it means paralleling them per channel) and is affordable to built?

 

[Dimonis]

Hi stewin!
I don't understand why are you using such old and clumsy mosfets like IRFP250? Take one IRFP4227 instead of a pair of '250'
It's if you build a full band amp.
For a subwoofer BTM i myself use IRFB4115 with frequency not more than 200kHz. With their 104A no need of paralleling.