The Class - H Amplifier

Jump to Latest
kASD

kASD

Member
Joined 2004
Fully Complementary Mosfet Grounded Source Class-H

SMD's are fun to work with.:D
PCB measures 18 X 8 cms and designed for 1.5kW@2ohms
IRFP240/9240 at output && IRFP140/9140 at switching
 

Attachments

  • grounded source class-h.png
    grounded source class-h.png
    50 KB · Views: 1,364
Last edited:
kASD

kASD

Member
Joined 2004
1. 3-Step Class-H waveforms

2. 3-Step SMD version using pairs MJW21195/95

4. 2-Step Mosfet SMD version using IRFP240/9240 & IRFP140/9140 complementary output and Switching


;)
 

Attachments

  • Image0001.jpg
    Image0001.jpg
    180.9 KB · Views: 1,249
  • Image0002.jpg
    Image0002.jpg
    154.8 KB · Views: 1,135
  • Image0004.jpg
    Image0004.jpg
    196.3 KB · Views: 1,131
kASD

kASD

Member
Joined 2004
blmn said:
Very nice work Workhorse, congratulations!

can you tell us something about the levels choice for Vcc steps and the results of your measurements on these beasts?

Regards.
Click to expand...

Hi

If you opt for 2 step i would recommend asymmetrical steps because the music has very high peak to average ratio. for example lower tier voltage =70V an upper tier voltage= 40V so total voltage will be 70+40=110V this gives best results in 2 step class-H, as most of the time music swing will be on its average level only goes to high tier during peaks.

In 3-Step, the rails are best chosen as same voltage difference steps like 55, 110, 165 in 3 steps as used in my prototypes. 3-step is actually good for highpower amps for subwoofer applications and they also facilitate the use of 250V VCE devices at rails +/-165VDC the upper most tier. Efficiency of 3 step is much higher than class-AB and surely cuts waste heat and cost.

I am getting less than 0.1% THD@1khz at full power from these prototypes. power is limited to 2KW@2ohms for 4 pairs of MJW21196/95 pairs.

Thnxz
 
microsim444

microsim444

Disabled Account
Joined 2008
diysmps.com
Workhorse said:
Hi

If you opt for 2 step i would recommend asymmetrical steps because the music has very high peak to average ratio. for example lower tier voltage =70V an upper tier voltage= 40V so total voltage will be 70+40=110V this gives best results in 2 step class-H, as most of the time music swing will be on its average level only goes to high tier during peaks.

In 3-Step, the rails are best chosen as same voltage difference steps like 55, 110, 165 in 3 steps as used in my prototypes. 3-step is actually good for highpower amps for subwoofer applications and they also facilitate the use of 250V VCE devices at rails +/-165VDC the upper most tier. Efficiency of 3 step is much higher than class-AB and surely cuts waste heat and cost.

I am getting less than 0.1% THD@1khz at full power from these prototypes. power is limited to 2KW@2ohms for 4 pairs of MJW21196/95 pairs.

Thnxz
Click to expand...



Hey. Workhorse


Can u share your schematics here :rolleyes:
 
ontoaba

ontoaba

Member
Joined 2009
Workhorse said:
Hi Gold,

Djk is right the actual efficiency with Musical waveform is much different as I already said, with pure sinewave its very much different thing, because the Musical Content features lots of High amplitude peaks in comparision with normal average program material and this is where the Class-H/G amp seeks its advantage and proving more efficiency....

Thanks DJK for valuable comments....

K a n w a r
Click to expand...

That may true:
Exploding the Efficiency Myth of Class D Amplifiers | ECN: Electronic Component News
http://www.newelectronics.co.uk/articles/22076/p21-22.pdf
Nothing more efficient than using headphones. Your big heatsink showing its efficiency problems:D ,nice work for that big one, but smaller is better (for me of course).
 
Last edited:
ontoaba

ontoaba

Member
Joined 2009
I am going to share my drivers here, may be too late but could be used as reference. I have many schematic but nothing of them could be used at my mosfets amplifier, so I make my self drivers. Do not judge from its simple, it is powerful driver.
 

Attachments

  • AMP4D_FET_CLSSH.PNG
    AMP4D_FET_CLSSH.PNG
    20 KB · Views: 963
Last edited:
kASD

kASD

Member
Joined 2004
ontoaba said:
That may true:
Exploding the Efficiency Myth of Class D Amplifiers | ECN: Electronic Component News
http://www.newelectronics.co.uk/articles/22076/p21-22.pdf
Nothing more efficient than using headphones. Your big Heatsink showing its efficiency problems:D ,nice work for that big one, but smaller is better (for me of course).
Click to expand...


Class-H is much efficient than Class-AB, but don't compare it with Class-D which is totally different thing.

Having made Class-AB/H/G/D, Class-D gave me highest of efficiency and reduced Heatsink size.;)
 
You must log in or register to reply here.
Top Bottom