900W H-class PA Amp with Limiter

Its strange to see that apex didn't mentioned about the testing procedure of this amp, this is not a simple class-AB in which you feed a signal and music comes at output and kids start enjoying the show, its class-H and you need to make sure the STEP is working.

I want to see what Apex has to say about this, I will input my 2 cents later on.
 
Last edited:
Its strange to see that apex didn't mentioned about the testing procedure of this amp, this is not a simple class-AB in which you feed a signal and music comes at output and kids start enjoying the show, its class-H and you need to make sure the STEP is working.

I want to see what Apex has to say about this, I will input my 2 cents later on.

If you can get higer voltage on amp out than MV, step driver work ;)
 
If you can get higer voltage on amp out than MV, step driver work ;)

Exactly,

But tell the procedure to the guys who don't even have an oscilloscope, all they have is a multimeter.

A simple solution:

Feed a 50-100hz sine at input and measure the voltage at output with your DMM and see its more than Low rail voltage and touching the high rail voltage or near to it. Its just an approximation of its working.
 
Exactly,

But tell the procedure to the guys who don't even have an oscilloscope, all they have is a multimeter.

A simple solution:

Feed a 50-100hz sine at input and measure the voltage at output with your DMM and see its more than Low rail voltage and touching the high rail voltage or near to it. Its just an approximation of its working.

If clip LED light on, output voltage is higer than low rail voltage.
 
can u pls tel us in details....?

If you use LED instead NSL32, and run amp with 0,+/-55v,+/-110v rail voltage, LED will light on if you have 95v peak voltage on amp out with high level audio input. With low rail voltage of +/-55v, amp will clip at about 50v output voltage, and LED can't light on if step driver not work and not increase rail to +/-110v at peaks.
 
vertex

vertex
 

Attachments

  • APEX%20VERTEX[1].jpg
    APEX%20VERTEX[1].jpg
    546.1 KB · Views: 2,102
Has any body making H900 ever felt the need of adding more transistors to the output.
so here it is PCB+ LAYOUT to add 6pairs of additional transistor . and this can used for VERTEX also.




regards
Sekhar
 

Attachments

  • H900 EXTENTION.PNG
    H900 EXTENTION.PNG
    12.5 KB · Views: 1,507
  • TOP LAYER H900 EXTENTION.PNG
    TOP LAYER H900 EXTENTION.PNG
    17.8 KB · Views: 1,459
  • TOP + BOTTOM  LAYER H900.PNG
    TOP + BOTTOM LAYER H900.PNG
    25.4 KB · Views: 1,408
Exactly,

But tell the procedure to the guys who don't even have an oscilloscope, all they have is a multimeter.

A simple solution:

Feed a 50-100hz sine at input and measure the voltage at output with your DMM and see its more than Low rail voltage and touching the high rail voltage or near to it. Its just an approximation of its working.

And make sure the hexfet isn't getting blazing hot. If the step is a slow boat to China or "sticking" it may "work" but blow up when the party gets started. You also need to make sure the hexfet stays saturated with a long pulse (low frequency) and under heavy load. You might not be able to verify that very easily under light duty test conditions with just a DMM.

You can build amps without an oscilloscope, but don't blame us if "the roof is on fire" means your amp rack. You don't need anywhere near as expensive of a scope as you do for building class D or SMPS, so I seriously suggest getting one.