My very first Class D pwm (switching) amplifier.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Thanks for suggestion Pierre!

I think I found the problem.

The others side of 10k resistor (R7) on the SD pin is connencted to the COM pin on the PCB.
Later I removed that resistor and connected it directly. But that was wrong, because the SD pin is referred to the Vss.
So, when I shorted the resistor (R3) between Vss and COM, that crackle disappeared.

That crackle before was not normal. Lets say, that was something like blatter.

I only wanted it to crackle like a class AB amp (or at least similar to it).
I dont want to add clip protection unless I have to.
 
It has always intrigued me why the designers of these amps choose such ridiculous values for some of the components -- on the 'main' pcb R246 is shown as a variable 10K resistor - and R245 is shown as 5.11K in series with it.

5.11K in series with a variable resistor! :confused:

I wonder how much else doesn't actually need to be there :)
 
It has always intrigued me why the designers of these amps choose such ridiculous values for some of the components -- on the 'main' pcb R246 is shown as a variable 10K resistor - and R245 is shown as 5.11K in series with it.

Are you wondering why there is a series resistor or why it is 5.11k ? There is a logical explanation for both of course. But before I give you the answer I let you think a little more !! ;)

Regards

Charles
 
Unfortunately I don't seem to be able to access the schematic again at the moment, but from what I recall the variable resistor enabled the adjustment of a 'monitor' of some sort.
What I don't understand is why it is necessary to be able to vary between 5.11K (+ the tolerance of the minimum pot position) and 15.11K (+ the tolerance of the pot end position).
If the adjustment is likely to be around the variable resistance centre point, and within a range that is reasonably 'easy' to calibrate, then I would have thought 4K7 would have been perfectly adequate.
If you need to calibrate to within 10R (the present resistor is shown to 3 decimal places) how on earth do you 'tweak' a variable to that degree of accuracy (20 turn device?)

And why? -- if your design is that fragile, you've probably got it wrong!
Heaven knows what happens if the ambient temperature changes!!

I've obviously missed the point somewhere here! :)
 
At my end of the pecking order, the purchasing people moan if they have to stock E24 values, let alone E 48, 96 or 192!!

And if I presented a design that could not accept a 4K7 to 14K7 range instead of a 5K to 15K range it would be thrown straight back at me as 'too fussy' !

Now if I worked a little higher up the scale---- :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.