J, why the h*** do you intentionally short a supply?
Wanted to save on the power switch?
This is definitely not a valid testing method to determine the supply's quality or the amps safety!
Most Chip-Amps aren't even specified for supply voltages below a two-digit range. I'd expect shorting the supply would also cause currents from capacitors (or even the speaker) to flow through the low power part of the chip (maybe even in reversed directions).
Fully symmetric designs, OTOH, continue to work with one half of the symmetric supply taken down. Given an exact symmetry of both 'upper' and 'lower' half of an amplifier circuit, no parts of the circuit share 'common' currents. Thus, the 'failed' half of the signal waves is missing in the output signal, leaving it half-rectified.
Prof. Leach claims this for his design (in the FAQ section of his website) as a security benefit, in case one of the supply fuses blows due to overload.
I understand this to mean that both amplifier and speaker are inherently protected by this working principle, as both heat sink temperature and speaker voltage/current swing are reduced.
Again, J, if your amp works with the supply not shorted, then your amp works at all! I understand that you're planning on protecting your design from as many failures as possible.
Would you like to explain why and when you expect an occasional supply short to happen (during normal use)?
edit:
Shorting a power amp's supply (even in low power amps) can be a major health risk. Not that I'm too picky about doing creative things with the own body, but I just had to imagine you
, holding a lead in each hand, closing your eyes, shifting your hands towards each other, waiting for the bang... 
I think you are trying to discharge capacitor
I don´t used to reply in forums, but it´s familiar problem to me, this chip is destruyed inmediatly if you make a spark when you short your supply, (of course with unplugged supply!!!), or if you connect the dc output of the supply to your amplifier with residual charge in capacitors. Any spark destroy the chip.
If you try to discharge capacitor. the best option is to put a discharge resistors welded in paralel with capacitors (500 ohm for ejample)
Jose Luis
my personal multi-amplification page: http://www.nmine.com/multiamp.htm
I don´t used to reply in forums, but it´s familiar problem to me, this chip is destruyed inmediatly if you make a spark when you short your supply, (of course with unplugged supply!!!), or if you connect the dc output of the supply to your amplifier with residual charge in capacitors. Any spark destroy the chip.
If you try to discharge capacitor. the best option is to put a discharge resistors welded in paralel with capacitors (500 ohm for ejample)
Jose Luis
my personal multi-amplification page: http://www.nmine.com/multiamp.htm
This was an interesting problem. You mean when you turn power off you want to disharge the power supply faster and the tool for this is a screwdriver or similar?jpattison said:
The solution is, as said above here, to have a controlled discharge current meaning using a resistor.
A golden rule is NOT to use a screwdriver when you want to discharge. Certain types of caps can actually become damaged due to these huge currents.
jpattison , can you tell more about your total setup and why you want to discharge? There are solutions for this.
peranders said:
I presume that this is directed to me!!
If so!
If you have read the thread from the begining you will see that i allways have tried to help...
In my last post i only try to show some humor...because everybody do this kind of things at the begining.
But who is calling other forum inmates "poor guys"??
The amplifier is used in a bus/coach and needs to be able to handle the power supply being shorted etc....
Have you seen the state of the wiring in some of these buses!?!?!?
I worked out the problem anyhow:
There is a 100uF bias electro on the input amps that biases the output to half-rail. When the supply is shorted, this cap is still charged and makes the bias pin a higher voltage than the supply voltage, destroying the chip.
A diode from this pin (anode) to VIN fixes the problem!!
Cheers for all your help
Have you seen the state of the wiring in some of these buses!?!?!?
I worked out the problem anyhow:
There is a 100uF bias electro on the input amps that biases the output to half-rail. When the supply is shorted, this cap is still charged and makes the bias pin a higher voltage than the supply voltage, destroying the chip.
A diode from this pin (anode) to VIN fixes the problem!!
Cheers for all your help
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.