Why not? I can't see any reason preventing to do the same with a crowbar.
As I explained in some detail in a previous comment, it's just a matter of the "logic" controlling it, and where you pick-up its "input" signal. Currently it is taken from the amp out, and of course that will not work. But in the "my_Ref" design there are resistors in series with the chip outputs, so all you need to do is to check for DC before those resistors instead of after them...
I think Panelhead is referring to the turn-on delay. Do you really want to short out the amplifier output every time the amp is turned on?
Tom
Tom
Thank you Sir
The LM3886 configured as a current pump has a nasty turn on transient. It spikes to almost rail voltage. Even with a 0.47 or now a 0.33 ohm resistor in series the smoke will be let out of the chip during regular use.
It is not designed to operate at unity voltage gain.
The amp needs to see an open circuit on power up, then the speaker load after a few seconds.
But the LM3886 does have a protective circui build in. Think National referred to it as SPIKE.
The LM3886 configured as a current pump has a nasty turn on transient. It spikes to almost rail voltage. Even with a 0.47 or now a 0.33 ohm resistor in series the smoke will be let out of the chip during regular use.
It is not designed to operate at unity voltage gain.
The amp needs to see an open circuit on power up, then the speaker load after a few seconds.
But the LM3886 does have a protective circui build in. Think National referred to it as SPIKE.
The SPiKe protection is SOA protection. There're over-current and over-temperature protection circuits as well. It's best to avoid engaging them, especially in a composite amp.
Tom
Tom
how many times did it happen, by accident, also to you, sometimes even with signal on the input?
With no signal on the input (input shorted, too) I don't see what's the problem.
How many times did you try to turn it on with output shorted..? (answer: never, there is the relay +delay applied..)
And the fact that the system endures accidental short circuits, does not mean provoking it mindlessly would be a good practice.
So the system looks like, from your point: a crowbar at output; a disconnect at both rails (strictly synchronized) before or after bypass caps (for each rail)
A short circuit on input
Oh yeah
Ciao, George
And the fact that the system endures accidental short circuits, does not mean provoking it mindlessly would be a good practice.
So the system looks like, from your point: a crowbar at output; a disconnect at both rails (strictly synchronized) before or after bypass caps (for each rail)
A short circuit on input
Oh yeah
Ciao, George
Thank you Sir
Some other My_Ref used a two section relay per channel. The sections are paralleled to minimize contact resistance and improve reliability. Those Schrack relays worked great for this.
That might be a thought is there is a GB15 some day.
Some other My_Ref used a two section relay per channel. The sections are paralleled to minimize contact resistance and improve reliability. Those Schrack relays worked great for this.
That might be a thought is there is a GB15 some day.
Unfortunately that doesn't improve the situation when the relay opens. The inductive kickback from the speaker causes arcing. Best case that destroys the switch contacts and the current stops. Worst case it'll keep arcing. In that case the relay won't protect your speaker against a high DC voltage on the output of the amp.
Have a look: Speaker DC protection with relays
Tom
Have a look: Speaker DC protection with relays
Tom
Adding a capacitor in parallel with the relay contacts to avoid arcing as has always been a good practice would be too hard? 😛
Last edited: