Pro: speaker is protected from every failure mode except playing too loud.
Con: Dynamic clipping behavior can be asymmetrical. The 1/2Vcc reference can NEVER fall as fast under load as the main supply (there is an RC constant for noise/hum rejection) so you can get asymmetrical clipping as the supply falls since the reference will move off center momentarily. Only a big deal if the supply sags a lot no load to full load, and if the change in voltage is rapid and repetitive. IOW don’t make the main filter cap too small, even if it’s just a tweeter amp. You still want 10,000 uF (the equivalent of the two 20,000‘s on a split supply in series). Ok to make the output cap smaller if not supporting low bass.
Other con: if the SOA protection kicks in it will be more pronounced and audible than with a split supply.
Con: Dynamic clipping behavior can be asymmetrical. The 1/2Vcc reference can NEVER fall as fast under load as the main supply (there is an RC constant for noise/hum rejection) so you can get asymmetrical clipping as the supply falls since the reference will move off center momentarily. Only a big deal if the supply sags a lot no load to full load, and if the change in voltage is rapid and repetitive. IOW don’t make the main filter cap too small, even if it’s just a tweeter amp. You still want 10,000 uF (the equivalent of the two 20,000‘s on a split supply in series). Ok to make the output cap smaller if not supporting low bass.
Other con: if the SOA protection kicks in it will be more pronounced and audible than with a split supply.
Doing some research I find plenty of chipamp examples with +/- supplies but none with single supply. I suppose the +- versions are preferred as being for the best and the single supply is "not a thing" except for 1 page in a tech document with no explanation.
I briefly tested a LM1875 singleton version which is worth more testing but I have ordered an ESP LM3886 PCB which has no spkr protection yet.
I briefly tested a LM1875 singleton version which is worth more testing but I have ordered an ESP LM3886 PCB which has no spkr protection yet.
It's not just chip amps, just about every solid state power amplifier uses +/- supplies. The single-supply thing was done in the early solid-state era (1960s) and required a capacitor between the output stage and the speaker as the quiescent voltage was half the supply voltage. The capacitor needed to be large (at least as large as the main power supply filter cap) to get good bass response. The advantage was that an output transistor failure would not put DC through the speaker. More reliable power transistors and better amp designs needed +/- supplies but allowed a direct connection to the speaker.
If you need a midrange-only or tweeter-only amp, a single-voltage supply amp with cap output may be appropriate and save money.
If you need a midrange-only or tweeter-only amp, a single-voltage supply amp with cap output may be appropriate and save money.
@benb said: If you need a midrange-only or tweeter-only amp, a single-voltage supply amp with cap output may be appropriate and save money.
I have a NAD 320 which has a lot of hours. I like the enclosure with it's channel relay buttons and tone controls .. it still looks like new. I fixed the 18v regulator which was shorted by a tiny cap. And I replaced a dry cap in the main supply. But now the mainboard is flaky because of heat from the various transistors on small heatsinks and poor ventilation. Maybe root cause can be found with a heat gun and a freeze spray but I am not inclined to do that. Instead I want to replace the mainboard with chipamps such as LM3886.
The NAD mainboard has a UPC1237HA with a POSISTER on the heatsink. If the heatsink gets too hot it will turn off the spkr relay .. eventually.
I remain undecided about the spkr protect issue for chipamp upgrade.
I have a NAD 320 which has a lot of hours. I like the enclosure with it's channel relay buttons and tone controls .. it still looks like new. I fixed the 18v regulator which was shorted by a tiny cap. And I replaced a dry cap in the main supply. But now the mainboard is flaky because of heat from the various transistors on small heatsinks and poor ventilation. Maybe root cause can be found with a heat gun and a freeze spray but I am not inclined to do that. Instead I want to replace the mainboard with chipamps such as LM3886.
The NAD mainboard has a UPC1237HA with a POSISTER on the heatsink. If the heatsink gets too hot it will turn off the spkr relay .. eventually.
I remain undecided about the spkr protect issue for chipamp upgrade.
I have experienced this LM3886 protection many times. The noise is unpleasant enough to turn the volume down, so long term overload of tweeter is not likely.
Never had a tweeter failure because of that.
Regards