Amplifier startup problem

Yes - have a look at the LTSpice model

If a large cap isn't desired, I would go for a bridge. Preferable to making a hard "virtual" ground or split supply. You only need 1/2 the voltage as well.

The zener suggestion is perhaps an alternative.

Isn't that the same thing, just from the perspective of a fixed ground, instead of from fixed power rails?

Sylvia.

.... Phil

There is no specification for common mode input range. Who knows what happens when both input go to -Vee.

In a more typical use, the -ve input would never go close to -Vee.

Can I suggest you check the behaviour of this device when the both inputs are close to -Vee? Its not unknown for gain to become inverted. The behaviour is certainly not characterised.

It simply does, after a fashion. Unconventional, perhaps, but there may be a good reason for all I know.

Bear in mind that the simulation isn't using the LM1875.

Put a 16R load on and set the generator to 1.2V, then look, and it's not too bad, albeit only about 1.6W into 16R.

If you increasing the frequency or increase the 100uF caps you can increase the input (hence the output) a bit.

Dunno.

Cheers

I would be interested to know its behaviour here, and also its behaviour while the power rails increase - it might explain the clicks.

Unfortunately, I don't have a spare one, and buying another five (they come in packs of five from my supplier) seems excessive just to satisfy an interest.

Sylvia.

It depends, some amps have other pins connecting indirectly to supplies too.

NT

Sylvia, how about using another LM1875 as a rail splitter. (How much current at maximum?)

George H.

If you intend to change the circuit to a more conventional one then I would agree, if not then I would have thought it wise to further investigate any form of common failure.

if you have a single supply use figure 2 (page 3) instead, unless you absoulutely need to pass DC to the speaker.