Load Modulated Ripple

[linguist]

I was reading an article on load modulated ripple where they say:


"For op-amp audio applications, the load-modulated ripple problem is not critical, because it means that the power rails will fluctuate with the music, which affects the music but it’s complementary to it. The effect is a “flabbier” sound, and an increase in stereo crosstalk. Contrast a constant 120 Hz ripple like you get from some wall power supplies: this will add a constant 120 Hz undertone to the music, which is far more audible."


I don't really understand, if something affects the sound & makes the sound Flabbier as written, isn't this an issue?

By flabbier do they mean the amplifier is slower to respond, like voltage sag in the old guitar amps etc?

What amount of modulation becomes audible in high quality amps?

If the amplifier can respond as quick as the demand calls for, isn't this best?

If someone has the time could they explain (PSRR) power supply rejection ratio, with an example in Laymans terms so I can understand a bit easier.

Cheers

 

[MikeMl]

The filter capacitor charges during a small part of the full-wave rectified 120 Hz cycle. At 60Hz, the half cycle time is 8.333ms. The charging current pulse lasts ~2msec. For about 6ms, the filter capacitor has to hold up the voltage.

The filter capacitor(s) discharge at a rate dependent on the instantaneous music power. During a quiet passage, the voltage sags very little. During a loud music passage, the voltage sag could be substantial, depending on the source impedance of the power transformer, the capacitance of the filter capacitor(s), if left and right channels have independent power supplies, the peak power being delivered to the speakers, etc. The voltage sag could be reduced by making the filter capacitor(s) bigger, and by having independent power supplies for right and left channels. You dont want a loud passage in one channel effecting the other channel; hence it is better to have independent power supplies.

PSRR is a measure of how good the amplifier is at not feeding the filter capacitor voltage fluctuations through to the speaker. If you have amplifier(s) with good PSRR, you can have a crappy power powe rsupply. If your amplifiers have poor PSRR, then you need well regulated supplies.

 

[Nigel Goodwin]

It's not really a problem for audio amps, they are designed to reject supply fluctuations to a high degree, it's only likely to be a problem at VERY high levels, above the rated specifications for a decent HiFi amp.

Some of the old top HiFi gear, such as the Quad 33/303 used a regulated power supply to reduce any such problems.

 

[linguist]

Thanks MikeMl & Nigel Goodwin for the replies,

I see what your saying, thanks for the explanations.

Cheers

 

[Mr RB]

Some of the old top HiFi gear, such as the Quad 33/303 used a regulated power supply to reduce any such problems.

I built one a couple of years ago Nigel!

Most of the amp is a big voltage regulator on a heatsink providing regulated +/- 35v at plenty of amps.

Then it uses two of the new very low-distortion LM3876 closed loop analog amp chips ("gainclone" style).

When the music stops between songs the speakers sound like a black hole, it's amazing to listen to. Also I believe the amp clarity is greatly increased as all the peaks are unattenuated due to the big regulated supply rails. I did many years of amp repairs and listened to some top units Perraux, Grundig etc and I think my amp beats them all hands down.

**broken link removed**

Roman Black's gainclone amp

 

[Nigel Goodwin]

I built one a couple of years ago Nigel!

Nice one!

I've got some kind of gainclone type amp knocking about at home, an old aquaintance bought it off Ebay from Hong Kong, and it didn't work when it arrived. I had a quick look at it for him, and there was nothing obvious, it looked like one of the chips might be duff?. Nice heatsink case and big transformer.

He didn't want to spend anything on it though, so he told me to dispose of it - it's still sat there in it's box - I might do something with it one day perhaps?.

 

[Mr RB]

It should be easy enough, the chips are so simple they are basically just an op amp. They are DC capable so you can inject DC into the + or - input and test the chip output drivers if the high or low side drivers (or both) are blown.

It's unlikely they blew becasue these chips have serious hardware inside that protects from both thermal overloads and current overloads! When I think of all the years repairing old amps trying to diagnose which tiny PNP on the board was making the crackle, or why the 6 power mosfets in channel B all just blew up again...

These new amp chips give truly amazing performance and reliability in a tiny cheap chip.