Clipping indicator in a class D amp?

[ThyDntWntMusic]

Looking to build one of these for a car amplifier.

Power Amp Clipping Indicator

its a single channel, 1500watt amp, class D.

is there anything I need to be aware of? as its a single channel, I assume I only need one pair of diodes from the amplifier output into the circuit?

thanks!

 

[ssanmor]

In order to keep it simpler, you can only measure the difference between speaker output and one of the rails (as audio signals are symmetric in mean by nature it will provide a good indication).
In order to do so, you only need a couple of transistors and some resistors, plus a LED.

 

[AP2]

In order to keep it simpler, you can only measure the difference between speaker output and one of the rails (as audio signals are symmetric in mean by nature it will provide a good indication).
In order to do so, you only need a couple of transistors and some resistors, plus a LED.

the circuit complexity has a right, just because the signals are asymmetric in nature (not symmetrical)

 

[Osvaldo de Banfield]

I saw much time ago, a clip detector very easy to use, and understand. Place a transformer across the speaker, with a primary voltage about the maximum expected speaker voltage, and a variable resistor in series. In the secondary, place a neon lamp in series with a resistor of about 100K 1/2W. The lamp will glow when near saturation of amp. Example, if the maximum expected voltage is 24V peak, use a 220 or 110V /24V reversed wired (Low voltage secondary as the primary), and a variable resistor of, say, 1K. In the secondary, the lamp. Then place the amp near the maximum power, and adjust the pot to just bright in audio peaks. It has been used in some older non-expensive tape recorders as a level indicator prior to recording head saturation.

 

[stoc005]

Neon light?? Transformer?? Very old skool from the 60s approach.

 

[Osvaldo de Banfield]

Yes, agree. But, does it do its job? I know that this scheme is easy, non expensive and works fine.

 

[ssanmor]

the circuit complexity has a right, just because the signals are asymmetric in nature (not symmetrical)

I said symmetrical in mean (will you see asymmetries in a blinking LED?) Otherwise the audio would have a DC component most likely.

 

[qusp]

I said symmetrical in mean (will you see asymmetries in a blinking LED?) Otherwise the audio would have a DC component most likely.

AP2, in case you have not heard this usage of mean, it means average

 

[ssanmor]

Thanks!, yes, perhaps "mean" can mean tough or other things in a different context: with "mean" I "mean" "average" 🙂

 

[ThyDntWntMusic]

so, ah, I should build the clipping indicator from that link?

 

[AP2]

Thanks!, yes, perhaps "mean" can mean tough or other things in a different context: with "mean" I "mean" "average" 🙂

hi,
of course, know well in mathematics "mean".
the correct way, regardless of that the result is a LED, or a cell in order to vary the gain "ALC" or other circuit. in principle is correct, the wiring diagram of the link. not be surprised that some companies have developed ALC, with the detect only on one side of the signal. as some stupid current protection in same way. one said...😛

 

[ThyDntWntMusic]

if I were to set the gains using an oscilloscope with no load on the amp, will it still clip at the same volume, even though with the load the voltage rails will sag?

 

[dangus]

I don't know if that circuit will work with a Class D amplifier, but it wouldn't cost much to find out. Or contact Rod Elliott and ask him.

A better approach is to use an amp chip that already has built-in overload detection. The TA2024 does; you just have to add something to buffer the output and stretch the pulse so an LED stays on long enough to be visible. Arjen Helder's TA2020 and TA2024 MkIII amps have overload outputs, but I can't tell if it includes a pulse stretcher. Still, it'll be much easier to work with than having to solder directly to a pin on the chip. http://helderhifi.webnode.com/products/product-2/ http://helderhifi.webnode.com/products/product-1/

The Yamaha YDA148 chip takes it one step further with selectable options that prevent clipping: you can set it so the amp turns itself down, or compresses the signal to prevent clipping.

 

[ThyDntWntMusic]

I'm not entirely sure how a class D amplifier IC would help me?

 

[ThyDntWntMusic]

just found this very simple circuit:

http://www.siliconchip.com.au/cms/gallery/article.html?slideshow=0&a=30810&i=5

I don't need it to be entirely accurate, just need a reasonable guide as to when the amplifier is being pushed into clipping, both on music, and with just tones 😉

 

[dangus]

The subject is "clipping indicator in a class D amp", so aren't you using a class D amp? And that's why I'm concerned that a circuit designed for conventional linear power amps may not work properly with a Class D amp with a bridged output.

Otherwise, sure, the voltage rails will sag. Whether that is a significant problem, I don't know. Small amounts of clipping should be inaudible, unless clipping causes the amp to misbehave in other ways.

 

[ThyDntWntMusic]

I think you might be missing the point.
I have a 1500watt, class D, car audio, monoblock amplifier.

I need an indication of when/how much its clipping.

Another amp I have has a clipping circuit built in, has an LED that just starts to illuminate just as clipping on an oscilloscope occurs, by the time its badly clipping, the LED is fully on.

Not really the point though.

Amplifier doesn't run a bridged output stage.

 

[AP2]

If I understand correctly, you want a circuit that maintains just before squaring the signal? i mean, you can adjust what should be the max?

 

[ThyDntWntMusic]

I want a clipping indicator that starts to illuminate as the amp just starts to clip.

 

[pinkmouse]

I'm going to build the negative rail variant, (I have loads of BD139s), of the SC circuit, I'll let you know how it works out. Maybe next weekend before I get around to it though...