https://www.emcstandards.co.uk/mains-filters-might-not-perform-to-spec-and-can
I've just traced down my 20kHz noise in my JCM amp.. turns out it's external and/or being spat out of the IEC filter I have (schaffner 6A jobbie).
With the amp off, the filter on and connected to the PT primary.. there's big spike at 20kHz that I can measure through the insulation of the negative mains output from the filter. There's little on the primary side.
The issue is that the Toroid then couples that directly onto the heater and B+ lines. Mainly the heater 6.3Vac that I'm using.
Edit - replacing the IEC filter, well it's still there.. so obviously external.. still an interesting article.
I've just traced down my 20kHz noise in my JCM amp.. turns out it's external and/or being spat out of the IEC filter I have (schaffner 6A jobbie).
With the amp off, the filter on and connected to the PT primary.. there's big spike at 20kHz that I can measure through the insulation of the negative mains output from the filter. There's little on the primary side.
The issue is that the Toroid then couples that directly onto the heater and B+ lines. Mainly the heater 6.3Vac that I'm using.
Edit - replacing the IEC filter, well it's still there.. so obviously external.. still an interesting article.
Last edited:
resonance merely means unity power factor, how is that amplification?
have you changed the laws of physics?
The laws of physics do not have to be broken to allow for resonance gain.
For example, consider the inductive ballast for a flourescent tube that obtains 1.2kV to strike the arc using only the regular 230V supply. Further, for the resonant example, consider an electronic ballast doing the same by exploiting the Q-factor of the output filter.
For example, consider the inductive ballast for a flourescent tube that obtains 1.2kV to strike the arc using only the regular 230V supply. Further, for the resonant example, consider an electronic ballast doing the same by exploiting the Q-factor of the output filter.
Last edited:
Why not ?
Are you saying that an underdamped mains filter cannot (similarly) amplify noise?
I think that article has valid points.
Reactive passive components are regularly spitting out more energy than they receive, and just as regularly ingesting more energy than they spit out!
You have to define your terms carefully - voltage amplification doesn't require power amplification, similarly current amplification doesn't require power amplification - consider a transformer. Amplify just means "makes bigger", that's all, without context it's ambiguous. We usually expect power amplification, but not always...
And even an RC circuit can amplify voltage (little known fact), though very feebly, it takes LC circuits or transformers for practical passive voltage amplification.
You have to define your terms carefully - voltage amplification doesn't require power amplification, similarly current amplification doesn't require power amplification - consider a transformer. Amplify just means "makes bigger", that's all, without context it's ambiguous. We usually expect power amplification, but not always...
And even an RC circuit can amplify voltage (little known fact), though very feebly, it takes LC circuits or transformers for practical passive voltage amplification.
Last edited:
Come on, reactive components are the one's that accept, store, and release energy. Try charging up a large electrolytic and placing a screwdriver across the terminals(!)
With reactive components you can get resonant voltages or currents exceeding the input (by large factors), and similarly at resonance the impedance can go to very high or very low (which is the point here about noise spectra having peaks with particular mains filter). So the output noise amplitude (voltage or current) can be larger than the input noise amplitude with these reactive networks. Sure the noise power isn't amplified, but that's not what we care about if the noise voltage is what gets into our voltage amplifier input...
With reactive components you can get resonant voltages or currents exceeding the input (by large factors), and similarly at resonance the impedance can go to very high or very low (which is the point here about noise spectra having peaks with particular mains filter). So the output noise amplitude (voltage or current) can be larger than the input noise amplitude with these reactive networks. Sure the noise power isn't amplified, but that's not what we care about if the noise voltage is what gets into our voltage amplifier input...
A resonant filter obtains energy from the input excitation and a part (depending on the Q-factor) of this energy gets focussed at the resonant frequency, which is how the gain happens.
For an input filter like the following, when L = 1mH, C = 1uF and R = 1 ohm, the response has a 30dB gain at 5kHz (below).
For an input filter like the following, when L = 1mH, C = 1uF and R = 1 ohm, the response has a 30dB gain at 5kHz (below).
Read my post again. Voltage amplification doesn't require power amplification. Therefore passive networks can amplify voltage. Or current. But not both at the same time.
People use "amplify" colloquially to mean making anything larger, but technically an amplifier must amplify power to be counted as an amplifier, so passive components can never amplify.
Good points - the concept of an inductor creating energy does't sit right with physics, I agree, however reflecting energy and resonating could but to maintain the laws of physics the energy must be coming from somewhere. Hence resonance isn't free.
It just happened to be a paper in an emissions circular.. however that doesn't mean it's above the laws of physics (in a StarTrek Scotty accent).
It just happened to be a paper in an emissions circular.. however that doesn't mean it's above the laws of physics (in a StarTrek Scotty accent).
