Dont tune in, its a tinitus trick you learn, hard to explain how not to, but its possible, think of something else , read a book or watch some interesting TV. If you focus on the sound it does your head in.
Or get a new power supply, when they make noises they are often not far from failing, and fialure rates for PC's are PSU number one.
Or get a new power supply, when they make noises they are often not far from failing, and fialure rates for PC's are PSU number one.
The router is a Siemens and the switch was a D-Link DIR601 and POE.
Its a shame that doctors and researchers haven't found a way to regrow nerves and the tympanic membrane back to the way it should be. I hope that they will in the future find a way to correct medical issues like your son experiences, marce.
Its a shame that doctors and researchers haven't found a way to regrow nerves and the tympanic membrane back to the way it should be. I hope that they will in the future find a way to correct medical issues like your son experiences, marce.
The router is a Siemens and the switch was a D-Link DIR601 and POE.
Its a shame that doctors and researchers haven't found a way to regrow nerves and the tympanic membrane back to the way it should be. I hope that they will in the future find a way to correct medical issues like your son experiences, marce.
Its a shame that doctors and researchers haven't found a way to regrow nerves and the tympanic membrane back to the way it should be. I hope that they will in the future find a way to correct medical issues like your son experiences, marce.
Back in the '60s and '70s, those drove me crazy. I hated the noise. I can't hear them any more, so my assumption is that they've improved the technology.
The unfortunate truth is that, like me, your hearing has deteriorated but also, there are a lot fewer of them around now..
Last edited:
There are several components in an SMPS you can hear, magnetic components undergo stress in response to changes in current, while ceramic capacitors undergo stress in response to voltage. If the SMPS feedback loop is not operating correctly, the switching waveform will have components in the audio band.
Switch mode power supplies (SMPS) generally have poor rejection of low frequency input signals (line harmonics) but do not pass higher frequencies from input to output (though they do introduce ripple of their own). A linear regulator has poor rejection of high frequencies, but very good rejection of lower frequencies.
From an electrical standpoint, following an SMPS with an LDO is valid, and a good practice. Placing a filter at the final output to eliminate switching noise (ripple) is a good idea. You are not trying to remove the switching frequency of, generally, 50 kHz to 1 MHz, but rather the fast edges associated with the switch. A filter with a cutoff of a few MHz between your SMPS output and LDO input will do wonders. Choose a cutoff that is 5 to 10 times higher than your SMPS switching frequency, or half the unity gain bandwidth of the LDO, whichever is lower.
Switch mode power supplies (SMPS) generally have poor rejection of low frequency input signals (line harmonics) but do not pass higher frequencies from input to output (though they do introduce ripple of their own). A linear regulator has poor rejection of high frequencies, but very good rejection of lower frequencies.
From an electrical standpoint, following an SMPS with an LDO is valid, and a good practice. Placing a filter at the final output to eliminate switching noise (ripple) is a good idea. You are not trying to remove the switching frequency of, generally, 50 kHz to 1 MHz, but rather the fast edges associated with the switch. A filter with a cutoff of a few MHz between your SMPS output and LDO input will do wonders. Choose a cutoff that is 5 to 10 times higher than your SMPS switching frequency, or half the unity gain bandwidth of the LDO, whichever is lower.
Getting back to the original question (cough, cough) most 3 legged regulators have high output impedance at high frequencies. Most datasheets will show this. Some datasheets also show PSRR with frequency. It's usually worse as frequency rises. So the LDO regulator may not be very good at all up where the SMPS is making its noise.
That's the first thing to look at. If you have some way to do an FFT analysis of the SMPS output, you'll clearly see the noise. You can then test for noise after the linear LDO reg.
That's the first thing to look at. If you have some way to do an FFT analysis of the SMPS output, you'll clearly see the noise. You can then test for noise after the linear LDO reg.
Pano,
It seems that you would have two types of problems to deal with with the smps supplies. One is the generation of RF and transfer through inductance into a following circuit and the primary you are talking about are the remnants of the high frequency components not being filtered through the wired connections to the next stage. Two separate issues needing two separate approaches to attenuate them.
It seems that you would have two types of problems to deal with with the smps supplies. One is the generation of RF and transfer through inductance into a following circuit and the primary you are talking about are the remnants of the high frequency components not being filtered through the wired connections to the next stage. Two separate issues needing two separate approaches to attenuate them.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.