with that inductor (5H) plus load(100mA), 50 uf in the first one and 220 uf afterwards. if you need more recovery, then knock the 220uf down to 50uf
ringing is not a bad thing in the first 100 msec or so, but there should be enough dampening that it never happens again. If it does, then your caps are not keeping up to the demand.
at the end of the power supply, I like to stick a .68 uf cap to snub high frequency noises coming in from dirty power.
ringing is not a bad thing in the first 100 msec or so, but there should be enough dampening that it never happens again. If it does, then your caps are not keeping up to the demand.
at the end of the power supply, I like to stick a .68 uf cap to snub high frequency noises coming in from dirty power.
Hi Dave I tried some simulation (my main is 230V not sure about the hom)
Looks like the best result according with PSDU would be first cap 50u and second cap 100u (the inductor is 10H 90 hom)
So this prove my original idea to increase the first cap to 470u to have more energy available is not a god idea bottom end
The small snub cap should be poly not electrolytic is that correct?
Looks like the best result according with PSDU would be first cap 50u and second cap 100u (the inductor is 10H 90 hom)
So this prove my original idea to increase the first cap to 470u to have more energy available is not a god idea bottom end
The small snub cap should be poly not electrolytic is that correct?
Right capacitance, yes. Best possible quality? - a good way to spend money and give yourself a warm glow, but electrons can't read the brand printed on the outside and don't know and don't care how much you paid for the caps.ciccio64 said:
PS A cap is not a snubber (snubber is RC). Adding a small cap across a big cap (often called 'bypassing') can sometimes make things worse so use with caution, especially if the big cap has low ESR. Bypassing an electrolytic does little harm (but little good); bypassing a big film cap with a small film cap can create a resonance peak somewhere above the audio band.
Last edited:
Did you mean that polyprop are unsuitable as snubber?
Caps for use as snubber are often made of polypropylene due to the fast respons and selfhealing qualities.
Charging a 63V capacitor to 50V is quite different to charging a 500V capacitor to 450V.
The rectifier and the capacitor could be damaged by the massive current pulse at first start up.
A slow charge circuit could be added to limit the charging current. NTC thermistors are designed for exactly this duty: charging a capacitor from a low source impedance.
Look up the design procedure given by all the big manufacturers of NTC Power Thermistors.
It's what you see at the input to every mains powered SMPS. A thermistor+rectifier to charge up the 400V capacitor/s
A resistor is the snub device. The capacitor in series with the snubbing resistor is only there to stop the resistor exploding under the effect of the high DC voltage.
Last edited:
No.
The transient current demand is met by the last capacitor in the power supply. That last capacitor might be the local supply rail decoupling capacitor, not the PSU smoothing capacitor.
The more you put between the first capacitor and the amplifier, the less it can affect the amplifier performance. and even less the audibility effects of that first capacitor.
There is nothing wrong with "designing in" better hum attenuation. But that will not affect the sound quality outcome, unless your amplifier has terrible PSSR, when some additional hum attenaution is welcome. But not the sound quality. The last capacitor will affect the sound quality. It is the one that supplies the transient current demand and much of what you hear is that transient current changing how the speaker cone moves.
Wow! Guys tks a lot I get a lot of info here and I shall need some time to digest all of them
Andrew so in my design were I have a first cap but than an additional LC filter and actually one more RC on the pre section (not show on the attachment) the effect on the sound of the first cap should be negligible and assuming I have a "robust" rectifier should be not a problem to increase the value from 100u to 470u to have a better hum attenuation is that correct?
When you say the last capacitor shall affect the sound quality are you talking about the type of caps used (electrolytic ,film etc) or you talk about the correct capacitance to be used?
Tks
Andrew so in my design were I have a first cap but than an additional LC filter and actually one more RC on the pre section (not show on the attachment) the effect on the sound of the first cap should be negligible and assuming I have a "robust" rectifier should be not a problem to increase the value from 100u to 470u to have a better hum attenuation is that correct?
When you say the last capacitor shall affect the sound quality are you talking about the type of caps used (electrolytic ,film etc) or you talk about the correct capacitance to be used?
Tks
The amplifier draws current pulses, not continuous DC when there is a speaker output.
The changes in current are a transient effect and inductance is their enemy.
This transient current will predominantly come from the nearest capacitor with the least inductance in the flow and return route. Those capacitors that have a higher inductance will supply proportionately less current.
Does that make any sense to you?
The changes in current are a transient effect and inductance is their enemy.
This transient current will predominantly come from the nearest capacitor with the least inductance in the flow and return route. Those capacitors that have a higher inductance will supply proportionately less current.
Does that make any sense to you?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.