Advise needed for DC coupling capacitors

Jump to Latest
Alavala

Alavala

Member
Joined 2020
#1
Hello all,
I was just wondering on "settling time after switch-on" of DC blocking capacitors of preamplifiers. Some designs have say 1uf , some 10uf and some have electrolyte 220uf or 470uf. Perhaps in case of 470uf we would have longer settling time. Is using 470uf ok? Or we would face a problem? And if the answer is yes what would be the problem of long settling time? (Although some famous designs use large value bi-polar capacitors. )
Thanks.
 
#4
Alavala said:
Hello all,
I was just wondering on "settling time after switch-on" of DC blocking capacitors of preamplifiers. Some designs have say 1uf , some 10uf and some have electrolyte 220uf or 470uf. Perhaps in case of 470uf we would have longer settling time. Is using 470uf ok? Or we would face a problem? And if the answer is yes what would be the problem of long settling time? (Although some famous designs use large value bi-polar capacitors. )
Thanks.
Click to expand...
Capacitors by themselves define nothing, you will always have an associated resistance or impedance after them, which defines a time constant. which is what is worrying you.
By consense , Audio band is considered 20Hz-20kHz ; capacitors are often rated so associated with next impedance they pass, lower than that say 5 Hz, or even 2 Hz , to have minimal influence at 20 Hz; nothing to gain reaching lower, but even a 1Hz lower limit implies a roughly 1 second time constant .... which is almost a non issue.
So you needn´t worry about any competently designed (pre)amplifier.
 
  • Like
Reactions: ejp, Alavala and NanoFarad
#5
JMFahey said:
Capacitors by themselves define nothing, you will always have an associated resistance or impedance after them, which defines a time constant. which is what is worrying you.
By consense , Audio band is considered 20Hz-20kHz ; capacitors are often rated so associated with next impedance they pass, lower than that say 5 Hz, or even 2 Hz , to have minimal influence at 20 Hz; nothing to gain reaching lower, but even a 1Hz lower limit implies a roughly 1 second time constant .... which is almost a non issue.
So you needn´t worry about any competently designed (pre)amplifier.
Click to expand...
Thank you so much JMFahey, it was helpful
 
#6
It is better to check by calculation if the value was chosen right and to check if there are no coupling capacitors in series in the chain.

This happens with sources that have output caps and amplifiers that have input caps. Two caps in series = lower value.
 
  • Like
Reactions: Alavala
#11
Time constant is just one parameter to consider. The other very important is the DC voltage you block with that capacitor.

Let's say your preamp has about 0.5V DC at the output and the time constant is 1s. That means DC voltage after capacitor will drop to benign ~100mV one second after turn on. But if DC voltage at the output was 20V immediately after turn on, 1s later you would still have dangerous 4V after the capacitor.
 
  • Like
Reactions: Alavala
#12
chip_mk said:
Time constant is just one parameter to consider. The other very important is the DC voltage you block with that capacitor.

Let's say your preamp has about 0.5V DC at the output and the time constant is 1s. That means DC voltage after capacitor will drop to benign ~100mV one second after turn on. But if DC voltage at the output was 20V immediately after turn on, 1s later you would still have dangerous 4V after the capacitor.
Click to expand...
Thank you @chip_mk
 
#13
chip_mk said:
Time constant is just one parameter to consider. The other very important is the DC voltage you block with that capacitor.

Let's say your preamp has about 0.5V DC at the output and the time constant is 1s. That means DC voltage after capacitor will drop to benign ~100mV one second after turn on. But if DC voltage at the output was 20V immediately after turn on, 1s later you would still have dangerous 4V after the capacitor.
Click to expand...
What would be a good cut-off frequency for low-pass filter of an amplifier, considering the least effect on high frequency?
 
#16
In general you don't need low pass, except maybe for RF blocking. There is no clear criteria afaik. In my opinion 200 - 400kHz. Typical implementation with several hundreds pF across input will depend much from the output impedance of the previous stage.
 
  • Like
Reactions: Alavala
#17
chip_mk said:
In general you don't need low pass, except maybe for RF blocking. There is no clear criteria afaik. In my opinion 200 - 400kHz. Typical implementation with several hundreds pF across input will depend much from the output impedance of the previous stage.
Click to expand...
Exactly, then in case of a preamp with wide bandwidth op-amps do you think a low-pass filter with cut-off frequency of 400kHz would be ok?
 
#18
There are two things regarding the upper freq limit of an amp.
1. Amp cirtuit and its compensation. Here you focus on stability and low distortion. Design and used parts dictate the outcome
2. RF input filter

Composite BW of these two factors in a modern amp should be at least 100kHz imho.
 
You must log in or register to reply here.
Top Bottom