Hi And thanks for reading.
Concerning the smoothing resistor capacitor combination found in most power supplies after the bridge rectifier, designers do add one resistor or more as a cap by itself does not get rid of noise instead shifts it to a different frequency. Adding a resistor does reduce noise far more effectively.
I am wondering if the resistor can be ommited altogether and high esr caps used solely, using the high esr of the cap in place of the resistor.
Is this possible, does it reduce noise effectively.
I have always paid extra and purchased more expensive caps in the belief that lower esr is better, but am wondering now.
Concerning the smoothing resistor capacitor combination found in most power supplies after the bridge rectifier, designers do add one resistor or more as a cap by itself does not get rid of noise instead shifts it to a different frequency. Adding a resistor does reduce noise far more effectively.
I am wondering if the resistor can be ommited altogether and high esr caps used solely, using the high esr of the cap in place of the resistor.
Is this possible, does it reduce noise effectively.
I have always paid extra and purchased more expensive caps in the belief that lower esr is better, but am wondering now.
If we are considering an SMPS then chokes are used to stop any interference from switching pulses returning to the main grid.
Most power supplies, SMPS especially, have Power Factor Correction which deals with any interference and attempts to set the Power Factor nearer 1, otherwise they can be as poor as 20% meaning that your Smart Meter will increase the power factor and charge you the extra.
Transformer fed linear power supplies don't produce interference and just have a class X capacitor across the primary supply.
Expensive over the top electrolytics make little difference except for your pocket!
Most power supplies, SMPS especially, have Power Factor Correction which deals with any interference and attempts to set the Power Factor nearer 1, otherwise they can be as poor as 20% meaning that your Smart Meter will increase the power factor and charge you the extra.
Transformer fed linear power supplies don't produce interference and just have a class X capacitor across the primary supply.
Expensive over the top electrolytics make little difference except for your pocket!
There is some portion of truth in what you say. But there isn't 'High ESR' type cap unfortunately. They are 'usual' or 'Low ESR' tybe. But:
But if we use CRC/CLC (I mean two stage filter but not one stage C-filter) - the we get the possibility to use first C with lower rated capacity (up to several times or up to an order) - it will have higher ESR and higher impedance than the second C (thus lowering the peak charging current), and thus improving the situation a bit. The first cap has the most influence to charging current and radiated electro-magnetic field by rectifier wires. I mean the most important wires in rectifier are all wires between secondary winding and first filtering cap.
As a suggestion is it not wiser to utilise caps with not too low ESR, thinking about the snubber resistor on film caps.
Is it possible the resistance is doing all the snubbing.
If that is true the esr can do the snubbing, if there is enough of it.
In standard practise a resistor soldered in series from the supply to the caps will not lower the oscillation currents flowing between the caps, but does it stop them from getting to the rails
Is it possible the resistance is doing all the snubbing.
If that is true the esr can do the snubbing, if there is enough of it.
In standard practise a resistor soldered in series from the supply to the caps will not lower the oscillation currents flowing between the caps, but does it stop them from getting to the rails
The way I see it is ESR should be as low as you can find, for two reasons:
1) ESR represents increased ripple voltage (worse filtering) on the output. For instance
the difference between a 0.02 ohm and a 0.1 ohm capacitor at 3A load is an extra ~1V
of charge-pulse ripple above that due to load. Charging pulses are typically 3 or more times the steady state load current.
2) The ESR converts ripple current to heat, causing the cap's life to be shortened - dissipating several watts in a cap will run it hot.
Snubbing of rectifier diodes is a separate concern, due to the inductance of the transformer winding causing ringing at RF, typical snubber resistances are of the order of 100 ohms, not 0.1 ohm.
Series resistance, if put anywhere, should be between the secondary and rectifier, to soften and spread out the charging pulses and reduce their rms value. Nothing is a free lunch in a PSU design, this will lose output voltage at load, but it also reduces dissipation in the transformer, the rectifiers and the capacitor's ESR.
Or you can add another RC stage after the main filter caps, which has the advantage of seeing much less of the charging pulses.
Note that ripple current ratings are more than just thermal limits, different caps with similar ESR can have very different ripple current ratings - I found a few 6800µF 63V caps around 0.02 ohms - they varied in ripple current rating between 3.3A and 11.5A... Clearly the latter are much safer choice.
1) ESR represents increased ripple voltage (worse filtering) on the output. For instance
the difference between a 0.02 ohm and a 0.1 ohm capacitor at 3A load is an extra ~1V
of charge-pulse ripple above that due to load. Charging pulses are typically 3 or more times the steady state load current.
2) The ESR converts ripple current to heat, causing the cap's life to be shortened - dissipating several watts in a cap will run it hot.
Snubbing of rectifier diodes is a separate concern, due to the inductance of the transformer winding causing ringing at RF, typical snubber resistances are of the order of 100 ohms, not 0.1 ohm.
Series resistance, if put anywhere, should be between the secondary and rectifier, to soften and spread out the charging pulses and reduce their rms value. Nothing is a free lunch in a PSU design, this will lose output voltage at load, but it also reduces dissipation in the transformer, the rectifiers and the capacitor's ESR.
Or you can add another RC stage after the main filter caps, which has the advantage of seeing much less of the charging pulses.
Note that ripple current ratings are more than just thermal limits, different caps with similar ESR can have very different ripple current ratings - I found a few 6800µF 63V caps around 0.02 ohms - they varied in ripple current rating between 3.3A and 11.5A... Clearly the latter are much safer choice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.