I am looking into implementing a +-45V PSU with synchronous rectification to avoid bus pumping effects when powering a half-brigde UcD. The problem is that I have too little knowledge of synchronous recticifation, I would appreciate if anybody could point me to a circuit of some sort or tell me a little about what I need to think of. One problem I have heard of is mosfets blowing up at PSU power off.
/Daniel
/Daniel
Hi Daniel,
the pumping effect of switched mode power amps (SMPA) Will not and cannot be removed by using (active/passive) synchronous rectification. It will happen more or less distinctive depending on one major factor:
The "topology" of the SMPA - well to be more specific - the output in single-ended or BTL topology is meant. The effect can usually be observed in single-ended amps only and you might see some relics based on pretty bad PCB layouts and/or power-supply wiring schemes with bridge-tied loaded (BTL) SMPAs.
And depending on one minor factor:
The size of the bulk capacitors used in connection with low impedance tracks between them and the switching transistors (basically the complete path where the current is flowing). The effect can be reduced (by using large capacitors) but never eliminated completely.
According to my practical experiences I've observed voltage peeks of 20% and more if the capacitors are too small (it could be fatal for them if the voltage rises above their maximum rating). The bigger the better in case of single-ended SMPA is the rule of thumb. That's by the way one of the reasons why I dislike the UcD amps (bought once one UcD180 for a subwoofer project) - not because they are bad (the performance, etc. is good) but they're always single-ended types (UcD180, UcD400, and now available UcD700). I prefer currently the Tripath chips in BTL configurations.
Speaking of Tripath, I found a paragraph in the datasheet for the TK2350 that might be of interest for you (an extract from the datasheet):
"One solution to the pumping issue it to use large power supply capacitors to absorb the pumped supply current without significant voltage boost. The low-frequency pole used at the input to the amplifier determines the value of the capacitor required. This works for AC signals only.
A no-cost solution to the pumping problem uses the fact that music has low frequency information that is correlated in both channels (it is in phase). This information can be used to eliminate boost by putting the two channels of a TK2350 amplifier out of phase with each other. This works because each channel is pumping out of phase with the other, and the net effect is a cancellation of pumping currents in the power supply. The phase of the audio signals needs to be corrected by connecting one of the speakers in the opposite polarity as the other channel."
This underlines my previous statements ;-)
In general speaking are synchronous rectifier good for improving the over-all efficiency (less power dissipation) with the disadvantage of a (sometimes) complex control of it. That's one of the reasons why it's so common to use Schottky diodes (with their low forward voltage (Vf) value) for rectification.
Last but not least I've never heard of MOSFETs blowing in a synchronous rectifier at PSU power off. This might be a specific design issue of that particular PSU (you've heard of) not a basic topology issue.
I hope this will satisfy you.
Greetings
Corax
the pumping effect of switched mode power amps (SMPA) Will not and cannot be removed by using (active/passive) synchronous rectification. It will happen more or less distinctive depending on one major factor:
The "topology" of the SMPA - well to be more specific - the output in single-ended or BTL topology is meant. The effect can usually be observed in single-ended amps only and you might see some relics based on pretty bad PCB layouts and/or power-supply wiring schemes with bridge-tied loaded (BTL) SMPAs.
And depending on one minor factor:
The size of the bulk capacitors used in connection with low impedance tracks between them and the switching transistors (basically the complete path where the current is flowing). The effect can be reduced (by using large capacitors) but never eliminated completely.
According to my practical experiences I've observed voltage peeks of 20% and more if the capacitors are too small (it could be fatal for them if the voltage rises above their maximum rating). The bigger the better in case of single-ended SMPA is the rule of thumb. That's by the way one of the reasons why I dislike the UcD amps (bought once one UcD180 for a subwoofer project) - not because they are bad (the performance, etc. is good) but they're always single-ended types (UcD180, UcD400, and now available UcD700). I prefer currently the Tripath chips in BTL configurations.
Speaking of Tripath, I found a paragraph in the datasheet for the TK2350 that might be of interest for you (an extract from the datasheet):
"One solution to the pumping issue it to use large power supply capacitors to absorb the pumped supply current without significant voltage boost. The low-frequency pole used at the input to the amplifier determines the value of the capacitor required. This works for AC signals only.
A no-cost solution to the pumping problem uses the fact that music has low frequency information that is correlated in both channels (it is in phase). This information can be used to eliminate boost by putting the two channels of a TK2350 amplifier out of phase with each other. This works because each channel is pumping out of phase with the other, and the net effect is a cancellation of pumping currents in the power supply. The phase of the audio signals needs to be corrected by connecting one of the speakers in the opposite polarity as the other channel."
This underlines my previous statements ;-)
In general speaking are synchronous rectifier good for improving the over-all efficiency (less power dissipation) with the disadvantage of a (sometimes) complex control of it. That's one of the reasons why it's so common to use Schottky diodes (with their low forward voltage (Vf) value) for rectification.
Last but not least I've never heard of MOSFETs blowing in a synchronous rectifier at PSU power off. This might be a specific design issue of that particular PSU (you've heard of) not a basic topology issue.
I hope this will satisfy you.
Greetings
Corax
Thanks for your reply.
Regarding runnig both channels in complementary phase configuration - this generally works, but what happens if the music shifts balance like some dj:s sometimes do? Then I guess pumping will likewise occur, and my protection circuitry will need to work anyway.
Regarding runnig both channels in complementary phase configuration - this generally works, but what happens if the music shifts balance like some dj:s sometimes do? Then I guess pumping will likewise occur, and my protection circuitry will need to work anyway.
Yes, in deed. Under this circumstance it will happen!
And the only way to reduce the excessiv pumping is to make the bulk capacitors as large as possible (depending on the "fill-state" of your wallet ;-) because this could be expencive) .
Take into account that too small capacitors will rise the rail voltage and this could not only be fatal for the maximum rating of the capacitor voltage - in addition it could be fatal for the used semiconductors too. Since you don't know how strong the pumping effect will be, my recommendation would be to select components (SMPA, bulk capacitors, etc.) with at least a 20% higher voltage rating. If it never reaches these voltage levels in your final design you're in luck and your component selection was good.
Have fun
Corax
And the only way to reduce the excessiv pumping is to make the bulk capacitors as large as possible (depending on the "fill-state" of your wallet ;-) because this could be expencive) .
Take into account that too small capacitors will rise the rail voltage and this could not only be fatal for the maximum rating of the capacitor voltage - in addition it could be fatal for the used semiconductors too. Since you don't know how strong the pumping effect will be, my recommendation would be to select components (SMPA, bulk capacitors, etc.) with at least a 20% higher voltage rating. If it never reaches these voltage levels in your final design you're in luck and your component selection was good.
Have fun
Corax
- Status
- Not open for further replies.