The output capacitor needs time to charge up to half rail voltage.
During that time the speaker should be disconnected and an alternative route provided for the charging current.
Similarly the input and feedback capacitors need to charge up, but they happen relatively quickly and don't need alternative routing for the currents.
During that time the speaker should be disconnected and an alternative route provided for the charging current.
Similarly the input and feedback capacitors need to charge up, but they happen relatively quickly and don't need alternative routing for the currents.
Maybe larger input capacitor could slow entire output capacitors charging process, then it would charge slowly - without audible thump?
Also i did found in other sources that reducing feedback capacitors capacity also minimizes thumps. But it can not be reduced too much...
Dear Nigelwright,
Yes, regarding HUM.
But thump is not HUM and in Single polarity "Gainclone" design it is not related directly with poor grounding. As Mr. AndrewT already mentioned it is related with relatively fast charge (but not as fast as the other in-chain capacitors) of the Output capacitor. This charging current can be also large - and that is why big thump ("POP") is produced to the speaker.
It is bigger problem with "Gainclone" type Chip Amps - because they have internal "undervoltage" protection. You can modify circuit that it could charge output capacitor slowly, but during power on the VDC voltage is rising "undervoltage protection" turns off at about +20VDC - then, at that moment output DC voltage will try to be the same as on the non inverting input terminal giving high current to charge the output cap and this current also goes through speaker.
Last edited:
Thump is usually the output transistors putting out a positive or negative pulse on power up.
Just turning on the output transistors (from standby or mute) shouldn't thump unless the chip amp has a design flaw.
A popular way to get rid of thump is to add a power up delay relay circuit.
It holds off relay for 4 seconds while power stabilises.
The relay could also be part of a DC detect failure circuit.
Have you used recommended standy/mute delay circuit ?
Just turning on the output transistors (from standby or mute) shouldn't thump unless the chip amp has a design flaw.
A popular way to get rid of thump is to add a power up delay relay circuit.
It holds off relay for 4 seconds while power stabilises.
The relay could also be part of a DC detect failure circuit.
Have you used recommended standy/mute delay circuit ?
Update " will try to slow down noninverting input terminals voltage rise during power on by larger RC timing constant. It should help. "
Yes, it helped in one way, but worsen problem in other way - slowly charges - slowly and discharges - if power is turned on, when this input capacitor is not fully discharged - then we have Thump. So I left "softstart-capacitance multiplyer" that helped earlier, and gave up finding another, more simpler way to cure the thump.
Yes, it helped in one way, but worsen problem in other way - slowly charges - slowly and discharges - if power is turned on, when this input capacitor is not fully discharged - then we have Thump. So I left "softstart-capacitance multiplyer" that helped earlier, and gave up finding another, more simpler way to cure the thump.
Big value (kOhms) resistors would not help as this output capacitor should be charged fast enough in this case, small value could help - but also will introduce main drawback - Ripple voltage directly to speaker.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.