Why is this a problem? 
That is the normal function for a chip amp, or any amp for that matter and depends on the size of the filter caps you're using and the current demands of the amp. Use small filter caps and have a short discharge time. Use big caps and have a longer discharge time. Use monster size caps, or a bank of caps and the thing will play for a long time after power down. 
If your using the National lm3886 controlling the mute pin will let you kill the sound at power down.
Later BZ
That is the normal function for a chip amp, or any amp for that matter and depends on the size of the filter caps you're using and the current demands of the amp. Use small filter caps and have a short discharge time. Use big caps and have a longer discharge time. Use monster size caps, or a bank of caps and the thing will play for a long time after power down. If your using the National lm3886 controlling the mute pin will let you kill the sound at power down.Later BZ
rs1026 said:
RS1026,
The short answer would be: No.
More specifically, every capacitor has some shunt resistance, causing it to drain itself slowly, even if nothing is connected to the terminals of the capacitor.
However, I would not call your issue a real problem. It just shows that your capacitors are in good shape, and capable of supplying your amp for a long time after the power transformer is switched off.
This phenomenon it related to, but not identical to the issue of filtering ripple from the supply. Yes, lowering the capacitor value to very low values cause poor ripple suppression and lower the peak current capability of your PSU, but this is just as much a matter of finding a capacitor with low ESR (Equivalent Series Resistance) and ESL (Equivalent Series Inductance) for ripple suppression. Alternatively, use ceramic capacitors in parallel with your electrolytic ones to lower ESR and ESL.
Your setup works fine, and nothing is wrong. As someone suggested, you may want to use the mute pin. Alternatively, use a relay on the live terminal of your speaker to disconnect them from the amp during power-up/ -down.
Jennice
In my non-chip-amp designs, I use the relay to avoid turn-on "popp" sounds in the speakers.
If you don't have any popp-sounds (probably not with a chip-amp), the use of a relay is only for the puspose of not hearing the capacitors drain. The phenonenon still happens... it's just a matter of avoiding to hear it.
Ceramic capacitors have lower ESR and ESL than electrolytics, especially at higher frequencies (the harmonics of the rectified mains frequency). If you like the sound of your setup, there's no need to make any change.
Jennice
If you don't have any popp-sounds (probably not with a chip-amp), the use of a relay is only for the puspose of not hearing the capacitors drain. The phenonenon still happens... it's just a matter of avoiding to hear it.
Ceramic capacitors have lower ESR and ESL than electrolytics, especially at higher frequencies (the harmonics of the rectified mains frequency). If you like the sound of your setup, there's no need to make any change.
Jennice
Values
I am using 6800uf per rail for filtering. Is this too much capacitance?? I am not sure. How much should the filtering capacitance be?I am using stk4192 based amp. I saw in some gainclone threads that 1000uf should be OK. Is that true?
I thought more would be better
If draining of capcitance is a known issue in chipamps then I will leave with it. Also with the turn on and turn off thump.
Thanks
I am using 6800uf per rail for filtering. Is this too much capacitance?? I am not sure. How much should the filtering capacitance be?I am using stk4192 based amp. I saw in some gainclone threads that 1000uf should be OK. Is that true?
I thought more would be better
If draining of capcitance is a known issue in chipamps then I will leave with it.
Also with the turn on and turn off thump.Thanks
Personally, I don't have a chip amp, but I could never get enough capacitance. The amp I'm buiding at the moment uses 10 x 10.000 uF / 63V in each PSU. (50.000uF for each voltage rail of a 500VA transformer). 
The turn-on/off thump sound is another issue, which you may be able to solve with the mute-pin of the chip (others in this forum know much more about this than I do).
Capacitor draining is not a chip-amp issue, but a natural phenomenon with all capacitors and amps.
Even though the PSU is switched off, there is still some electronics connected to the capacitors. This uses power, which (when the PSU is switched off) only comes from the capacitors. When you use the energy, it will naturally discharge the capacitors.
Even if you would have all electronics disconnected via a relay, whern you switch off the PSU, the capacitors would slowly discharge themselves. All capacitors have an Equivalent Shunt Resistance, which can be compared to a resistor across the terminals of a capacitor. (except that in this case, it is electrons leaking inside the capacitor, as if there was a resistor to conduct).
Thus, you will never be able to switch off your amp and PSU, and then expect the capacitors to maintain full voltage until next time you turn on the amp.
The turn-on/off thump sound is another issue, which you may be able to solve with the mute-pin of the chip (others in this forum know much more about this than I do).
Capacitor draining is not a chip-amp issue, but a natural phenomenon with all capacitors and amps.
Even though the PSU is switched off, there is still some electronics connected to the capacitors. This uses power, which (when the PSU is switched off) only comes from the capacitors. When you use the energy, it will naturally discharge the capacitors.
Even if you would have all electronics disconnected via a relay, whern you switch off the PSU, the capacitors would slowly discharge themselves. All capacitors have an Equivalent Shunt Resistance, which can be compared to a resistor across the terminals of a capacitor. (except that in this case, it is electrons leaking inside the capacitor, as if there was a resistor to conduct).
Thus, you will never be able to switch off your amp and PSU, and then expect the capacitors to maintain full voltage until next time you turn on the amp.
Re: Values
In my SMD Gainclone I have 4700 uF per rail and channel. I get a totally silent turn on and turn off (meaning that the music fades away in a controlled manner)... but... the LM3886 has a MUTE pin which I use. You could use the LM3876 with mute.rs1026 said:I am using 6800uf per rail for filtering. Is this too much capacitance?? I am not sure. How much should the filtering capacitance be?I am using stk4192 based amp. I saw in some gainclone threads that 1000uf should be OK. Is that true?
I thought more would be better
Re: not good to hear
I'd like to return to this statement...
Are you playing loud music?
Is it slow (clearly visible) cone movement, both when playing music and when not, or is it only when playing loud?
How big are your speakers (in particular the drivers you use for bass)?
Jennice
rs1026 said:
I'd like to return to this statement...
Are you playing loud music?
Is it slow (clearly visible) cone movement, both when playing music and when not, or is it only when playing loud?
How big are your speakers (in particular the drivers you use for bass)?
Jennice
driver
The cone movement is only when I play some music. That too when I try to increase the volume from the pc/tv.Since stk4192 has a fixed gain.The only way I can change the level of the music is through vloume control of winamp or say TV.
If I play in low volume then there is no cone movement but only when the signal level increases I notice cone movement.I am using 8 inch drivers for bass.
When I swtich on the amp the cone moves backward once and sticks to the base of the magent.But once I start playing the music it becomes alright. Also there is clear movement of cone when I swtich channels in the TV.
The cone movement is only when I play some music. That too when I try to increase the volume from the pc/tv.Since stk4192 has a fixed gain.The only way I can change the level of the music is through vloume control of winamp or say TV.
If I play in low volume then there is no cone movement but only when the signal level increases I notice cone movement.I am using 8 inch drivers for bass.
When I swtich on the amp the cone moves backward once and sticks to the base of the magent.But once I start playing the music it becomes alright. Also there is clear movement of cone when I swtich channels in the TV.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.