Ehm.. Im not smart enough to exacly understand what you just said..
Its the standard circuit (The one posted at LM3886 - High-Performance 68W Audio Power Amplifier with Mute ).. Either with 10k pot or a 100k pot..
Its the standard circuit (The one posted at LM3886 - High-Performance 68W Audio Power Amplifier with Mute ).. Either with 10k pot or a 100k pot..
in other words, tell us more about the amplifier and then we can try to advise on the capacitors that may suit either of the two resistance you have provided.
That standard schematic has some errors.
No RF attenuation/filter.
Ci too low in value.
No DC blocking at the input.
Input pin resistances not matched.
No output Zobel.
That standard schematic has some errors.
No RF attenuation/filter.
Ci too low in value.
No DC blocking at the input.
Input pin resistances not matched.
No output Zobel.
Last edited:
RF filter/attenuator is just two small components R + C. these can be added on the front and/or back of the PCB.
Ci may require a physically bigger electrolytic.
DC blocking can be added off board if there is no room on board.
Input pin resistances is just selecting different values of resistors.
Output Zobel is two components, R + C, this time a bit bigger, but can usually be tacked on somewhere near the output connection and near the power ground.
Every thing is sortable.
The one that might be more difficult to sort is separating the Power Ground from the Signal Ground. That is not an error in the schematic. It's understanding that schematic symbols do not differentiate between different grounds used for different purposes.
Ci may require a physically bigger electrolytic.
DC blocking can be added off board if there is no room on board.
Input pin resistances is just selecting different values of resistors.
Output Zobel is two components, R + C, this time a bit bigger, but can usually be tacked on somewhere near the output connection and near the power ground.
Every thing is sortable.
The one that might be more difficult to sort is separating the Power Ground from the Signal Ground. That is not an error in the schematic. It's understanding that schematic symbols do not differentiate between different grounds used for different purposes.
Last edited:
You need to do more research.
But here goes, repeating previous posts by dozens of members.
The Zobel can see big voltages if the amp tries to oscillate or protects itself. Use >=100Vac cap=100nF and use >=2W resistor=2r7
Ci >= Cin * sqrt(2) * Rin / Ri.
Cin=2u2F, Rin=10k, Ri=1k0. then Ci >= 2.2 * 1.414 * 10k / 1k0 >= 31uF. Use 33uF
This will roll off the bass of the amplifier.
Changing Rin to 47k would bring back the lost bass, but this requires Ci>=146uF. Use 150uF or 220uF.
Rf filter can be between 0.33us and 1.5us to listener's taste.
Rb =1k0 requires a cap between 330pF and 1n5F. The top end is severely rolled off if you go as big as 1.5us.
But you have a variable Rs because there is a passive pot in front of Rb.
You end up with a variable RF filter.
You cannot solve this with a passive pot. You have to compromise.
The -ve input pin sees Ri+Ci // Rf1. The cap prevents DC current passing so the -ve pin sees 20k.
The +ve pin sees 1k0 + [effective resistance of pot + Rs] the effective resistance of the pot varies from zero to 0.25 * {Rs + pot value], say from zero to 2625ohms, assuming Rs=500r.
+ve pin sees between 1k0 and 3k625. This is not equal to 20k. The input offset current passes through these two input resistance and generates an input offset voltage of Iinoff * {diff input resistances}.
The output offset voltage is amp DC gain * input offset voltage.
The input offset currents vary with temperature. Look at the datasheet for typical and maximum over the temp range of the chip.
But here goes, repeating previous posts by dozens of members.
The Zobel can see big voltages if the amp tries to oscillate or protects itself. Use >=100Vac cap=100nF and use >=2W resistor=2r7
Ci >= Cin * sqrt(2) * Rin / Ri.
Cin=2u2F, Rin=10k, Ri=1k0. then Ci >= 2.2 * 1.414 * 10k / 1k0 >= 31uF. Use 33uF
This will roll off the bass of the amplifier.
Changing Rin to 47k would bring back the lost bass, but this requires Ci>=146uF. Use 150uF or 220uF.
Rf filter can be between 0.33us and 1.5us to listener's taste.
Rb =1k0 requires a cap between 330pF and 1n5F. The top end is severely rolled off if you go as big as 1.5us.
But you have a variable Rs because there is a passive pot in front of Rb.
You end up with a variable RF filter.
You cannot solve this with a passive pot. You have to compromise.
The -ve input pin sees Ri+Ci // Rf1. The cap prevents DC current passing so the -ve pin sees 20k.
The +ve pin sees 1k0 + [effective resistance of pot + Rs] the effective resistance of the pot varies from zero to 0.25 * {Rs + pot value], say from zero to 2625ohms, assuming Rs=500r.
+ve pin sees between 1k0 and 3k625. This is not equal to 20k. The input offset current passes through these two input resistance and generates an input offset voltage of Iinoff * {diff input resistances}.
The output offset voltage is amp DC gain * input offset voltage.
The input offset currents vary with temperature. Look at the datasheet for typical and maximum over the temp range of the chip.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.