Dear friends!
I'm designing a subwoofer amplifier using TDA7294. The summing amplifier and active LP filter is based on the NJM4580.
My questions:
1) I have 3 decoupling capacitors C1, C4, C5. Are they all needed or is it possible, for example, to remove C5? Or vice versa, remove C1 and C4, leaving C5?
2) I have the opportunity to use NJM4580, LM833DT, TL072CDT and LM358DT op amps. Which one is the best?
Thank you very much!
I'm designing a subwoofer amplifier using TDA7294. The summing amplifier and active LP filter is based on the NJM4580.
My questions:
1) I have 3 decoupling capacitors C1, C4, C5. Are they all needed or is it possible, for example, to remove C5? Or vice versa, remove C1 and C4, leaving C5?
2) I have the opportunity to use NJM4580, LM833DT, TL072CDT and LM358DT op amps. Which one is the best?
Thank you very much!
Attachments
C1, C4, and C5 are coupling capacitors, not decoupling capacitors.
C1 and C4 can safely be removed, since the source likely has an output capacitor,
and U1A and U1B have low gains.
C1 and C4 are redundant with C5, which also is a high pass filter.
Do not remove C5, since U2 has much more gain.
Try the TL072 for U1A/B.
C1 and C4 can safely be removed, since the source likely has an output capacitor,
and U1A and U1B have low gains.
C1 and C4 are redundant with C5, which also is a high pass filter.
Do not remove C5, since U2 has much more gain.
Try the TL072 for U1A/B.
Those are coupling capacitors, a.k.a. DC-blockers, not decoupling caps.
I'd say the input caps are good to have - you have no idea what's going to be plugged in(!) However you could sum the inputs with R3 and R6, then pass through a single coupling cap to the input opamp. Either way it usual to have high-value bleeder resistors on the inputs to discharge input capacitors when nothing's plugged in, 1 megaohm or so. This reduces the chance of clicks and thumps when plugging in.
You should add RF-suppression caps directly on each input to ground at the connector(s), perhaps 560pF ceramic, so reduce mobile phone pickup and the like.
C5 allows the TDA7294 to set its bias voltage correctly I think - definitely keep that one.
Of those opamps the LM358 is to be avoided at all costs as it can clips from 5kHz upwards and has high levels of cross-over distortion in this circuit. NE5532 will work fine.
Make sure there are actual decupling caps for all active devices too...
I'd say the input caps are good to have - you have no idea what's going to be plugged in(!) However you could sum the inputs with R3 and R6, then pass through a single coupling cap to the input opamp. Either way it usual to have high-value bleeder resistors on the inputs to discharge input capacitors when nothing's plugged in, 1 megaohm or so. This reduces the chance of clicks and thumps when plugging in.
You should add RF-suppression caps directly on each input to ground at the connector(s), perhaps 560pF ceramic, so reduce mobile phone pickup and the like.
C5 allows the TDA7294 to set its bias voltage correctly I think - definitely keep that one.
Of those opamps the LM358 is to be avoided at all costs as it can clips from 5kHz upwards and has high levels of cross-over distortion in this circuit. NE5532 will work fine.
Make sure there are actual decupling caps for all active devices too...
It is not worth omitting these DC blocking capacitors, but 0.68uF is relatively obscure and I would just use 1uF or 10uF. The schematic asserts that they are a 7Hz high-pass but that is not critical. 100Hz low-pass is pushing the range of a sub and I would put it a bit lower about 80Hz or wherever the full range speakers quit. But you should connect the pot slider to the "X" end of the pots so that if the slider loses connection, the op-amp does not slam to the supply rail. LM358 is not suitable for audio.
There is a problem with the power amp gain control. The gain ranges from about 30dB to ~infinity where nasty things may happen. This can be done a couple different ways but not as shown in the schematic. The low-pass is dubious with 1.6K resistors and 1.5uF cap, and the topology is also not a normal Salen-key. Sorry, I'll not take the time to redesign it for you. Put it into spice and check it.
Thank you very much, fiends!
1) I added bleeder resistors 1M (R15, R16).
2) I connected pots sliders as was recommended. Here I must say, these pots are for initial tuning only inside case and never be set to 0. So protective resistors are not required in my opinion.
3) I increased C1, C4, C5 to 1uF (5Hz now, more then enough for my tiny woofer).
4) I added RF capacitors C18, C19 560pF. But why do I need them if I have low-pass filter set to ~100Hz? Wouldn't it block all interference above that frequency (radio, mobiles, chargers)?
5) I originally planned to use an audio op amp (NJM4580), but was advised to use the generic TL072 and NE5532. Are they really better than NJM4580?
6) To steveu. It's not a Salen-key. It's Multiple Feedback Low-pass (http://sim.okawa-denshi.jp/en/OPttool.php). With these values it can filter from 70 to 150Hz (I wll set exact frequency in initial tuning).
1) I added bleeder resistors 1M (R15, R16).
2) I connected pots sliders as was recommended. Here I must say, these pots are for initial tuning only inside case and never be set to 0. So protective resistors are not required in my opinion.
3) I increased C1, C4, C5 to 1uF (5Hz now, more then enough for my tiny woofer).
4) I added RF capacitors C18, C19 560pF. But why do I need them if I have low-pass filter set to ~100Hz? Wouldn't it block all interference above that frequency (radio, mobiles, chargers)?
5) I originally planned to use an audio op amp (NJM4580), but was advised to use the generic TL072 and NE5532. Are they really better than NJM4580?
6) To steveu. It's not a Salen-key. It's Multiple Feedback Low-pass (http://sim.okawa-denshi.jp/en/OPttool.php). With these values it can filter from 70 to 150Hz (I wll set exact frequency in initial tuning).
