Hi Daniel,
To your orange question: Elna.
The zobel network capacitor will only affect the sound quality if it is not working. Think about it. A low impedance output will not be loaded by the reactance of that capacitor. Figure out the resistance at 10 KHz to see what I mean. That, and it has a resistance in series with it. All it does is stabilize the amplifier by providing a load at high frequencies. Also, you need a good film type cap here. Some cheap ones will lose their foil over time (= reduced capacitance).
Hi craigg4c, ionomolo,
Solid tantalum caps short when reversed biased or "over voltaged". Nothing has changed here and I still don't trust solid tantalum. Look up the wet slug type.
Hi Andrew,
-Chris
To your orange question: Elna.
The zobel network capacitor will only affect the sound quality if it is not working. Think about it. A low impedance output will not be loaded by the reactance of that capacitor. Figure out the resistance at 10 KHz to see what I mean. That, and it has a resistance in series with it. All it does is stabilize the amplifier by providing a load at high frequencies. Also, you need a good film type cap here. Some cheap ones will lose their foil over time (= reduced capacitance).
Hi craigg4c, ionomolo,
Solid tantalum caps short when reversed biased or "over voltaged". Nothing has changed here and I still don't trust solid tantalum. Look up the wet slug type.
New electrolytics are much better than they were. The main trick is to keep the voltage across the capacitor close to zero. A proper DC bias will correct most of the bad behavior. Film types are still better.
Hi Andrew,
I agree for amplifiers that we are use to. However, comparing the 2R7 to a 4R load isn't as bad. I think they designed the amp for the lower impedances. Do as National has recommended here. Watch for oscillations as board layout can also affect this. Your PCB is a component.
-Chris
Re: Ci yet again still ...
This part of the reason that the input filter and the NFB filter must be designed to work together.
The usual recommendation is that F-3@input>=1.4*F-3@NFB.
The input filter can be the DC blocker back at the source. But that makes interfacing products from different designers more problematic.
this will apply only if the input filter attenuates the LF before they generate a voltage across the NFB cap.craigg4c said:
This part of the reason that the input filter and the NFB filter must be designed to work together.
The usual recommendation is that F-3@input>=1.4*F-3@NFB.
The input filter can be the DC blocker back at the source. But that makes interfacing products from different designers more problematic.
Hi Andrew! I just saw this in action on a TDA7294. It had terribly distorted bass, like every speaker was boomy. . . until the input filter cap was made sufficiently small value capacitance.
And then, finally, the bass played at correct pitches. . . much lower. So what you're saying is observable and applicable.
Thank you!
And then, finally, the bass played at correct pitches. . . much lower. So what you're saying is observable and applicable.
Thank you!
You will get terrible distorted bass if you:
( i ) feed non-HP-filtered audio to your speakers (most probable)
( ii ) make LF signals pass trough an underrated electrolytic (Less noticeable effect).
Of course no one will hear terrible distortion bass as a result of using a normal, properly rated, electrolytic cap as Ci. Caps with lower capacitance or of very poor quality are usually heard as "bright" sound.
BTW This should seem obvious to anyone reading that forum.
EDIT: Myself found LM3886 to be slightly boomy, without input and NFB cap, so it hasn't anything to do with Ci. I remember posts here talking about "bass emphasis" of this amplifier. It might not be a serious issue as it's intended to be used in "compact stereo" and "high end stereo TV's".
Daniel, if you want to feel better go for one of these saying "Audiophile power amps" in its datasheet. If you're only having fun you can wind a transformer like me, but please, try to avoid making people think that choosing a wrong part will make theyr amplifier sound awful. This might have stopped me from diying when it was possible!
Hi ionmolo. Why would you get terrible distorted bass if you feed non-high pass filtered audio to your speakers? If the signal is not distorted and is DC coupled to the amplifier there should not be any problems with distortion. If the amp is DC coupled and the signal fed to the amp has DC offset then you'll just get an amplified DC offset.
This might probably apply only to poor quality loudspeakers, but with my lab speakers subbass intermodulates with low frequencies in a very ugly way. These are $70 loudspeakers!
Note that i'm not saying that an HP filtered signal has less distortion than a DC-coupled one, but that loudspeakers that go to 60 Hz sound awful when reproducing lower frequencies, so the overall appreciation of the sound is better when the signal is filtered.
I havn't tryed any completely dc coupled amplifier with my best loudspeakers but one friend has a vintage amplifier with "remove infrabass" control and bass really sounds better when is turned on.
Note that i'm not saying that an HP filtered signal has less distortion than a DC-coupled one, but that loudspeakers that go to 60 Hz sound awful when reproducing lower frequencies, so the overall appreciation of the sound is better when the signal is filtered.
I havn't tryed any completely dc coupled amplifier with my best loudspeakers but one friend has a vintage amplifier with "remove infrabass" control and bass really sounds better when is turned on.
Well spotted about the bottom plate - it's just a compromise I had to make in the manual manufacturing process
. However, sinks are already oversized, so they never get hot, just warm. If anyone can point me to a place to buy alu plates in Benelux, I would be more than happy to build second one in metalNo shielding too, but no noise is present at all. All wires are twisted though (CAT5E UTP
No black gates or any other snake oiled stuff, sorry
Some people might spot the difference, but not me.
anatech said:
Hi Chris. Yes, that is true. I guess the point I was trying to make was that it really depends on the drivers, the type of enclosure, and how loud the speakers are being asked to play. Even crappy drivers in a properly designed enclosure can play below their -3dB without sounding terrible.
It's also worth pointing out that most recorded material doesn't even have much information at or below around 20Hz, and whatever is down there is often at a lower level relative to the rest of the music. As with anything though, there are exceptions to this.
Hi Brian,
I'm sure that is more than sufficient for this purpose. I guess I was wondering if this would equal or better what you have done. My personal feeling is not, but there are many view out there.
I'll probably try some switching supply caps in a prototype to see if there is any difference. I normally use regulated supplies, so I'll need to rig up a "standard" power supply.
-Chris
I'm sure that is more than sufficient for this purpose. I guess I was wondering if this would equal or better what you have done. My personal feeling is not, but there are many view out there.
I'll probably try some switching supply caps in a prototype to see if there is any difference. I normally use regulated supplies, so I'll need to rig up a "standard" power supply.
-Chris
I don't think there would be any noticeable improvement by using film caps over the 0805 ceramics. Plus, you can't get film caps closer to the pins than the ceramics without soldering them directly to the pins, and even then you probably have lead inductance greater than that of the surface mount ceramics very close to the pins. Don't forget that the film caps will be more expensive as well. Sometimes higher prices make things sound better
Hi Brian,
Exactly!
Yes to every one of your points. However, the film caps would be located across the main filter caps to try and reduce the amount of HF noise from the rectifiers. Ceramics may still work better. At these frequencies we need to watch for the self resonant frequencies of the film types or small electrolytic types. Mind you, small electrolytic types may be too lossy to resonate.
-Chris
Exactly!
Yes to every one of your points. However, the film caps would be located across the main filter caps to try and reduce the amount of HF noise from the rectifiers. Ceramics may still work better. At these frequencies we need to watch for the self resonant frequencies of the film types or small electrolytic types. Mind you, small electrolytic types may be too lossy to resonate.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.