I'm thinking of buildling a chip amp but I'm still unclear on a few things.
1. What's the difference between inverting and non-inverting? Obviously one inverts the signal, but why do people seem to prefer wiring up their chip amps so they invert? What're the sonic differences?
2. What is "snubberizing"?
3. Why do higher capacitances on the power supply hurt the high frequency responce? Surely the bigger the better?
and finally
4. What are the benefits of putting a buffer in front of the chip?
Any info would be much appreciated
1. What's the difference between inverting and non-inverting? Obviously one inverts the signal, but why do people seem to prefer wiring up their chip amps so they invert? What're the sonic differences?
2. What is "snubberizing"?
3. Why do higher capacitances on the power supply hurt the high frequency responce? Surely the bigger the better?
and finally
4. What are the benefits of putting a buffer in front of the chip?
Any info would be much appreciated
bigwill said:How about I make my buffer inverting and have an inverted gainclone so it's the sound of inverted but not inverting!
Actually that works quite well, and it is the approach several very good designers have taken, including Mauro Penasa.
Here is a link.
Non inverting GC with inverting power stage.
bigwill said:I'm thinking of buildling a chip amp but I'm still unclear on a few things.
3. Why do higher capacitances on the power supply hurt the high frequency responce? Surely the bigger the better?
They don't. Anyone who's taken an introduction to analog circuit design course know's that that is simply a ridiculous claim. More capacitance means less ripple on the power supply which is ALWAYS a good thing (if you're final goal is DC, which it is).
and finally
4. What are the benefits of putting a buffer in front of the chip?
An ideal opamp has infinite input impedance. Whatever chip (or circuit in the case of discrete transistors) you endup using will, of course, have a finite input impedance. The addition of a buffer (in the form of another opamp/tube/fets) will increase the input impedance of your amplifier, which should make it more ideal. If you're really wanting to put a buffer in, just use an opamp with a gain of 1 or maybe 2 if you want some extra kick.
Any info would be much appreciated
Hope that helps...
Doovieman
Not to disagree with Doovieman, but too much capacitance can be a problem, from a listening point of view.
e.g. for some time I was running my Rotel bx940 cd player with pair of "huge" 22000 microfarad caps, 10x the original.
It certainly made the sound more robust, punchier, and more time coherent.
BUT, switch to a more moderate size cap (6800) and the above mostly stays but the musicality of the midrange, and dynamics are much improved (probably because there is more silence when instruments stop playing)
I know that seems contradictory, how can "punchy" make less dynamic, but I think these effects are quite frequency dependent.
This is just what I hear, no idea what scoping it would show.
e.g. for some time I was running my Rotel bx940 cd player with pair of "huge" 22000 microfarad caps, 10x the original.
It certainly made the sound more robust, punchier, and more time coherent.
BUT, switch to a more moderate size cap (6800) and the above mostly stays but the musicality of the midrange, and dynamics are much improved (probably because there is more silence when instruments stop playing)
I know that seems contradictory, how can "punchy" make less dynamic, but I think these effects are quite frequency dependent.
This is just what I hear, no idea what scoping it would show.
bigwill said:I'm thinking of buildling a chip amp but I'm still unclear on a few things.
1. What's the difference between inverting and non-inverting? Obviously one inverts the signal, but why do people seem to prefer wiring up their chip amps so they invert? What're the sonic differences?
Any info would be much appreciated
If you look at the inverting and noninverting circuits you will see that the inverting has a much lower input impedance, which is why a buffer prior to the input is preferred for inverting...I suppose.
big caps
Hi, ito big capacitance...
My experience so far has been that big caps (5 pll 10000uf for six channel active) with normal 2200uf(mid/woof) and 1000uf (tweets) near/on chips sound fine, add 150nf MKT's in pll on chips and treble gets smoother, but also softer - wasn't sure if real improvement or just softer, but pleasant sound.
I replaced 150nf's with MKP's (On tweeter amp I used 220nf pll with 22nf just because I had a limited amount of 150nf - but I think 100nf/150nf will be fine) and definite improvement in treble... I can't really say if it was the change in values or the change in dielectric...
Cymbals etc really better - used to be a bit thrashy or just "tsk tsk tsk" when loud or fast (compared to before MKT's didn't listen long enough with MKT's to really evaluate thoroughly)- This didn't really bother untill I listened to much more costly tweets. Now treble is better defined and ringining of cymbals better reproduced even really high. In my system I can now completely hear what is being done with which cymbal - where and how it gets struck, percussive instruments lower down freq. scale also sound better defined probably due to better treble as drums get struck. Violin is better, massed strings I think me tweet is still not up to that sounding real... Oh yes, this is with a ZAR250 tweet (about 20 euro/pounds?) All amps are NI... standard Natsemi diagram without mute. Hitano EXR caps for 1000uf/2200uf
Replaced all MKT's with MKP's am not sure if midrange or bass improved - I think the improved treble just makes what seems to be better midrange - most midrange noises is sandwiched... open or closed with treble.
Cheers
Andre
Hi, ito big capacitance...
My experience so far has been that big caps (5 pll 10000uf for six channel active) with normal 2200uf(mid/woof) and 1000uf (tweets) near/on chips sound fine, add 150nf MKT's in pll on chips and treble gets smoother, but also softer - wasn't sure if real improvement or just softer, but pleasant sound.
I replaced 150nf's with MKP's (On tweeter amp I used 220nf pll with 22nf just because I had a limited amount of 150nf - but I think 100nf/150nf will be fine) and definite improvement in treble... I can't really say if it was the change in values or the change in dielectric...
Cymbals etc really better - used to be a bit thrashy or just "tsk tsk tsk" when loud or fast (compared to before MKT's didn't listen long enough with MKT's to really evaluate thoroughly)- This didn't really bother untill I listened to much more costly tweets. Now treble is better defined and ringining of cymbals better reproduced even really high. In my system I can now completely hear what is being done with which cymbal - where and how it gets struck, percussive instruments lower down freq. scale also sound better defined probably due to better treble as drums get struck. Violin is better, massed strings
I think me tweet is still not up to that sounding real... Oh yes, this is with a ZAR250 tweet (about 20 euro/pounds?) All amps are NI... standard Natsemi diagram without mute. Hitano EXR caps for 1000uf/2200ufReplaced all MKT's with MKP's am not sure if midrange or bass improved - I think the improved treble just makes what seems to be better midrange - most midrange noises is sandwiched... open or closed with treble.
Cheers
Andre
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.