Member
Joined 2009
Paid Member
Consider the distortion levels, Kean, they are VERY low and the non-linearities in the CCS will have little effect on the audio signal for two reasons; output is controlled by the buffer EMITTER, which is in turn set by the VAS, and the negative feedback is very powerful as we have such high OLG. A mosfet is cool, sure, but it takes about four volts off the output, limiting us to 16Vpp with clip first appearing on the positive rail. With a bipolar, we can get around 20Vpp. This added headroom is good use of available resources. A 90dB/mW headphone of 2K impedance needs 18Vpp to achieve around 99dBA, which is loud, but we all know some like it HOT.
Hugh
Hugh
Last edited:
Hello Hugh and guy's
I've just seen my email and reply to it.
As I reply in my email it's a very good circuit Hugh, low distortion, stable, and it should sound like an SE amp with better resolution than the previous SE circuit.
All-right, I vote yes for this one.
Ok for a CCS.
Thank
Bye
Gaetan
I've just seen my email and reply to it.
As I reply in my email it's a very good circuit Hugh, low distortion, stable, and it should sound like an SE amp with better resolution than the previous SE circuit.
All-right, I vote yes for this one.
Ok for a CCS.
Thank
Bye
Gaetan
Last edited:
Member
Joined 2009
Paid Member
Wow, I can't even get my computer started as fast as you can complete a sim!
It gets better with practice.
I should know considering I've been doing all the specialized simulation here...
- keantoken
It gets better with practice.
I should know considering I've been doing all the specialized simulation here...
- keantoken
Hello keantoken
You are the Lucky Luke of the simulation... you sim faster than your shadow
Bye
Gaetan
Member
Joined 2009
Paid Member
It gets better with practice.
I should know considering I've been doing all the specialized simulation here...
- keantoken
You've certainly made more contributions that I have managed. Still, it seems we have a 'team' that comprises of lots of very useful skills for the project.
Have you considered if it's worth the trouble extending the simulation to include a small passive model of the headphone rather than a simple 'pure resistor' ?
I'm like the backup-guy who simulates when things get sticky.
I'm certainly not alone here in simulations, though I make an effort to show the differences that usually aren't simulated for whatever reason. I make it a point, if it comes down to guessing and speculating, to simulate instead of guesstimate. I've made some contributions, but I'm limited in that anything I say can't be definitive; I haven't built more than 10 real circuits in my life.
In short, I talk a lot, but don't have anything to show for it.
As for a reactive load simulation, I don't know how necessary it is and I heard it said that headphones are much less reactive than loudspeakers. I think the largest issue where this is concerned is with an AB amp, where it might cross over into B during low load impedances; but this isn't a problem with the current class A design.
- keantoken
I'm certainly not alone here in simulations, though I make an effort to show the differences that usually aren't simulated for whatever reason. I make it a point, if it comes down to guessing and speculating, to simulate instead of guesstimate. I've made some contributions, but I'm limited in that anything I say can't be definitive; I haven't built more than 10 real circuits in my life.
In short, I talk a lot, but don't have anything to show for it.
As for a reactive load simulation, I don't know how necessary it is and I heard it said that headphones are much less reactive than loudspeakers. I think the largest issue where this is concerned is with an AB amp, where it might cross over into B during low load impedances; but this isn't a problem with the current class A design.
- keantoken
Last edited:
a single ended ClassA amp will go into current clip. It cannot supply more than what the CCS is set to.
Hi Lohk,
Nice to see you here!!
This is the precise value of resistor needed to achieve zero offset. In practice, it would simply be a 4K7 resistor supplying a blue LED, around 2.5V, with a 1K pot across it. The wiper would supply the bias voltage, suitably decoupled with a cap.
There is, of course, no readily available resistor of that size! However, even there were, the bias would still vary with applied voltage, so it's adviseable to have a voltage source in that position. An LED is useful, of course, as it gives a visual indication of operation.
Hugh
Nice to see you here!!
This is the precise value of resistor needed to achieve zero offset. In practice, it would simply be a 4K7 resistor supplying a blue LED, around 2.5V, with a 1K pot across it. The wiper would supply the bias voltage, suitably decoupled with a cap.
There is, of course, no readily available resistor of that size! However, even there were, the bias would still vary with applied voltage, so it's adviseable to have a voltage source in that position. An LED is useful, of course, as it gives a visual indication of operation.
Hugh
