AndrewT said:
Thats what i was saying too. You can use it to earth your speaker, which in that case is just using them as decoupling caps again. Which isnt actually achieving anything.
You wouldnt be able to use it as an earth reference/return point, because it WONT be. Its going to bounce around according to your music signal.
Has anyone here experimented this ? I did, ( in class AB i forgot to precise) many times a long time ago and it was an improvement every time, sonically speaking. But at this time, one did not expect an hifi amp to amplify DC.
In fact the capacitors feed alternatively the output devices with a small quantity of the energy they contain and stay charged at VCC/2. The resistors were 2K2 such as bleeding resistors.
In fact the capacitors feed alternatively the output devices with a small quantity of the energy they contain and stay charged at VCC/2. The resistors were 2K2 such as bleeding resistors.
Hi,
what is the lowest frequency you require your load to reproduce?
What is the impedance of your load?
What should the impedance/resistance be for the resistor ladder?
What should be the impedance of the capacitor ladder for that frequency you chose.
You must select an appropriate impedance for the Virtual Earth to suit the load and frequency it is required to operate at.
I am not referring to DC. I can't hear DC, can you?
what is the lowest frequency you require your load to reproduce?
What is the impedance of your load?
What should the impedance/resistance be for the resistor ladder?
What should be the impedance of the capacitor ladder for that frequency you chose.
You must select an appropriate impedance for the Virtual Earth to suit the load and frequency it is required to operate at.
I am not referring to DC. I can't hear DC, can you?
You do realise bobodioulasso that using the caps as you suggested ARE coupling caps in that role right?
They will still pass the current from the speaker to the rails.
I thought you were wanting to get rid of ALL caps in the output?
As long as the amp has decent dc performance, it will zero itself using the reference from the two resistors, but that reference will be through an inductive messy speaker when you are playing music. That cant be good.
I know what you are trying to do, but for optimal amp performance, its not a good idea, and it does put caps in the output stages, which you are trying to avoid.
And you are also relying on the amp to zero itself properly, without an accurate or stable "earth" value.
They will still pass the current from the speaker to the rails.
I thought you were wanting to get rid of ALL caps in the output?
As long as the amp has decent dc performance, it will zero itself using the reference from the two resistors, but that reference will be through an inductive messy speaker when you are playing music. That cant be good.
I know what you are trying to do, but for optimal amp performance, its not a good idea, and it does put caps in the output stages, which you are trying to avoid.
And you are also relying on the amp to zero itself properly, without an accurate or stable "earth" value.
SpittinLLama said:
The above suggestion was brushed over pretty fast without much consideration.
To the OP: Is it at all possible to separate the left and right channel ground on the headphones?
If it is possible, I think BTL would be the best solution. Not only because it will remove the need for coupling caps with a single supply, but it will also allow for twice the voltage swing over SE.
Of course you will need double the components.
Bob: I guess I'm dim (je suis un peu bête). It just looks to me like you put the coupling cap on the other side of the load. And since we're talking about AC, that changes nothing.
Now you might like the sonic results and I'm sure it works fine, but as a technical matter, there's still a coupling cap in series with the load.
Now you might like the sonic results and I'm sure it works fine, but as a technical matter, there's still a coupling cap in series with the load.
A am not an ingeneer but, IMHO the impedance of the virtual earth, from an AC point of view is the impedance of the paralleled two caps, as rails and ground can be considered as the same thing in an AC point of view. Can we say this impedance high? about half an ohm at 20hz?
May be i am wrong when trying to explain this but i know by experiment that it works nicely.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.