Hi!
Firstly, I post in analogue because I think that class D is analogue rather than digital.
Anyways, I have grown very fond of my ice-power based power amplifier and wonder if it would be possible to implement a preamp based on class d as a gain stage?
The amount of voltage fed to the ice-power modules does not influence the gain, but what does? Does different implementations behave differently in this respect ?
I would like to avoid passive potentiometers if possible..
Any thoughts about this? Is it feasable?
/
Firstly, I post in analogue because I think that class D is analogue rather than digital.
Anyways, I have grown very fond of my ice-power based power amplifier and wonder if it would be possible to implement a preamp based on class d as a gain stage?
The amount of voltage fed to the ice-power modules does not influence the gain, but what does? Does different implementations behave differently in this respect ?
I would like to avoid passive potentiometers if possible..
Any thoughts about this? Is it feasable?
/
Because it's pointless. One can make a case for a class D power amp in terms of cost and weight savings. No case can be made for a pre-amp because they are flea-power to begin with.
But, if you can make a case I'm all ears.
On another note, It's a shame I have to make this reply in a different browser but it seems that DIY has become dysfunctional. I'm going to be complaining about this in every post I make until it's fixed.
But, if you can make a case I'm all ears.
On another note, It's a shame I have to make this reply in a different browser but it seems that DIY has become dysfunctional. I'm going to be complaining about this in every post I make until it's fixed.
Hi,
class D gain stages usually have fixed gain. In self oscillating designs power supply rejection is pretty good, so you would need to atenuate the input like in classic preamp (pot in feedback path is nogo in class D). Carrier based modulation class D designs without feedback have zero power supply rejection, so power supply modulation could be used to some extent. But do not expect more than 20dB gain control in that case. I remember also that there are PWM based analog multiplier designs, but right now I can not think of implementation.
Best regards,
Jaka Racman
class D gain stages usually have fixed gain. In self oscillating designs power supply rejection is pretty good, so you would need to atenuate the input like in classic preamp (pot in feedback path is nogo in class D). Carrier based modulation class D designs without feedback have zero power supply rejection, so power supply modulation could be used to some extent. But do not expect more than 20dB gain control in that case. I remember also that there are PWM based analog multiplier designs, but right now I can not think of implementation.
Best regards,
Jaka Racman
Re: Re: Class D preamp ??
It seems that altering things in the time domain seems easyer than changing the level/svinging voltage. Could this be the reason for class d amps sounding so "free" ?
Anyone have a genious idee about controlling the level without an traditional autenator?
/
peranders said:
It seems that altering things in the time domain seems easyer than changing the level/svinging voltage. Could this be the reason for class d amps sounding so "free" ?
Anyone have a genious idee about controlling the level without an traditional autenator?
/
Hi,
I think class d amps sound so "free" due to the AD conversion being ideally 100% linear, kind of like an ideal class A. I guess that's kind of what you said.
Since you can't control the amplitude of it with traditional means, I think realistically the best bet would be a very good class A pre amp stage. Since it's very low power.. it really shouldn't be a problem.
Regards
I think class d amps sound so "free" due to the AD conversion being ideally 100% linear, kind of like an ideal class A. I guess that's kind of what you said.
Since you can't control the amplitude of it with traditional means, I think realistically the best bet would be a very good class A pre amp stage. Since it's very low power.. it really shouldn't be a problem.
Regards
richie00boy said:
I was thinking if a d stage could replace the ordinary output buffer maybe. If that could be combined with some very smart attenuator, "somehow" within the bridge, I would like to try it. Normal input circuits could be used. My knowledge in "d" is really shallow so ....
Are Elektor articles available on the web? I never subscribed to that mag.
Best regards /
A class-D pre will definitely be a lot more complex than a class-A or B one. I think JLH used an output buffer to ensure the PWM filter was loaded correctly, so the benefit you are hoping to achieve wouldn't be done.
EW (not Elektor
) articles are not available very easily other than looking up the article on softcopy and then ordering the back issue from EW.
I may have the mag somewhere and I have just purchased a scanner though... However, I don't recall the article having a board layout. Maybe not a good first design project.
EW (not Elektor
) articles are not available very easily other than looking up the article on softcopy and then ordering the back issue from EW.I may have the mag somewhere and I have just purchased a scanner though... However, I don't recall the article having a board layout. Maybe not a good first design project.
Aha - I hoped that it would be quite insensitive to it's load. Is this due to the typical high impedance of the input for a typical power amp?
I'm shure I could do anything with some scematics but it would maybe be a good read (if I can grasp it contents).
Please - You must be dying to try that scanner
BR /
I'm shure I could do anything with some scematics but it would maybe be a good read (if I can grasp it contents).
Please - You must be dying to try that scanner
BR /
Maybe I misunderstood, but I got the impression you were trying to eliminate the typical output (op-amp) buffer. If I have remembered correctly, the JLH design does *not* do this, as a buffer is placed after the PWM stage to ensure it's loaded correctly. Thus the power amp input impedance is of little consequence.
TNT,
Didn't you use ICEpower?
As far as I know the ICEpowers run at +12VDC / -12VDC for the Opamps, perhaps you can upgrade the opamp in the inputsection of the ICEpower to an AD8620 or OPA2134.
Perhaps this will make the sound of your setup much better?
As thinking of a Class-D preamp...........
I would definatly go for a passive preamp and increase the input impedance of your amplifier, or a very nice discrete Class-A preamp.
Cheers,
Jan-Peter
Didn't you use ICEpower?
As far as I know the ICEpowers run at +12VDC / -12VDC for the Opamps, perhaps you can upgrade the opamp in the inputsection of the ICEpower to an AD8620 or OPA2134.
Perhaps this will make the sound of your setup much better?
As thinking of a Class-D preamp...........
I would definatly go for a passive preamp and increase the input impedance of your amplifier, or a very nice discrete Class-A preamp.
Cheers,
Jan-Peter
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.