Anyone familiar with this company?
http://www.digitalamp.com/
They make some interesting claims, such as this one:
"Does the DAC4800A wrap feedback around the output filter?
No. Wrapping feedback around the output filter doesn't provide significant improvement in output impedance. We keep the output filter outside the loop, which lowers distortion and provides a superior listening experience. We also use super heavy duty components to insure the lowest natural filter impedance (practically zero)."
And this:
"Why feedback around the filter is bad news...
Well, let me start by saying that the filter's reactance destroys phase margin so the feedback then needs to go through more stages to compensate. The compensation causes delay, and the loop becomes sloppy. Looking at it in the time domain, you never recover from delay.
The difference in THD+N can be quite large, especially when driving a speaker (not just a load resistor as is done for most testing). There's no magic in control theory. I have seen (and tried) modulation methods that actually use the filter to control the modulation, and they suffer from an even worse problem --- the speaker reactance and output current can swing the modulation frequency into the audio band during peaks. There are ways to protect against this effect or minimize it, but that just adds more circuitry. I prefer keeping simple things simple and just doing them right from the start. A properly designed output filter is invisible audio wise.
The bottom line is that I've tried both methods and even spent a few years on this issue! Compensation networks can get complicated, and when you start piling filter on top of filter the effect on the sound can be quite noticeable...
Amps with feedback around the filter just don't sound as accurate. I'll explain later why most Class-D designs on the market aren't all that good, and feedback topology is on the list."
Any comments?
http://www.digitalamp.com/
They make some interesting claims, such as this one:
"Does the DAC4800A wrap feedback around the output filter?
No. Wrapping feedback around the output filter doesn't provide significant improvement in output impedance. We keep the output filter outside the loop, which lowers distortion and provides a superior listening experience. We also use super heavy duty components to insure the lowest natural filter impedance (practically zero)."
And this:
"Why feedback around the filter is bad news...
Well, let me start by saying that the filter's reactance destroys phase margin so the feedback then needs to go through more stages to compensate. The compensation causes delay, and the loop becomes sloppy. Looking at it in the time domain, you never recover from delay.
The difference in THD+N can be quite large, especially when driving a speaker (not just a load resistor as is done for most testing). There's no magic in control theory. I have seen (and tried) modulation methods that actually use the filter to control the modulation, and they suffer from an even worse problem --- the speaker reactance and output current can swing the modulation frequency into the audio band during peaks. There are ways to protect against this effect or minimize it, but that just adds more circuitry. I prefer keeping simple things simple and just doing them right from the start. A properly designed output filter is invisible audio wise.
The bottom line is that I've tried both methods and even spent a few years on this issue! Compensation networks can get complicated, and when you start piling filter on top of filter the effect on the sound can be quite noticeable...
Amps with feedback around the filter just don't sound as accurate. I'll explain later why most Class-D designs on the market aren't all that good, and feedback topology is on the list."
Any comments?
See the PCB layout: Thin power tracks subject to high di/dt and no ground plane. For sure he had a bad time implementing UcD (doesn't tolerate ringing) and gave up. $6000, anyone? 
Edit:
http://findarticles.com/p/articles/mi_m0EIN/is_2005_March_7/ai_n11851528
Ouch... $100000
Edit:
http://findarticles.com/p/articles/mi_m0EIN/is_2005_March_7/ai_n11851528
Ouch... $100000
raintalk said:
What does Digital Amplifier Company have to do with DIYaudio?
Well, it is just possible that their approach to feedback/filtering may be of interest to the DIY community wishing to adapt or implement it for their own projects, just like people have discussed the UCD from hypex and their own homebrew version here....Hope that answers your question.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.