Class i and siblings

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Keantoken,

I appreciate you have figured this out.
I also share your concern about the quasi-saturation and how well that is modeled.
Vce of Q5 & Q6, for example, is very low. Happily, these trannies have little work to do, as they only provide the bias voltage for the other trannies (Q3, Q4, Q7 & Q8).

BTW, how well matched are those NXP duals? They are not monolithic, is it?

Cheers,
E.
 
I notice that Pioneer had Class I technology in the past and yet as far as I can see they do not use it in their products today.

Are there any subjective reports of sound quality comparing Class I with the simpler double EF output stage ?

And I have the impression that the Circlophone approach is equivalent in many ways and reports are that it sounds very good.
 
I notice that Pioneer had Class I technology in the past and yet as far as I can see they do not use it in their products today.

Are there any subjective reports of sound quality comparing Class I with the simpler double EF output stage ?

And I have the impression that the Circlophone approach is equivalent in many ways and reports are that it sounds very good.

Psst.. no cimpiraibli i using ears...
if you looking for earable comparison try another one like this DOT. It is comparable with SE 805 + 300B configuration. Even better in treble(high).

But class i+siblings really are fun and interesting for engineering, marketing and visual comparison. Indeed..
 

Attachments

  • hexfet source follower single ended censored.PNG
    hexfet source follower single ended censored.PNG
    66.7 KB · Views: 642
Experiments

Edmond's BFF, syn08, is experimenting with autobias schemas, comparing various non-switching arrangements (from the Linear Technology LT1166 to optocoupler drivers, including Kendall's Class I, Edmond's cloned LT1166 autobias schema, etc...) in terms of long and short term thermal and HF stability, distortion performance, integration with various front ends, etc...

Results with Edmond's autobias are, so far, mixed. The THAT Spice models are as bad as the next, so some changes in the original design are required to make it work. Other changes are required for stability, including managing a nasty RHP zero that is otherwise killing the gain margin when a global NFB loop is closed.

Unfortunately, I am not at liberty to further discuss his stuff here. Edmond or anybody else interested in this stuff may contact him directly.
 
Despite the high complexity of the output stage of class-i works well on resistive load, the distortion at 1 kHz, no more than 0.005% at 30 V (peak) output.
At the same time he did not hold the external load and external load distortion increased more than 100 times (to 0.8%), which in my opinion is not suitable for audio amplifier.

best regards
Petr
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.