moderator, apologies If I have started a second thread in error, but here goes anyway.
This is my attempt below at Hawksford-Cordell (H-C) error correction in a bipolar output stage.
Simulated at 20KHz and 8 Ohm load, EC gives 3.5 times improvement (0.12 down to 0.035) and with 2 ohm load 2x improvment (from 0.39 down to 0.15).
I don't want to get into the feedback/feedforward debate (because its a bit beyond me to be honest) but would appreciate some feedback.
BTW, I see opportunities here to combine EC with sliding bias but there may be loop stability issues that need to be investigated.
This is my attempt below at Hawksford-Cordell (H-C) error correction in a bipolar output stage.
Simulated at 20KHz and 8 Ohm load, EC gives 3.5 times improvement (0.12 down to 0.035) and with 2 ohm load 2x improvment (from 0.39 down to 0.15).
I don't want to get into the feedback/feedforward debate (because its a bit beyond me to be honest) but would appreciate some feedback.
BTW, I see opportunities here to combine EC with sliding bias but there may be loop stability issues that need to be investigated.
Attachments
Some minor corrections and I've used more conventional resistor values. Distortion now down to 0.02 at 20KHz into 8 Ohms and 0.1 into 2 Ohms.
This is quite a good result and I'd expect after placing 20-30dB global feedback around this stage (i.e. as in a complete amplifier system), distortion would be very low.
This is quite a good result and I'd expect after placing 20-30dB global feedback around this stage (i.e. as in a complete amplifier system), distortion would be very low.
Bonsai said:
this will be very interesting, to see your results good linearity in each stage, also in the output stage
via local resistor degeneration feedbacks
or using other local feedback techniques
this is the way to good linearity and low overall power amplifier distortion
- and possibly fast amplifiers keeping linearity
-----------------------------------------------
we have read many times in forum:
global feedback is good .... used in the right way
but not when used as a substitute for bad design and great in-linearity in the local stages
good properly designed local stages
reduces the call for high amounts of global feedback
= high closed loop feedback factor
-------------------------------------------------
This has been expressed, one way or another,
by just about any good knowledge amplifier person
in this forum
Including the Very Master of practical use of This design method: NELSON PASS
and another sensible clever OldTimer: HUGH 'AKSA' DEAN
good luck boys!
you are on the very right track here
and in very very good company
Regards
lineup
.............................................................
Lineup Audio Amplifier Laboratory
http://lineup.awardspace.com/
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.