Another series from Castor-Perry

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

bcarso

Well-known member
Joined
Feb 20, 2005
Messages
4,055
Location
San Fernando Valley, Los Angeles
This is some good stuff. There is an RF emphasis but it's clearly applicable to audio nonlinearities.

The first article gives a link to the second, and there are more to come.

http://www.rfdesignline.com/howto/208200124

EDIT: As an brief summary of key results, he shows that the way many use some of the typical series expansion treatments of typical circuit nonlinearities to estimate 3rd harmonic distortion are invalid. The later, higher-order terms are significant contributions to the third, and some of the appropriate series don't converge very fast.

I also appreciated his mention of the Lanczos reference, dating back to the days when insanely powerful computers and math software didn't exist.

I see from his LinkedIn profile that he's looking for work as a consultant, having just finished up on a recent assignment. I wish I could afford his services. At least he's taking advantage of his idle time to think and write :cool: .
 
> The later, higher-order terms are significant contributions to the third

They aren't IF you stick to vacuum tubes and 9:11 rulers laid on the composite curves. The problem is BJTs, and expectations of accuracy.

I think we all knew that quitting at the Third was sloppy. I thought n^2 was needed for very-small error. I'm startled he says 2n+1. But I say 9th, he says 7th, not a big difference in terms of the Third.

> the days when insanely powerful computers and math software didn't exist.

I neatened my clutter, lost my calculator. Yeah, there's a calc in Windows and two on my Palm Pilot, but.... I spent a few days with just a slide-rule. It DOES do resistor-ratios faster than finger-banging.

This flash-back ended when I added-up a cost proposal. Mentally, of course. I came out $500 wrong, and the old grey matter didn't notice for several hours after I posted the numbers to my boss. So I found my 1970s techno-marvel (well, the 1990s implementation which does not eat batteries so bad as LEDs did).

The link to Gilbert's essay on page 2 is nothing new but (if you must use BJTs) worth reading ...the heart and soul of the BJT... ...the well-spring of BJT magic.
 

Latest posts

Back
Top