precision full wave recitifier on a chip?

[Terry Given]

Glenn Gundlach wrote:

That is one of my worst consistent errors-- transposing digits 2 and 3
of a 4 digit sequence. BUT, the AD8036/7 does make a good fullwave
rectifier.
GG

Years back the company accountant poked his head into the R&D lunchroom,
and announced he knew a maths trick we didnt - namely that when adding a
string of numbers up, if you get 2 different answers and the difference
between the two is divisible by 9, then you have transposed some digits.

So we spent the next 10 minutes analysing it, and came up with a general
proof - for any arbitrary amount of transposition within each number,
the resulting error is always an integer multiple of BASE-1. Not very
helpful in Binary though.

We frightened the poor guy off.

Cheers
Terry

 

[Jim Thompson]

On Wed, 16 Mar 2005 14:57:59 +1300, Terry Given <my_name@ieee.org>
wrote:

Glenn Gundlach wrote:
That is one of my worst consistent errors-- transposing digits 2 and 3
of a 4 digit sequence. BUT, the AD8036/7 does make a good fullwave
rectifier.
GG


Years back the company accountant poked his head into the R&D lunchroom,
and announced he knew a maths trick we didnt - namely that when adding a
string of numbers up, if you get 2 different answers and the difference
between the two is divisible by 9, then you have transposed some digits.

So we spent the next 10 minutes analysing it, and came up with a general
proof - for any arbitrary amount of transposition within each number,
the resulting error is always an integer multiple of BASE-1. Not very
helpful in Binary though.

We frightened the poor guy off.

Cheers
Terry

In my day it was called "Casting Out Nines".

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

 

[Terry Given]

Jim Thompson wrote:

On Wed, 16 Mar 2005 14:57:59 +1300, Terry Given <my_name@ieee.org
wrote:


Glenn Gundlach wrote:

That is one of my worst consistent errors-- transposing digits 2 and 3
of a 4 digit sequence. BUT, the AD8036/7 does make a good fullwave
rectifier.
GG


Years back the company accountant poked his head into the R&D lunchroom,
and announced he knew a maths trick we didnt - namely that when adding a
string of numbers up, if you get 2 different answers and the difference
between the two is divisible by 9, then you have transposed some digits.

So we spent the next 10 minutes analysing it, and came up with a general
proof - for any arbitrary amount of transposition within each number,
the resulting error is always an integer multiple of BASE-1. Not very
helpful in Binary though.

We frightened the poor guy off.

Cheers
Terry


In my day it was called "Casting Out Nines".

...Jim Thompson

In Hex, would that be "F-off" ?!

Cheers
Terry

 

[legg]

On Mon, 14 Mar 2005 11:10:15 -0700, Jim Thompson
<thegreatone@example.com> wrote:

On 14 Mar 2005 10:01:42 -0800, "BobG" <bobgardner@aol.com> wrote:

Thanks all. I need 8 of em to sniff 8 audio channels.. using an 8 ch
a/d avr.... I showed the standard 2 opamp fwr abs val to the layout
guy... he says... dont they have that on a chip yet.... guess he didnt
want to draw 8 of em... thought I'd ask.......

You and what ONE other person would buy it ?

That's what I thought, but illier things have been integrated.

RL