[Shameless Plug] A New Book

On Wed, 12 Apr 2006 12:05:22 -0700, Jim Thompson wrote:


Yeah, well, so was Cinderella!
Cheers! Rich
--
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Congratulations and good wishes.
How software specific is it? In other words, do you think it covers the basic concepts in an accessible way that would also help for someone who might want to implement parts or all of the system in an analog way?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
rex wrote:

I was involved in the copyediting / technical commentary side of this book. _My_ answer to your question (as a disinterested third party who has read the book in probably more detail than the average skimmer) is that as long as you can transform mathematical models into analog circuits, the text is pretty agnostic as to HW vs SW implementation.
A filter is a filter. A summing stage is a summing stage. It doesn't really matter if this is bits moving inside a DSP or implemented as a fistful of R/C/L/opamps. If I recall correctly, there's no sample code with the book, or if there is it's minimal/pseudo, so there is really no bias.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Does it hit on the practical stuff, like windup, slew issues, saturation, auto/manual bumpless transfer, the nonlinear nasties?
John
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
John Larkin wrote:

ISTR talk of saturation and nonlinearity issues. I'd better not answer this in any detail without going back to my archives of the manuscript (if I even still have them). It was quite a while ago when I went through it, so my memories are fuzzy and I don't want to do Tim a disservice.
The book does talk rather a lot about behavior I would describe as idiopathic :)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
John Larkin wrote:

I think I passed over bumpless transfer entirely -- I have an extreme distaste for changing modes, so even though I've implemented systems with bumpless transfer it didn't make it into the book.
I do go into how to limit slew rates elegantly, although not by that name.
There's a chapter on practical nonlinear control: it describes the effects that nonlinearities have on system behavior in general, why you must pay attention to nonlinearities, common nonlinearities including saturation, friction and backlash, and what to do with them without needing a Cray instead of an 8051, and without twisting your brain into a pretzel.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

That seems like a lame excuse for leaving something out of what is supposed to be a "practical" book. Regardless of your "distaste" for it, mode changing and bumpless transfer have been requirements for all of the control systems I've ever seen or worked on.
Perhaps a subtitle should to be added:
"except for the bits the author finds distasteful"
I find Microsoft Windows distasteful, but were I writing a book about desktop PC OSes I don't think I could justify leaving it out...
--
Grant Edwards grante Yow! Yow! I'm imagining
at a surfer van filled with
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Grant Edwards wrote:

It wasn't a conscious decision to exclude it. Over the years I've put in more energy into designing mode changes out of systems to avoid the subject than I have gracefully implementing bumpless transfer. At any rate, I just plain overlooked (d'oh!) the issue, either for the purposes of illustrating how you can avoid the mode changes _or_ for the purposes of explaining and illustrating cases of bumpless transfer.
I've already penciled it in to my markup copy, so if the book makes it to a 2nd edition it'll be in there. In the meantime I'll write a supplement for my website.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I've never found it particulary difficult to impliment mode changes and bumpless transfer, but it's definitly something worth mentioning.
--
Grant Edwards grante Yow! ... this must be what
at it's like to be a COLLEGE
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Grant Edwards wrote:

Agreed, and I'm sorry it didn't occur to me to include it -- It should be doable in a few pages in the right chapter.
As I said, I'll put a supplement on my website.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I make one temperature controller that *cannot* overshoot... it might poach a biological sample that a thousand rabbits died to make. We slew-limit the user's setpoint before we go into the loop and, when we get within 5 degrees of the target, cut the slew rate by 5:1 to creep up on the final value. Anti-windup on the integrator, too, just in case.
Seems to me that control theorists love advanced linear algorithms, when 95% of servo loops are fine with P-I and occasionally P-I-D, but it's the startup/shutdown/windup/slew conditions that are the real challenges.
John
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
John Larkin wrote:

My favourite control paper:
"An Electronic Throttle Control Strategy Including Compensation of Friction and Limp-Home Effects" J. Deur, D. Pavkovic, N. Peric, M. Jansz, D. Hrovat, IEEE Industrial Applications May/June 2004 vol. 40 no. 3 p.821
Cheers Terry
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
John Larkin wrote:

That's probably because its so hard to make any nonlinear results apply to anything that isn't the specific system that generated the results. General nonlinear results are hard to understand, hard to compute, and hard to determine whether they apply. So much of the thrust of control theory is about linear systems, and you're expected to make your nonlinear systems fit.
I like your solution for three reasons: One, it's pragmatic -- I bet the PID controller inside is pretty standard. Two, you didn't have to use many nonlinear elements to make it work right. Three, (and this is my own personal quirk), the need for a nonlinear controller arises not because the _system_ is nonlinear, but because the cost function is not x^2 -- it's an important side of nonlinear control that's not always pointed out well.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
larwe wrote:

Tsk tsk. Not only is their sample code, but the CD has a makefile and a test framework to prove that it's 100% bug free (well, maybe rounded up to the nearest 10).
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

Elsevier never gave me the CD contents ;)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
rex wrote:

I would say that about half the book is software specific, and half is general control. While there is a whole chapter devoted to implementation in software there simply isn't any space devoted to implementation in hardware. If you already know a bit about the Laplace transform and implementing transfer functions with op amps then I think it could be a very useful book.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

http://www.elsevier.com/wps/find/bookdescription.cws_home/707797/description#description

"If you like the food, tell your friends; if not, tell us." Well, OK!
Congratulations! I'm sure this is the culmination of more work than you imagined when you undertook the task. Is a vacation in order?
Jerry
--
Engineering is the art of making what you want from things you can get.

  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Jerry Avins wrote:

I turned the manuscript in at around 9:30 on Christmas Adam (12/23), after pretty much doing nothing but book for all of November and December.
Now I'm starting to think about articles, and supplements that will go onto my website until it's time for the 2nd edition.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 06 Apr 2006 23:21:33 -0700, Tim Wescott

--
Congratulations!


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

"This site requires scripting to be enabled in your browser. Please change your Internet Options, then click here to go to our home page."

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.