Subject: Derivative Action, Blessing or Curse
From: dwstclair@aol.com (Dwstclair)
Date: 4 Oct 1996 10:04:14 -0400
To paraphrase the children's story girl "with a curl in the middle of her
forehead", when derivative action is good it can be very, very good, and
when
it is bad, it can be horrid. Of the three controller adjustments typically
available, derivative action is the most unforgiving of a guess-and-test
approach to tuning. If you don't know what you are doing, you are better
off
not using it. Indeed, if you do know what you are doing, you will often
decide you are better off not using it.
Derivative action must be set relative to certain lags in the loop. To
determine these lags it is usually, if not always, necessary to make an
open-loop or a closed-loop test. If too little action is used, it might as
well not be used at all. If too much is used, it can cause instability.
For more information you may refer to a booklet "Controller Tuning and
Control Loop Performance, a Primer" [SECOND EDITION, twice the size of the
first], which is available in the US for $13,
postpaid if prepaid. It has sold over22,800 copies to date. Control
Engineering Magazine has referred to it as "PID Without the Math".
Information is available at
http://members.aol.com/pidcontrol/booklet.html,
or if you wish, I could e-mail you a four-page brochure on the booklet.
David W. St. Clair
Subject: Re: Solid Works CAD/CAM
From: mdmiles@teleport.com (Michael D. Miles, PE)
Date: Fri, 04 Oct 1996 14:40:07 GMT
AEP Inc wrote:
>I'm looking for a competitive product to Pro-E at a much lower cost.
>Solid Works was mentioned as being a close 2nd at around $4,000.
>I want to draw 3-D and generate tool paths 3-Axis for complex shapes for
>our Vertical Mill. We are starting to build plastic injection molds.
>We may also want to hire some freelance for CAD/CAM and CNC files
>to get started. We have more projects than time or manpower.
>Any thoughts would be appreciated.
I'm reading two distinct applications in this description:
1) Generate tool paths for CAM & CNC
2) Design modelling of parts and/or mold design
The market has generally maintained a veil of separation between the CAD and the CAM applications.
Manufacturers of CAM software typically have a 'design' portion in their software that is tuned to
the surfaces thought process (SmartCAM, MasterCAM, SurfCAM, etc). Manufacturers of CAD software
typically have add-in CAM modules or propose a neutral file format to connect to CAM packages. I
don't know of any CAD oriented packages that have CAM as a fundamental part of the product.
Your Application:
If you are generating the part design (creating the geometry to begin with), then using a CAD
package with a CAM module or with a neutral exchange format makes sense. A combination of
SolidWorks and SurfCAM or Pro/Jr and MasterCam or any similar mix and match would be suitable. The
designer would focus on the modelling software and the NC programmer takes the output and generates
the tool paths.
If on the other hand, you are receiving CAD files from other systems (customers), you may want to
consider only the CAM side. Most of the fabricators I deal with only import into their CAM package
and generate the part programming. What little modelling they need to do, they do in the internal
design portion of the CAM package. It is not very productive to model a part again if the geometry
is provided to you. A fundamental principle in the service industry is "be flexible on what you can
import and use". Don't lock yourself into only accepting Pro/E files (or some other package) since
the market will be providing IGES wireframe, IGES surfaces, ACIS, DXF, STEP, and who knows what
else. If you choose to exclude any of those, you are excluding a potential customer.
Since you include building injection molds, that adds a new wrinkle. Proper mold design involves
taking the part geometry and scaling for shrink, designing ejector plate and pins, routing cooling
lines, laying out a runner system, designing slides, lift cores, etc. Using a CAM package alone for
this type of design, while possible, is not really optimum for the design power you might want. If
you subcontract out to a mold designer, then you can simply import those design details into the CAM
package, program the tool paths, and start cutting metal (or graphite).
Suggestions:
For CAM software, the mainstream of SmartCAM, MasterCAM and others are very suitable. My personal
favorite (although I haven't had to use it for any length of time) is SurfCAM (from SurfWare).
For CAD software, there are several 'hot' products and every designer I know will swear by some
particular package to the complete exclusion of all others. There is no 'BEST' until the criteria
by which it is judged is established and agreed upon. My personal favorite is for my design
requirements (sheet metal, injection molded parts, system design) is HP SolidDesigner (to be
released for NT in the next two months) but a lot of good things are being said about SolidWorks. I
have yet to experience the demo.
Entropy Happens!
Michael D. Miles, PE
Consulting Design Engineer
mdmiles@teleport.com
Subject: Re: Comma versus decimal point
From: Kevin O'Connell
Date: Fri, 04 Oct 1996 11:42:22 -0400
Mike wrote:
[snip]
> Also, I couldn't agree more about using
> SI units for engineering and physics. (I was exposed to both systems in
> school over 20 years ago and the choice to make is obvious.)
I might agree more with you if ANYONE truely used the SI system.
I too was taught both for virtually my entire education. SI was pushed
as the "correct" system. But there are two big problems. 1) No one
over here uses it. All construction equipment, and most commerical
equipment is marked, evaluated, rate, and sold in American (British?, English?)
units. Anyone who pretends to use it also publishes in American units
and frequently is actually working to American standards (all in American
units) and just converting (and frequently rounding). 2) No one uses
true SI. They all want to throw in some metric units or even old english
units (metric tonnes, metric feet, or mm or centigrams or whatever).
The one thing I will say for those who use American units is that
once they choose some, they tend to stay with them. They will string
a bajillion zeros after a measurement in inches before they convert them
to feet. And they will stick to seconds even if they are measuring
years. The "SI" crowd wants to skip all around various factor of ten
versions (mm mixed with cm mixed with kilometers) and this spills over into
time which seems to get expressed in seconds, minutes, hours years, decades
whatever. They do the same with angle measurements. They will speak in
degrees and calculate in radians.
The bottom line is I know both, I can work with both (all?) and
the only difference I find is those who work with American units are
much more unit concious and therefore careful.
Kevin
Subject: Re: Impact forces
From: klingener@aol.com (Klingener)
Date: 4 Oct 1996 12:47:06 -0400
In article <325310C3.D6D@nando.net>, Matt Ridge wrote:
> I am in the process of desinging a shear for wire and was wondering how
> to account for momentum of the shear in my calculations. It is obvious
> that if it takes a certain of force (according to direct shear eq.) to
> cut the wire, this can be amount reduced considerably if the shear is
> moving rapidly when it contact the wire. I think it has to do with
> energy or something like that, but I am only guessing.
>
> Any help would be appreciated.
One useful model would be the inelastic collision. The shearing
operartion itself is nonconservative, so forget momentum. First,
estimate the energy required to shear the wire. You may have good data on
this already, but if you have nothing else, multiply the calculated shear
force (handbooks will give guidance on this) by the wire diameter.
Second, compare this to the kinetic energy of the shear tooling before the
'collision.' If the shear energy is much greater than the prior KE, then
you'll have to plan on driving the shear during the operation; if it's
much less, then you can probably simply drop it.
---------------------------------------------
Fred Klingener
Brock Engineering PC, Roxbury CT
klingener@aol.com
---------------------------------------------
Subject: THERMAL SOLUTIONS '97 (Call for Papers)
From: jfritz@js-a.com (Jim Fritz)
Date: Fri, 04 Oct 1996 10:48:30 -0700
CALL FOR PAPERS
Infrared Thermography Topical -- THERMAL SOLUTIONS Œ97
June 24-26, 1997
Cleveland, OH USA
This topical conference will bring together people who are using thermal
tools to find NDT solutions to today¹s industrial, manufacturing and
maintenance problems. While these tools are used in diverse applications,
THERMAL SOLUTIONS Œ97 will provide a common forum for people from all
sectors. Applications using both infrared and various other thermal tools
will be featured.
PAPERS ARE BEING SOUGHT ON THE FOLLOWING TOPICS:
*Maintenance and Manufacturing Applications
*Utility Applications
*NDE in the Aerospace Industry
*Facilities Applications
*NDE in the Petrochemical Industry
*Thermal/Infrared Thermography Standards
Interested authors should submit, BEFORE NOVEMBER 18, 1996, a paper title,
abstract (about 200 words) and co-author names and addresses to:
Topical Coordinator
c/o ASNT Headquarters
1711 Arlingate Lane
PO Box 28518
Columbus, OH 43228-0518
614/274-6003
Fax: 614/274-6899
The topical chairman, John Snell, can be reached directly at 802-229-9820;
fax 802-223-0460; e-mail .
--
Jim Fritz---John Snell & Associates, Inc.
Training, Certification & Support for Thermographers
jfritz@js-a.com http://www.js-a.com/ir
Subject: This is a FLAME! Was Re: Yull Brown's Gas changing the world
From: "Paul Passarelli"
Date: Fri, 4 Oct 1996 17:54:41 GMT
Heather Smith wrote in article
<3253FEF4.1B6B@regina.cochrane-group.ca>...
> Alan Browne wrote:
> >
> > In article <4p4ek5$fp@nnrp1.news.primenet.com>, ltseung@primenet.com
says...
> > >
> > >What should a scientist do if he has a discovery that can bring
> > >immense benefit to the world but on the other hand, can destroy the
> > >world totally?
> > >
Sorry Folks, plug your ears...
THEY SHOULD TAKE THEIR DAMN CAMPAIGN BACK TO SCI.ENERGY.HYDROGEN !!!
SINCE YOU SEEM TO HAVE RUN OUT OF SUCKERS OVER THERE, TAKE THE HINT !!!
My apologies to the sensible readers in the rest of these groups.
Look,
I don't really believe that you are so naive that you don't know what
has been going on in s.e.h But you've spanned 14 groups. Even Lawrence
(self proclaimed Internet big mouth) has publicly admitted that there
might be something fishy going on here. Please stop this. The original
promise was to limit this spam to s.e.h
Face it, no _real_ engineer in any of the groups you spammed is going to
believe the nonsense anyhow, all you're doing is spreading misinformation
and disinformation. The few granola crunchers that you do succeed in
rallying are ultimately going to see through the scam. Then when all is
said and done, all you've accomplished is a waste of bandwidth, period.
> >
> > << snipped >>
> >
> > Does the original message smell of hoax, or is it just me???
> >
> > Alan
> > (This opinion is my own)Just testing this out
>
Subject: Re: friction: impossible to modelize?
From: HW Raedt
Date: Fri, 04 Oct 1996 21:18:16 -0700
Try http://www.ibf.rwth-aachen.de
They have a group dealing with modelling surface effects in mechanical
metallurgy processes. Some of their work is related to numerical
friction analysis. Unfortunately, the page is only in German. Have a
look at it. If it seems to be any good for you (there are some pics,
too), I could try to translate a bit for U.
HW Raedt
Enrique Juaristi wrote:
>
> After having read some (old) references, I am almost convinced that friction
> is almost impossible to modelize, be it analytically or numerically.
> There are a lot of factors to be taken into account and most of them are either
> impossible to measure or unpredictable.In the project I am working on, the
> friction between two hot rolled steel products is being studied.
>
> I'd like to know whether someone knows of techniques, hypotheses,
> experimental settings or references that could allow me to create a
> realistic model.
>
> Thanks,
>
> --
> Enrique Juaristi, mechanical engineer
> Centre de Recherche Public Henri Tudor
> 6, rue Coudenhove-Kalergi
> L-1359 Luxembourg
> Phone: +352 42 59 91 244
> fax: +352 43 65 23
Subject: Temperature Sensor/Logger
From: Norman Carlson
Date: Fri, 04 Oct 1996 15:24:05 -0700
I work for a company that has just released a new temperature
sensor/logger that
has many unique features and is very cost effective. It will interface
directly with an RS232C port. It can be used as a data logger in a stand
alone mode or can be polled periodicaly. Its main attractiveness is
price, size(0.6" x1.5"x2.13") and it will operate for 2 years on a single
battery. I would be interested in your feedback. Here is a recent press
release.
If you are interested, please let me know.
Worlds’ smallest, reusable battery operated, single channel
temperature data logger available for $49.
MadgeTech introduces the Temp101 Temperature Recorder, a cost effective,
NIST
traceable, next generation temperature data recorder. The Temp101 is the
latest addition to a series of miniature, economical, battery powered
intelligent temperature dataloggers. The Temp101 is a stand-alone device
used
to automatically record up to 16,535 temperature measurements from -40
deg F
to +165 deg F. Its 2 year battery life and real time clock insures all
data
is time and date stamped. Its reading rate can range from 1 per 2 seconds
to
1 per day. There are no buttons or switches to add unnecessary
complexity.
Its small size allows it to fit almost anywhere. Applications include
temperature monitoring, temperature recording of perishable goods in
transit,
replacement of expensive and cumbersome strip chart recorders, HACCP
compliance, HVAC, environmental research, ocean or pond
studies, refrigeration, artworkpreservation, museum monitoring,
agriculture,
etc. In offering the Temp101,
MadgeTech has combined the latest in lowpower technology with our
temperature
logging software, making it easy for users to graph and analyze data.
Our
software requires no programming skills and enables users to effortlessly
select reading rate, device ID and initiate the start of data collection
within moments after users connect their hardware. MadgeTech’s
temperature
logging software automatically determines what hardware is being used and
configures itself appropriately. Programming of the device is automatic.
All that is required of the user is to select device specific
configuration
parameters such as device ID, reading rate, etc. Data retrieval is
simple.
Plug it into an empty com port and our software does the rest. The user
may
choose among numerous graph formats and has access to a wide variety of
data
analysis options. For further analysis, all data is easily exportable
to
spreadsheets.
* Memory capacity: 4096 readings expandable to 16,535. Readings are
saved
even when battery is removed.
* Dimensions: 0.6" x 1.4" x 2.15" (15mm x 35mm x 54mm)
* Operating temperature range: -40 deg F to +165 deg F
* Temperature Accuracy: ±0.5 deg C from 0 to +60 deg C
±1.0 deg C from -40 to +75 deg C
* Downloading Data: Data may be downloaded any anytime to a host
computer at
2,000 readings/min.
* Temperature Sensor: Internal
* Response Time: 15min
* Battery Life: 2 years(comes with 3.6 Volt user replaceable battery).
* Time Accuracy: ±100 ppm at 20°C
* User selectable recording time: 2 seconds, 5 seconds, 10 seconds,
15 seconds, seconds, 1 minute, 5 minutes, 10 minutes, minutes,
30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, 24 hours.
* Data Format: Date and time stamped, programmable in deg F, deg C, deg K
or deg R. Data exportable to spreadsheet programs (Lotus, Excel,
etc.)
* Weight: 0.9oz. (25.5 grams)
* Visual indicator: LED flashes at selected reading rate.
* Computer Interface: Using our interface cable, the Temp101 is
compatible
with any serial or RS232C port.
* Software: Compatible with MadgeTech’s Windows based software for
complete
control and operation.
1996 Price List
* Temp101-1:
battery operated, single channel temperature data logger
for 4,096 readings: $49
* Temp101-2 :
battery operated, single channel temperature data logger
for 8,192 readings: $64
* Temp101-3 :
battery operated, single channel temperature data logger
for 16,535 readings: $79
* IFC-1 :
Interface Cable and Software and manualto work with all
MadgeTech’s data logger products: $19
Delivery is from stock.
Please call or fax if you need any additional information.
Thank You
Norman Carlson
phone: (603) 456-2011
fax: (603) 456-2012
email: madgetech@conknet.com
Subject: Re: Question: How to get starting ME job
From: hippy@acca.nmsu.edu (Hippy The Original)
Date: 4 Oct 1996 20:41:07 GMT
In article <531oeq$lm9@newsbf02.news.aol.com> you wrote:
: Specializing in not the problem "interest" might be the problem. What
: part of ME did you enjoy, why? Look for job in that area. I'm looking for
: a ME but not one who has no particular interst. GPA, I think
My interests are in general heat transfer and thermal systems. in
particular I am a gear head, internal combustion engines. I have tryed
getting a position with ford, gm, and honda r&d.; Ford will not look at me
since I do not have a grade point average greater than 3.2/4.0. The other
companies I just keep getting letters saying that they have recieved my
resume and "will keep it on file...".
My other interest is in advanced composites. My senior design
project was a carbon fiber/ aluminium mountain bike frame. If I could do
both I think that I would be a very happy employee. After looking for a
(any) position for over a year the problem that I see is either my
apperance (long hair), my attitude (a little cut and dry till I get to
know people), or my grade point. And as we all know they can't take the
first two into consideration when hireing or that would be discrimination
(yeah right). --
*************************************************************************
* The Original Hippy on Acca *
* Rich Murphy BSME *
* hippy@acca.nmsu.edu hippy@corholio.NMSU.EDU *
* rmurphy@nmsu.edu *
* *
* The opinions expressed are mine, If I was employed they would *
* STILL be mine. But if anyone would like to change the *
* employment part I would gladly listen to them. *
* *
*************************************************************************
Subject: Re: Which FEA package? Does anyone use ALGOR?
From: Mike Porter
Date: Fri, 04 Oct 1996 09:22:45 -0500
Jon Juhlin wrote:
>
> I see much discussion in this newsgroup about various FEA packages, and
> seldom see ALGOR mentioned. On the other hand, the package I see
> advertised most in Mechanical Engineer and various other trade magazines
> is ALGOR. Presumably they are selling their software to someone.Does
> anyone use this package? Those people who respond to queries regarding
> FEA seem to be deeply involved in the field, whereas the persons asking
> for advice seem to be looking for an auxiliary tool to use when when
> classical methods require too many broad assumptions...just ANOTHER tool,
> not a replacement for ALL tools. Is ALGOR a reasonable choice of FEA
> software for the rest of us?
> Putting aside the advertizing issues, Algor is a good general purpose FEA tool.
It is structured a bit differently than most of the other FE packages (you
don't directly specify elements, you build the geometry and a "decoder"
constructes the elements), but once you get the hang of the way the system
operates IMHO it is quite efficient.
I reciently publised an ASME PVP paper where we compared answers to a piping
nozzle problem run with 5 of the PC based solvers. All came up with
reasonablly close answers. Algor was one of the packages used.
Like _ALL_ FE packages, it has its quirks and occasional bugs. Nonetheless, it
will do most of what many of use need to do in every day design work.
--
Mike Porter Stress and Vibration Consulting Services
Dynamic Analysis FEA Consulting and Training
(913) 341-3269
Subject: Re: friction: impossible to modelize?
From: rongraham1@aol.com (RonGraham1)
Date: 5 Oct 1996 02:55:08 -0400
In article ,
enrique.juaristi@crpht.lu (Enrique Juaristi) writes:
What's the .lu domain? Never seen it before. (Please don't let's
have everyone in the world answer this question. :-))
>After having read some (old) references, I am almost convinced that
friction
>is almost impossible to modelize, be it analytically or numerically.
>There are a lot of factors to be taken into account and most of them are
>impossible to measure or unpredictable.In the project I am working on,
the
>friction between two hot rolled steel products is being studied.
I'm assuming the problem you are having is with Coulomb friction,
where the opposing force is probably some constant times the
sign of the relative velocity of the two bodies in contact. Am I right?
I also assume you are primarily concerned with simulation, since
this behavior really doesn't lend itself very well to linear analysis.
When I was at NASA, we handled that case in simulation with an
iterative procedure, iterating on (a) the system accelerations, based
on the friction found in the previous step, and (b) the friction, based
on the acceleration found in (a) -- *except* for replacing the contact
acceleration by (-velocity / time increment), the acceleration that
gets you stuck in one time step. When the iteration converges, then
limit the resulting friction to what it should be, and solve the equations
of motion with the result. (This procedure doesn't lend itself well to
a verbal explanation, does it? Well, it was pretty fast, anyway.)
What the folks who sell simulation software are doing now are offering
"switch states." They solve *independent sets* of motion equations
depending on which way the contacting surfaces are moving. This is
a cleaner method than what we used to use. But you have to be able
to define your relative velocity as a switch state. I don't know if you
can
do that in Matlab without Simulink. But someone else here will. EASY5
is a program that has a standard component to do this -- and I can't
really imagine EASY5 having a leg up on both Matlab and Matrix_x --
so I'm guessing they can do it too.
>L-1359 Luxembourg
Oh. Never mind that first question. :-)
Dr. Ron Graham
Project Engineer for Robotics, GreyPilgrim LLC, Philadelphia
founder of sci.engr
EMMA Robotic Manipulator now online - http://www.greypilgrim.com/
Subject: Re: PE exam: Preparation
From: slester@molalla.net (Steven Estergreen)
Date: Sat, 05 Oct 1996 07:04:13 GMT
gyanesh@osborne.engin.umich.edu (Gyanesh Gupta) wrote:
>hello,
>Is there anyone out there taking PE II exam in Oct.? If yes, I would
>like to hear from you. I have to seriously start preparing for the
>exam. Its only 3 weeks left! I do not have much success in finding a
>pattern for the exam or any guide lines to prepare for the exam. NCEES
>book is really hopeless! If you would like to communicate with me,
>please let me know. Please reply me at
>gyanesh@engin.umich.edu
>CC: usfmdc39@ibmmail.com
>Hope to hear from some one.
>GYanesh
If you don't know the subject matter and can't find time to cram,
don't bother to take the exam. The whole point is to demonstrate that:
1) you are a knowledgeable engineer who can be trusted to be thorough
on matters of public safety and knows when to admit you're not
qualified for a job
or 2) you have taken advantage of preparing for the exam as an
opportunity to brush up on more of "Mechanical Engineering" than was
required to complete the last project you worked on as an "Engineer in
Training."
If you can't afford to spend some time preparing for the exam, it
probably doesn't matter enough to you. Wait and take it when it's more
important to you.
It doesn't take more than the 70-80 hours recommended in the other
reply (I passed on the first try after that much preparation 9 years
after graduation from college, and the problems were not very similar
to my work experience), but you need to have learned the basics of all
the topics in school and have a systematic approach to problem solving
AND test-taking.
=============================
Steven Estergreen, MSME, PE
Mulino, OR
slester@molalla.net
PP,ASEL,Instrument,Tailwheel,Complex
My employers don't pay for my Internet account
and I don't care WHAT they think of the opinions
I express here!
Subject: Re: need to purchase adhesive for electronics appl.
From: Woody White
Date: Sat, 05 Oct 1996 15:39:24 -0700
Perhaps - a thick varity of cyanoacrylic ("super glue") adhesive....
_Woody_
lionsgate.com@lionsgate.com wrote:
>
> I need to purchase an adhesive within the next 2 weeks. Requirements are, 1)must not initiate
> corrosion, 2)contains no silicone, 3)non-conductive. The application is to secure (hold down)
> a small inductor on PCB for our production line. Can you supply this adhesive? or .... direct
> me to a source for further investigation ? I have used search engines to contact approx. 10
> on-line vendors, still waiting for feedback. This is second avenue.
>
> Thanks in advance
>
> date : Oct 4th
> name : Mike Shaw
> email : mike_shaw@argus.iceonline.bc.ca
> Canadian fax : (604) 436 - 2713
--
de Woody (WB4QXE) - woody.white@worldnet.att.net
Electron Microscopist/Microanalysist,
Ham radio "homebrewer", shade tree mechanic,
'90 Nissan 240SX, wish still had my Mcycle too!
http://www.geocities.com/capecanaveral/3722
.