If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.
Special Thanks To:
Aaron Holmes
Aaron Wheeler
Adam Horden
Alan Scrimgeour
Andre
Andrew Haynes
Anonymous000
asabase
Austin Weil
barney
Barry
Bert Hickman
Bill Kukowski
Blitzorn
Brandon Paradelas
Bruce Bowling
BubeeMike
Byong Park
Cesiumsponge
Chris F.
Chris Hooper
Corey Worthington
Derek Woodroffe
Dalus
Dan Strother
Daniel Davis
Daniel Uhrenholt
datasheetarchive
Dave Billington
Dave Marshall
David F.
Dennis Rogers
drelectrix
Dr. John Gudenas
Dr. Spark
E.TexasTesla
eastvoltresearch
Eirik Taylor
Erik Dyakov
Erlend^SE
Finn Hammer
Firebug24k
GalliumMan
Gary Peterson
George Slade
GhostNull
Gordon Mcknight
Graham Armitage
Grant
GreySoul
Henry H
IamSmooth
In memory of Leo Powning
Jacob Cash
James Howells
James Pawson
Jeff Greenfield
Jeff Thomas
Jesse Frost
Jim Mitchell
jlr134
Joe Mastroianni
John Forcina
John Oberg
John Willcutt
Jon Newcomb
klugesmith
Leslie Wright
Lutz Hoffman
Mads Barnkob
Martin King
Mats Karlsson
Matt Gibson
Matthew Guidry
mbd
Michael D'Angelo
Mikkel
mileswaldron
mister_rf
Neil Foster
Nick de Smith
Nick Soroka
nicklenorp
Nik
Norman Stanley
Patrick Coleman
Paul Brodie
Paul Jordan
Paul Montgomery
Ped
Peter Krogen
Peter Terren
PhilGood
Richard Feldman
Robert Bush
Royce Bailey
Scott Fusare
Scott Newman
smiffy
Stella
Steven Busic
Steve Conner
Steve Jones
Steve Ward
Sulaiman
Thomas Coyle
Thomas A. Wallace
Thomas W
Timo
Torch
Ulf Jonsson
vasil
Vaxian
vladi mazzilli
wastehl
Weston
William Kim
William N.
William Stehl
Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Registered Member #4659
Joined: Sun Apr 29 2012, 06:14PM
Location:
Posts: 158
So I was trying to simulate coilguns in FEMM, but I noticed some odd behavior. When I analyze a simulation, then go to the answer page, then use integration to determine the force on the projectile, I determined that the force result changes a lot based on the size and shape of the FEMM problem's boundaries.
When I made the FEMM boundaries tight around the coil and projectile (meaning that FEMM was only computing the field in that small area), the force on the projectile was almost twice as much than when I expanded the problem boundaries to encompass a lot of free space beyond the coil and projectile.
This means that the force integration results are based on one of two things - 1) The amount and shape of free space that the problem encloses 2) The density of mesh triangles in the projectile and coil
I'm thinking that #2 is the likely culprit, and furthermore, that I should make the boundaries tight so that the projectile and coil have more mesh triangles (that makes it more accurate, right?).
Has anyone else experienced this before? What's going on here?
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600
Yes, I find it's best to not use the default mesh size and make it smaller right around the bits you're analysing.
Also, I usually put a couple of circles around the thing I'm modelling, about twice as big as each other with progressively bigger mesh. Then I make a separate circle, completely away from the modelled area (and set the boundary to be linked to the outside circle around my model), and stick air in it (with quite big mesh).
This greatly reduces the distortions due to having a finite area, the flux will leak out of the model and then connect in the extra circle rather than having to set the boundary to zero potential, which will often give quite bad distortion.
Registered Member #4659
Joined: Sun Apr 29 2012, 06:14PM
Location:
Posts: 158
wrote ...
Yes, I find it's best to not use the default mesh size and make it smaller right around the bits you're analysing.
so it is better to have more triangles? This does slow down the simulation, though.
wrote ...
Also, I usually put a couple of circles around the thing I'm modelling, about twice as big as each other with progressively bigger mesh. Then I make a separate circle, completely away from the modelled area (and set the boundary to be linked to the outside circle around my model), and stick air in it (with quite big mesh).
This greatly reduces the distortions due to having a finite area, the flux will leak out of the model and then connect in the extra circle rather than having to set the boundary to zero potential, which will often give quite bad distortion.
that sounds interesting. Can you upload a FEMM file so I can look at it?
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600
Yes, only have small triangles where something particularly interesting is happening. It's worth putting extra boundaries in just to be able to make the mesh smaller, even though there's air both sides.
The analysis times are not usually bad at all in fact if you're careful, just a few seconds.
It also helps enormously to avoid using non linear things like iron in the analysis, if you're not saturating, replace it with a linear or hard material with otherwise similar properties.
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600
For the boundary trick, check out diagrams A.8 and A.9 in the appendices of the FEMM manual, under the heading 'Kelvin transformation'. The general problem you're analysing is an 'open boundary problem'.
Registered Member #2939
Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615
You can use an iterative process to refine your mesh. Run a coarse mesh, then look for those areas with big field gradients - shrink the mesh size in those areas. repeat until you are happy with the results. For boundaries, I usually just make the boundary about five times the size of the area of interest: ie keep the boundary a long way from the problem to minimize its influence.
Registered Member #2529
Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600
2Spoons wrote ...
For boundaries, I usually just make the boundary about five times the size of the area of interest: ie keep the boundary a long way from the problem to minimize its influence.
Yeah, the manual describes that as 'crude'; if you do the trick in the appendix you can make the calculated area a lot smaller, and it's a lot faster to calculate as well, and more accurate.
If you're about to do more analyses, run, do not walk to the appendix of the manual. Do it now!
Registered Member #4659
Joined: Sun Apr 29 2012, 06:14PM
Location:
Posts: 158
for simulating coilguns, I did a test setup with a huge far away boundary. Then I manually set the mesh size for the iron. What i found is that the mesh size in the iron has little effect on the results of force calculations - there was less than half a percent of difference between mesh size .01 meters and .0001 meters.
Registered Member #2939
Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615
BigBad wrote ...
Yeah, the manual describes that as 'crude'; if you do the trick in the appendix you can make the calculated area a lot smaller, and it's a lot faster to calculate as well, and more accurate.
Oh, I agree with you. Its just that I'm usually doing quick sims to check an idea rather than going for super accuracy, and a giant boundary is fast to set up. What I would really like is full 3D fea - though the axisymmetric mode works pretty well for solenoids. I used to have access to Ansys, boy could that program bring a pc to its knees!
On meshing: rather than setting mesh size by area, you can get more efficient meshing by specifying mesh sizes on edges and/or points. That way you can get a fine mesh on a boundary while the interior of a region is fairly coarse.
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.
Odd behavior in FEMM simulations
