Jump to content

Andy@LamteqSolutions

Members
  • Content Count

    76
  • Joined

  • Last visited

1 Follower

About Andy@LamteqSolutions

  • Rank
    Advanced User

Profile Information

  • Are you University user?
    No

Recent Profile Visitors

1854 profile views
  1. Many thanks Adriano, Am typically used to creating windows using destroy .selection set f .selection; frame $f hm_framework addpanel $f "Tool Selection"; hm_framework drawpanel $f; Although I will give your version a go. Something which does bother me is the padding "pack $cb -side top -fill x -pady 2 -padx 4" Say I have 20 options and I use this code, a column of 6 tickboxes will appear in the window and the rest will simply not appear What I would like it to do is create columns. How do I get it to create several columns?
  2. Hi all, I'm trying to put a script together which uses tickboxes but can't seem to find the command to produce a tickbox. Anyone have any ideas? Cheers Andy
  3. Many thanks Adriano, For some of the work we do, we do look to stress in elements at a certain distance from a weld. Alas this block of work we look to the stresses actually within the weld elements themselves against proof and fatigue... which I suspect based on what you've written cannot use a contact tie against dissimilar mesh (bit of a shame because that would be a real game changer for us). I suspect going forward we may be able to use this approach in areas away from regions of interest. Interesting stuff. In the case above, yes the slave nodes are in the weld which is a finer mesh. Many thanks again Andy
  4. Hi all, I have a question about using contact ties to attached weld elements to a structure. In a nutshell, I've been asked to bring 2 models together with a dissimilar mesh and weld them. In the attached image on the left is a typical merged node to node approach with no contact - this would be our traditional approach. The middle contour plot is the same model but the weld is detached from the structure and is connected via a tied contact. The right contour plot is with a dissimilar mesh (similar to what I have been tasked with) and using the same contact tie set up as the middle contour. As you can see there is some variation between all three contour plots. Ideally I want to be able to use the approach in the right plot as this would cut out significant amounts of remeshing. However I do need to justify this approach. 1. Why is there such a difference between plot results? 2. Has this approach been used by others and if so why? 3. Is the contact tie approach acceptable for this application? 4. Are there any published references which can be used to support this approach? Any help is always appreciated, I don't always get a notification when people reply so apologies if there is a delay. Kindest regards Andy
  5. Many thanks for your reply Simon and sorry I didn't see this earlier. Yea, we're still having issues with this value... I'm getting 36800 using v = 3e6 mm/sec (TNT based number) p = 1.68e-9 Mg/mm^3 Sel = 10x10 = 100 mm^2 Gap = 38 We have a test model which implies we need to be looking StFac 360 (as 36500 was barely making the component deform).. but when we apply that to another model the damage is vastly beyond what we would expect.
  6. Hi all, I've been tasked with a problem and need to get my head around LaGrange-Eulerian contact. Overview. I have a plate structure positioned over an explosive. The Structure is mesh size 10mm. The explosive, air and ground are eulerian elements 25mm size. Problem When the explosive strikes the structure, one of two things occur. Either the volume fraction of explosive penetrates the plate and the plate deforms (I find this occurs with Stfval = 38500, gap = 38) or the volume fraction of explosive does not penetrate the plate but the plate remains largely undeformed with very small displacement (I find this occurs with Stfval > 100,000, gap = 38). Am assuming speed of sound to be 3000m/sec (best I could find for explosive). To my mind, the volume fraction should not penetrate the plate. Am a bit puzzled by the apparent lack of deflection in the structure when the explosive hits the structure. Output for the contact is giving me approximately the same force in all cases - although higher values of Stfval are giving higher forces (between 6.5-8e6N). There does seem to by much more contact energy in the lower Stfval runs. So I am tad confused. Any help would be much appreciated. Andy
  7. Hi, My pc is a 1 socket 4 core CPU (Intel(R) Core I7-6820HQ 2.7GHZ. Whilst running with (-checkel NO -out -core IN -np 4) I noticed that only 50% of my CPU was being used. Hyperthreading is off. I was just wondering, is there anyway I can get maximum CPU usage? Cheers in advance. Andy
  8. Hi Rahul, I thought I'd just quickly touch base on this one. I ran a simplified model and it actually worked. Really bizarre but it did examine more thicknesses before settling on a final design. Not sure what was going on in my initial model (perhaps I'm challenging it a little hard and it's accepting the first answer... bit odd). Thanks for all your help. Cheers Andy
  9. Hi Rahul, I managed to create a HSTX file however, I noticed that it contained the TPL file which is unfortunately commercially sensitive so I cannot share that file. However I have grabbed every screenshot I can. Failing this I can put together a simple model which should have the same issue (am assuming operator error on my part). Cheers, Andy
  10. Hi Rahul, How strange. I did open it in 2019 as a test but I don't believe I saved it in 2019. I've rerun it using 2017, this time I set it to minimise mass but the problem is still the same. Kindest regards Andy Study_1.hstudy
  11. Rahul, I was wondering if you could see where I am going wrong with my test. I've managed to set up hyperstudy to look at specific values and what I have is a plate with varying thicknesses on a structure with a loadcase applied. What I would like is the hyperstudy to minimise the volume without exceeding a stress value. I'm using Hyperstudy 2017.3 and have attached images from the response, objectives, constraints and the Evaluation Data. What the Hyperstudy seems to be doing is assessing 5mm thickness then 4.25mm thickness - which exceeds my stress limit but then stopping without examining 4.5 or 4.75mm. Do you see where I am going wrong? My stress limit is 341.3MPa. Many thanks in advance. Andy
  12. Hi all, This question may be reasonably straight forward to answer (it isn't a trick question at all). When I perform a hyperstudy optimisation study with plate thickness being the variable. Is it possible to dictate what thicknesses I want hyperstudy to investigate? For instance, I have a structure with a plate where I want to know what thickness I can reasonably use without going over a stress limit but I also know that the thicknesses available are 5, 10, 16, 20 mm. How do I get hyperstudy to only look at those values? Cheers Andy
×
×
  • Create New...