dann2707 Posted January 16, 2014 Report Share Posted January 16, 2014 Just wondering if anyone has used it before as I need help with the angle of air flow! Thanks Quote Link to comment Share on other sites More sharing options...
forteh Posted January 16, 2014 Report Share Posted January 16, 2014 13° No experience myself, Dr Mashall might be your best. Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 Unfortunately I haven't used Ansys' CFD capabilities... It's not as simple as defining it as a vector is it? Like U velocity and V velocity? Edit: This is what Fluent's velocity inlet set up used to look like: Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 (edited) Ahh well i'm not sure if it's specific to Ansys but basically. I have a truck with a flow domain around it and one of the questions is asking to alter the air approach angle to 20degrees, so it's coming from the side as we are supposed to analyis side load and drag etc. I have changed the flow of the X velocity to Cos(20) and the Y velocity to sin(20) but i'm thinking this is going to bring the air coming down at an angle as in pointing downwards or upwards when really I want the air coming in at a side angle. Just wondering if I have my cos and my sins correct?! I'm basing my original pythag above on this webpage https://confluence.cornell.edu/display/SIMULATION/ANSYS+WB+-+Airfoil+-+All+Pages but I think thats altering the angle as said above, at a downwards/upwards angle which I don't want Does that make sense? Cheers! P.S - i'm using Fluent! Edited January 16, 2014 by dann2707 Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 How is the domain oriented? If x is streamwise, is y spanwise? Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 Assuming x is streamwise, and y is spanwise I get what you've done. i.e. if you wanted a 25m/s effective wind speed at a 20 degree angle to the direction of travel the x velocity would be 25cos20 and the y velocity would be 25sin20. The next question is whether you need to be travelling at a certain speed as that would then come into play. Quote Link to comment Share on other sites More sharing options...
stirlingpowers Posted January 16, 2014 Report Share Posted January 16, 2014 Ahh well i'm not sure if it's specific to Ansys but basically. I have a truck with a flow domain around it and one of the questions is asking to alter the air approach angle to 20degrees, so it's coming from the side as we are supposed to analyis side load and drag etc. I have changed the flow of the X velocity to Cos(20) and the Y velocity to sin(20) but i'm thinking this is going to bring the air coming down at an angle as in pointing downwards or upwards when really I want the air coming in at a side angle. Just wondering if I have my cos and my sins correct?! I'm basing my original pythag above on this webpage https://confluence.cornell.edu/display/SIMULATION/ANSYS+WB+-+Airfoil+-+All+Pages but I think thats altering the angle as said above, at a downwards/upwards angle which I don't want Does that make sense? Cheers! P.S - i'm using Fluent! Haven't used Ansys, but there should be some global graphics view with a coordinate system somewhere. If x is truck moving direction and y perpendicular and flat, so z is the vertical axis, then x-cosine and y-sine will give you a front/side wind direction. BTW: Use the right unit, sine and cosine are typically programmed to take rad as a unit, making the argument pi/9. 1 Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 Thanks for the replies!! This is a screenshot of the question highlighted on the left, so the speed is to be at 20m/s and on the right shows my model and what plane it is in And this shows the boundary conditions im to set Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 Ah so it's x and z. so y velocity should be 0 and z velocity should be 20sin20. Edit: 20 degree case: x=18.79m/s, y=0m/s, z=6.84m/s and for the 40 degree case: x=15.32m/s, y=0m/s, z=12.86m/s. Edit 2: In that last pic it looks like your inlet is at the back of the volume (behind the truck)... I'm it's not but just in case. 1 Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 The thing is my lecturer via email told me to leave Z at 0 however he didn't see what plane my domain was in so perhaps he's wrong? Not doubting you at all, just conflicting replies and as an engineer assuming in wrongggg!! haha So x and y should be left at 0 and Z should be 20sin20? thanks Dave Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 See my edits. Yeah without knowing what way round you'd created the domain he would've assumed you had z as the vertical axis and told you to leave z as 0 but because of your orientation it's y that needs to be left zero. 2 Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 Thank you so much! Running the calc now I feel terrible for the photoshop thread now though haha Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 Why? Danny deserved that and I still quite like the version you did with the pale green wheels! Hopefully it'll converge ok and give you some sensible numbers out. You should quickly be able to tell from the vectors whether it's doing what you expect. Some of that will depend on what the sides of the domain are defined as- obviously if they are 'walls' the inlet velocity will be forced to flow down the rest of the domain and won't give the desired results... Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 Still computing. My cpu is dire. With regardsto flow field analysis, would that be putting the view into contours and looking at the ccoloured chart and seeing where most of the force is generated? Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 If you can view a velocity vector field horizontally through the domain (x-z plane) somewhere within the height of the model you'll be able to see what happens at the boundaries. Looking at pressure contours won't necessarily make it obvious. I'm not 100% sure how it will deal with the inflow at 20 degrees or 40 degrees since the flow incident on the lorry will actually need to be coming through the side wall of the domain rather than the 'inlet'. It have been easier to make the domain a bit wider and yaw the model while keeping the inlet conditions the same... Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 16, 2014 Author Report Share Posted January 16, 2014 Ahh sorry to not elaborate further. It was just generally speaking, is flow field analysis looking at pressure force etc. Google images for that phrase leads me to believe it is but just wanted confirmation! Quote Link to comment Share on other sites More sharing options...
monkeyseemonkeydo Posted January 16, 2014 Report Share Posted January 16, 2014 Ah gotcha. Looks like they're after force coefficients etc. initially but beyond that velocity vector plots, surface pressure plots etc. well show areas of separated flow round the leading edge of the cab and how the wake changes with changing flow angle. If there are extra credits for going a little further think about adding radii to the trailing edge of the trailer or giving the top surface some curvature like some lorries or there to prove the effect small changes have on the wake and hence forces. You could probably even relate those back to improving fuel economy to give it some weight. Quote Link to comment Share on other sites More sharing options...
Ash-Kennard Posted January 18, 2014 Report Share Posted January 18, 2014 (edited) I did a coursework like this last year, but we were looking at whether tailgate down or up was more efficient on a pick up truck. Used StarCCM+ for it, so not sure if this applies entirely but make sure you look at your y+ values as boundary layers will be important here as the flow will by high after the stagnation point. and reflect your mesh so you can get a properly even flow pattern. Also the question was too small to read but are you sure he didn't mean the truck having 20 degrees of yaw? you may also benefit from setting the wind tunnel length to 3 bodies ahead and 5 bodies behind and maybe increasing the cross section of the tunnel. will require more cpu power, but that just means you can make a mesh independence study and really impress, i would go quite coarse on the mesh too, will be very intensive on cpu times otherwise Edited January 18, 2014 by Ash-Kennard Quote Link to comment Share on other sites More sharing options...
dann2707 Posted January 21, 2014 Author Report Share Posted January 21, 2014 Thanks forthat ash mate! When it says lift force does it mean upwards force or downwards. My number is positive? Quote Link to comment Share on other sites More sharing options...
Jake. Posted January 21, 2014 Report Share Posted January 21, 2014 I did a coursework like this last year, but we were looking at whether tailgate down or up was more efficient on a pick up truck. Used StarCCM+ for it, so not sure if this applies entirely but make sure you look at your y+ values as boundary layers will be important here as the flow will by high after the stagnation point. and reflect your mesh so you can get a properly even flow pattern. Which was more efficient? Quote Link to comment Share on other sites More sharing options...
Ash-Kennard Posted January 21, 2014 Report Share Posted January 21, 2014 Thanks forthat ash mate! When it says lift force does it mean upwards force or downwards. My number is positive? negative uplift is downforce. it also depends how you set it up i guess, ive z is vertcally positive then a positive lift number will be uplift not downforce. Which was more efficient? The investigation into Pickup truck aerodynamics has yielded varied results showing that many of the adaptations have minimal impact on the coefficient of drag but can have massively detrimental impacts on lift and pitching coefficients. However some of the modifications (Covered Curved Taper) can have very positive effects upon the drag. Significant improvements have been made where models have maintained the flow and separation has been controlled to produce a “clean” and properly formed wake. straight out of the conclusion, if you want pics ill do them later cba right now 1 Quote Link to comment Share on other sites More sharing options...
Jake. Posted January 21, 2014 Report Share Posted January 21, 2014 Pictures would be nice... Cheers haha! Quote Link to comment Share on other sites More sharing options...
Ash-Kennard Posted January 21, 2014 Report Share Posted January 21, 2014 Pictures would be nice... Cheers haha! this is bad this is better this was best cheers, ash Quote Link to comment Share on other sites More sharing options...
Jake. Posted January 21, 2014 Report Share Posted January 21, 2014 Ahh I see, that's pretty cool. I don't think I'm going to adapt my caddy to the shape of that though. Cheers for that.anyway Ash! Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.