sail7 Activity
Sail7 is a design and optimisation tool for sail performance.
Status: Abandoned
Brought to you by:
techwinder
Activity for sail7
-
Jarred Swalwell posted a comment on discussion General Discussion
Is there any way to summarize the induced drag calculated by the model? The streamlines show the tip vortices, but it'd be helpful to have a number if possible. On Fri, Jul 12, 2019 at 10:25 PM "André" techwinder@users.sourceforge.net wrote: OK , will correct the wind gradient bug in the next few days. I needed to do some code cleaning anyway and to promote the application to v1.0 status. This will probably be the last version before it is merged in xflr5. jib/main angle in graphs, also streamlines...
-
André posted a comment on discussion General Discussion
Hi everyone, Following the release of flow5, I'm closing the sail7 project. André
-
Yes, the density and kinematic viscosity can be set in the analysis definition. André
-
Claire-Marie Vautier posted a comment on discussion General Discussion
Hi ! Thank you for your work on Flow5, it's amazing ! I'm wondering if there's a way to change the fluid properties ?
-
Hi everyone, This is to announce the release of the next version of sail7. This new version is in fact an almost new application, which merges the functionalities of both xflr5 and sail7 together with a bunch of new features. It has been widely re-written in modern code style, in a huge overhaul effort intended to optimize the stability and to facilitate the application's continuous development. To mark this change, it takes the new name of flow5. You can find more about it on https://flow5.tech,...
-
- demoting the version from 1.00 down to 0.07 due to unsoved memory
-
- code cleaning
-
- code cleaning
-
- code cleaning
-
- code cleaning
-
- restored the missing image files
-
- updates for macOS
-
updating metadata
-
- Restored the saving of wind gradient parameters in the polar
-
OK , will correct the wind gradient bug in the next few days. I needed to do some code cleaning anyway and to promote the application to v1.0 status. This will probably be the last version before it is merged in xflr5.
-
- code cleaning
-
Cleaning up repository trunk
-
deleting obsolete directories
-
- code cleaning
-
tagging version 0.05
-
I think Sail7 is fantastic. I only meant to point out the errors one should expect given how sails are supported and used in the real world. When sailmakers design sails, it's for an idealized shape. The controls other pull it back into that shape, or are used to reduce lift (de-power) in situations where less heel is needed. Most sailmakers, other than the big lofts who produce grand prix sails, have no budget or expertise for CFD base design. They use experience to determine draft, draft location...
-
Thanks for sharing this, I'm learning a lot about sail operation. If flow separation is more or less always present in sail operation, and is significant to a greater extent than it is for plane wings, then this raises the question of whether potential flow theory is of any use for sail analysis. Also, the fact that the sail takes an unknown shape under wind pressure and boom tightening, means that the rigid sail model using sections and angles may not be very relevant. Anyway, I'll merge sail7 into...
-
Thanks for the video link, I'll watch them. Yes, in theory and in practice the elliptical wing tip minimizes induced drag, for wings. But imagine welding a 4" pipe on the leading edge of a spitfire wing. This is what a sail is. The trailing edge (leech) can be elliptical. But there are some practical issues with the elliptical leech. Angle of attack is adjusted through a mainsheet attached to a boom that pulls on the clew of the sail. When the mainsheet is tightened, the leech tightens. But because...
-
Yes, that's the reason. Panel methods cannot predict viscous effect such as flow separation. I tried to explain why in these videos. Fathead = long gaff? I'm surprised that the intent of these shape is to minimize the induced drag. The shape which does that is the elliptical wing e.g. the Spitfire's wing which ends with a zero length chord. If there is a reason why it should be different for sails, it escapes me.
-
Oh. Think I answered my own question by reading up on the limitations of VLM. Turbulence is not resolved so seperation would not be realized. Sorry, my background in fluids is acoustics, sailing is a hobby. I am looking at a comparison of fathead vs. pinhead main sail in a fractional rig. Induced drag should be calculated with VLM, but it is odd that the fathead main is not demonstrating a reduction in induced drag in the analysis; less induced drag is really the aerodynamic purpose of using a fathead...
-
One more useful parameter would be Lift/Drag ratio that could be plotted against other variables. As for analysis, I've plugged in typical jib and mainsails and fixed them to designed sheeting angles. I then varied Beta (wind direction) and in plotting lift/drag there is no inflection. Lift just continues to increase. I would expect seperation to kill off the lift as drag increases significantly. Am I missing something?
-
One more useful parameter would be Lift/Drag ratio that could be plotted against Ctrl. As for analysis, I've plugged in typical jib and mainsails and fixed them to designed sheeting angles. I then varied Beta (wind direction) and in plotting lift/drag there is no inflection. Lift just continues to increase. I would expect seperation to kill off the lift as drag increases significantly. Am I missing something?
-
OK, got that. Will implement this in the next version. I'll post some news about it in a few weeks.
-
Hi. This is really great, thanks for your efforts. As a first order trim in sailing upwind it's typical to sheet the main so the boom is centered and then trim the headsail close to stall. Two things that would determine the trim angles around which to optimize sails are the streamlines ahead of the sail plan and the ability to plot the angle of a sail against lift or drag. Thanks!
-
Cris Bolle posted a comment on discussion General Discussion
OK, thank you! Should I post any bugs I find here or somewhere else?
-
I'm afraid that I'm no longer able to develop sail7, so you'll have to do some testing. Sail7 will be part of the xflr5v7, so all new features and bug fixes will be introduced as part of the new app.
-
I could increase the wind speed by 10x, that would increase the forces by 100x right? Since it's a Potential Flow, the wind speed is just a proportional factor, it doesn't affect the nature of the solution.
-
I see that I can change the units of length and weight to something that is better suited to a model boat. But the calculated forces and torques are still in N and Nm. So I end up getting a 0.004N lift. Is there a way to change this or get the number in an exponential format? My other option is to scale the model by 10x. Thank you!
-
Since the last post, sail7 has been merged into the upcoming version of xflr5, which hopefully will be released in a few months. The goal is to minimize the future development effort because the two applications share the analysis engine and a good part of the GUI. There is no plan to make a web-based app. Also as mentioned above, and given the amount of work involved, I'm not sure I will open-source the code just yet. The problem of the sail definition is still very much open. As it stands, I've...
-
Keith Gray posted a comment on discussion General Discussion
I know it's been a while since anyone has mentioned on this thread. But we're profesional sailmakers that believe in the open source concepts. The various sail design software systems tend to have their own file structure, but there is one file format that is 'standardized' for sending designs between loft networks. It also happens to be an open text file describing the panels that are used to construct the sail. I know the current app is in win32, but for some help porting it to a web delivered...
-
KSP modified a comment on discussion General Discussion
-
The best way to have proper angles in the Start, End and Delta boxes is set into Polars > Define an Analysis > Design Variable section values for min = 0 and max = 1
-
The best way to have proper angles in the Start, End and Delta boxes is set into Polars > Define an Analysis > Design Variable section values for min = 0 and max = 1
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
It seems that problem is not in the trailing edge... If this were so, then high aspect ratio surfaces should have relatively greater influence from long trailing edge than low AR surfaces, but my last experiment show exactly opposite behaviour! Now I took flat plates for simplicity - one with high aspect ratio and second with low. As a result: twin High AR plates increased lift just for 5%, but Low AR plates increased lift for more than 50%!!! Maybe problem hides in the tip vortices? They are huge...
-
The only interaction between sails which can cause numerical issues is when the wake shedded in the wind direction from the front sail's trailing edge intersects the rear sail. Doesn't seem to be the case in you images, but it's hard to tell. Other than that, seems that you are getting the correct solution to the given problem.
-
André, the problem is not in adjacent surfaces, I did it just for clear view. Problem is in abnormal increasing in lift and drag for sequential surfaces, no matter is there gap or not between them. Firstly I noticed it with normal jib and main sails when compared them with one big sail with the same sail area. So I did this example for understanding what happens. Below is shots made separately for front and rear sails from previous post. We can see that front sail taken alone is mostly yellow and...
-
André, the problem is not in adjacent surfaces, I did it just for clear view. Problem is in abnormal increasing in lift and drag for sequential surfaces, no matter is there gap or not between them. Firstly I noticed it with normal jib and main sails when compared them with one big sail with the same sail area. So I did this example for understanding what happens. Below is shots made separately for front and rear sails from previous post. We can see that front sail taken alone is mostly yellow and...
-
André, the problem is not in adjacent surfaces, I did it just for clear view. Problem is in abnormal increasing in lift and drag for sequential surfaces, no matter is there gap or not between them. Firstly I noticed it with normal jib and main sails when compared them with one big sail with the same sail area. So I did this example for understanding what happens. Below is shots made separately for front and rear sails from previous post. We can see that front sail taken alone is mostly yellow and...
-
André, the problem is not in adjacent surfaces, I did it just for clear view. Problem is in abnormal increasing in lift and drag for sequentional surfaces, no matter is there gap or not between them. Firstly I noticed it with normal jib and main sails when compared them with one big sail with the same sail area. So I did this example for understanding what happens. Below is shots made separately for front and rear sails from previous post. We can see that front sail taken alone is mostly yellow and...
-
André, the problem is not in adjacent surfaces, I did it just for clear view. Problem is in abnormal increasing in lift and drag for sequentional surfaces, no matter is there gap or not between them. Firstly I noticed it with normal jib and main sails when compared them with one big sail with the same sail area. So I did this example for understanding what happens. Below is shots made separately for front and rear sails from previous post. We can see that front sail taken alone is mostly yellow and...
-
There is a technical reason behind this: the pressure distribution is linked to the gradient of the doublet density on a surface. When two surfaces are adjacent, the program has no way of knowing this, therefore the shared edge is seen as a free boundary for each surface, and this causes a discontinuity in gradient and pressure. The streamlines should be the same though, as long as the panels shape and number are the same for the two configurations - which does not appear to be the case in your example....
-
André, it seems that something strange happens with calculations of sequential sails. In this example I made one rectangular sail 16x8 with camber 10% and second set from two sails 16x4, joined one to other without gap and fully repeated shape of the one sail. It was expected that forces would be equal or at least with minimal difference, but we see that in the case of two sails lift is higher at 13% and drag at 4.5%. Also we see that rear sail have almost the same pressure distribution as one sail,...
-
André, it seems that something strange happens with calculations of sequential sails. In this example I made one rectangular sail 16x8 with camber 10% and second set from two sails 16x4, joined one to other without gap and fully repeated shape of the one sail. It was expected that forces would be equal or at least with minimal difference, but we see that in the case of two sails lift is higher at 13% and drag at 4.5%. Also we see that rear sail have almost the same pressure distribution as one sail,...
-
André, it seems that something strange happens with calculations of sequential sails. In this example I made one rectangular sail 16x8 with camber 10% and second set from two sails 16x4, joined one to other without gap and fully repeated shape of the one sail. It was expected that forces would be equal or at least with minimal difference, but we see that in the case of two sails lift is higher at 13% and drag at 4.5%. Also we see that rear sail have almost the same pressure distribution as one sail,...
-
André, it seems that something strange happens with calculations of sequential sails. In this example I made one rectangular sail 16x8 with camber 10% and second set from two sails 16x4, joined one to other without gap and fully repeated shape of the one sail. It was expected that forces would be equal or at least with minimal difference, but we see that in the case of two sails lift is higher at 13% and drag at 4.5%. Also we see that rear sail have almost the same pressure distribution as one sail,...
-
André, it seems that something strange happens with calculations of sequential sails. In this example I made one rectangular sail 16x8 with camber 10% and second set from two sails 16x4, joined one to other without gap and fully repeated shape of the one sail. It was expected that forces would be equal or at least with minimal difference, but we see that in the case of two sails lift is higher at 13% and drag at 4.5%. Also we see that rear sail have almost the same pressure distribution as one sail,...
-
Thanks for the feedback Antonio. As you may have understood I'm a poor sailor - my sailing experience is limited to windsurfing. So all comments and suggestions are very welcome. The things I plan to do when I resume the development of Sail7, presumably in the second half of this year: - implement import/export capability of the hull and sails from/to STEP and IGES files - implement all the GUI and analysis improvements that have already been introduced in xflr5 since sail7 was released - implement...
-
Thanks for the feedback Antonio. As you may have understood I'm a pool sailor - my sailing experience is limited to windsurfing. So all comments and suggestions are very welcome. The things I plan to do when I resume the development of Sail7, presumably in the second half of this year: - implement import/export capability of the hull and sails from/to STEP and IGES files - implement all the GUI and analysis improvements that have already been introduced in xflr5 since sail7 was released - implement...
-
Antonio Campos posted a comment on discussion General Discussion
After using Sail7 a bit, and after some thought about it, I think it is an issue that you have to manually define Phi (bank angle or lateral tilt of the sailboat) in your polars. In a normal sailboat, Phi is "automatically" defined, when the lateral moment (My in Sail7) becomes equal to the lateral moment of the weight of the ship. In the RC model ships I am interested in (mainly IOM class), this is very much driven by the keel (the bulb weight and its arm that define the moment), and at a lower...
-
Ok, thank you ! I did not know that was possible, because there is no manual. If I can help with that when you do the revision, I volonteer. I am spanish-speaking and fluent in french and english. Merci !
-
I understand. I was interested in controlling the vortex that is creatred by the wind going betweein the sails and the hull, and streamlines were interesting to understand it. Thanks for the answer.
-
What if you duplicate the boat?
-
This is not really an error, but just a consequence of the discrete way that streamlines are calculated. The streamline is initiated at the trailing edge, the velocity vector is calculated, a segment is drawn in the direction of the vector, and so forth. At some point downsteam, the segment length may be such that it goes through another sail. Not much to do about it. The streamlines display is there more for information than anything else.
-
This probably should have been posted here instead of another part of the forum, sorry about that: I see you are updating the code, hope to see what's new! I would like to report a possible error: I have found a way to define a very large horizontal NURBS sail, so I could simulate the water surface. Of course that fake sail surface tilts when you define a bank angle for the boat in the polar menu, so I have to "tune" the surface so as to compensate the bank angle, and keep it horizontal after running...
-
I mean that you can compare two polars of the same boat, but if you modify the boat, you cannot compare the result (modified boat vs. original version). To compare modifications on the hull, I have to export to Excel and re-do the graphics there.
-
Code cleaning
-
Hi Antonio, Not sure that I understand the question, because the comparison of polars is precisely what the polar view is intended for. Use the F8 key or click on the litle graph icon in the top toolbar? André
-
I have been playing a lot with Sail7 in the last months. One feature that I find missing is the ability to compare two setups: what happens if I modify one sail? what happens if I modify the hull?... : the best way to figure it out is to be able to see the new polar vs. the old polar in the same graphic ideally, or at least in the same page. OpPoints cover only playing with the parameters, but not playing with the shape of the boat, hull, sails, etc. The solution I have found is to export to excell,...
-
I have been playing a lot with Sail7 in the last months. One feature that I find missing is the ability to compare two setups: what happens if I modify one sail? what happens if I modify the hull?... : the best way to figure it out is to be able to see the new polar vs. the old polar in the same graphic ideally, or at least in the same page. OpPoints cover only playing with the parameters, but not playing with the shape of the boat, hull, sails, etc. The solution I have found is to export to excell,...
-
I have been playing a lot with Sail7 in the last months. One feature that I find missing is the ability to compare two setups: what happens if I modify one sail? what happens if I modify the hull?... : the best way to figure it out is to be able to see the new polar vs. the old polar in the same graphic ideally, or at least in the same page. OpPoints cover only playing with the parameters, but not playing with the shape of the boat, hull, sails, etc. The solution I have found is to export to excell,...
-
The subroutine to create hull parts allows to create a horizontal sea surface (even if it was not designed for that), and you can even add small waves that way, but if when you bank, phi angle affects the surface as the rest of the hull. This is annoying, because most of the analysis to make is when climbing against the wind on close haul, which always involves some tilting. Maybe a fast way to handle the issue would be to allow to create surfaces that don't tilt with the rest of hull and sails.
-
Hi Antonio, I'm taking note of the suggestions. At this point I don't have a solution to propose for the modelling of the water surface, but I plan to resume the work on sail7 at some point this year, as part of an expanded xflr5. I'll post back then. André
-
First, André, thans for this program, it is very intersting. I have been playing around for a few weeks, and I think the most annoying thing to correct is that there is no "water" surface, when you display streamlines in the lower deck area, the air clearly goes under the hull. I have thought wether I could "patch" this defect by creating a fake flat sail, as a large horizontal surface air cannot pass through, but I could not do it with the program. It would also be interesting that there is some...
-
Hi Jaap, Yes, there 's no easy way to define a twist distribution along the trailing edge. This brings us back to the main issue being that there seems to be no easy way to define a sail at all. It could be possible to do it the way a wing is defined in xflr5, but I don't know if this makes sense. It would require some feedback from sail designers and manufacturers. And yes technically it is possible to add features such as the display of drag and lift, but I don't intend to continue the development...
-
Jaap van der Heide posted a comment on discussion General Discussion
Dear André, Lately I have been playing around with the ideas presented in the NASA PRANDTL-D glider project with regards to limiting wing tip vortices and with that perhaps induced drag in rig/sail design. The NASA project employs unlineair twist distribution, which as far as I can tell cannot be done in Sail7, right? It would be nice IMHO if that could be done. BTW should it be possible to show a "spanwise" lift distribution curve and/or induced drag distribution curve in/from Sail7, much like it...
-
Hi everyone, In view of a long-term future update of sail7, I'm opening this thread...
-
sheldon haynie posted a comment on discussion General Discussion
Andre, in general the old symmetric spinnakers were more for drag than lift, modern...
-
Hi Sheldon, I have only a limited experience of sailing, but my understanding is...
-
Looking at Staysails that may not tack on centerline. Can a Y dimension be provided...
-
Thanks, that answers the question on the parameterization. Is there any way to add...
-
Ok, found the import as a subfunction of the edit. Now trying to parameterize an...
-
Not yet successful importing my boat/rig/sails into Sail7. It's a Yawl and with jib,...
-
Not yet successful importing my boat/rig/sails into Sai7. It's a Yawl and with jib,...
-
Once again, this is just an application of the VLM, with its limitations. I'm reasonably...
-
Poul Polli posted a comment on discussion General Discussion
My problem is that a simulation with sail7 and 2 sails provide a Lift lower than...
-
To put it simply, I don't know the answer to that question. The potential solution's...
-
Hi, i’ve got some questions about sail7. 1) It seems that sail7 compute all surfaces...
-
Hi, i’ve got some questions about sail7. 1) It seems that sail7 compute all surfaces...
-
Hi, i’ve got some questions about sail7. 1) It seems that sail7 compute all surfaces...
-
Hi, i’ve got some questions about sail7. 1) It seems that sail7 compute all surfaces...
-
There is no "sequentiality" in steady state panel analysis. The problem is soved...
-
but the error of considering all lifting surface simultaneously is that if i've got...
-
It's a raw potential solution, no boundary layer nor flow separation effects.
-
do sail7 take account of boundery layer and flow separation with some coefficients,...
-
Yes, it does. The VLM considers all lifting surfaces simultaneously and solves the...
-
hi, it seems that sail7 don't consider sails interaction, how do you solved this...
-
le pluart posted a comment on discussion General Discussion
or is it possible to design a sail directly in sail 7 ? thanks!
-
hi, i try to make a mainsail but there is a bug when i import it in saildef. . luff...
-
Corrected the bug which blocked user input in t...
1 >
