What's new

New Version of the 4 Link Calculator

Should be pretty easy. Side view max positive and negative deflection, relative to ride. And top view angle relative to axle tube. Top view shouldn't change, so just at ride. Should just be a little bit of trig. I'll add it to the to do list for v7.0.
 
Brief update. Finished: link end angles, add new material, add new rod end, convert inputs when switching units, and switch front X values when setting changes. Up next is save and load suspension, and then for the user interface.
 
And a second progress update this weekend. Save is done and tested. Load is setup, just need to do the tedious part of having the right saved value go to the right working value. Also need to do the select file portion. And save as needs the file name and location input as well but is otherwise done.
 
Load and save functionality is tested. Still need to do the select / choose file portion of them. Caught a few issues while testing them. I think I got them figured out; they were mostly data type and variable handling related.
 
I haven't used the calculator in years and when I did, the version I used didn't have the capability of articulating the axle in question to see the arc in which the links would travel as one side was fully compressed and the other side fully drooped.

I'm at a point designing a chassis on paper (I'm compooter illiterate so modeling is outside my wheelhouse) and I'm not sure if I need to narrow up the front frame rails to ensure the upper link mounts/captures at the axle end (XX 4 link) in the front clear the frame on articulation. The upper link mounts will clear the frame rails on both sides on full compression/bump with some amount of 'cushion' to allow for the link/capture/joint/bolt/hardware to still clear it during articulation should my chosen link geometry pull the tire and link inboard on compression. The question I need to answer is: how much 'cushion' do I need?

Any guidance is appreciated.
 
I haven't used the calculator in years and when I did, the version I used didn't have the capability of articulating the axle in question to see the arc in which the links would travel as one side was fully compressed and the other side fully drooped.

I'm at a point designing a chassis on paper (I'm compooter illiterate so modeling is outside my wheelhouse) and I'm not sure if I need to narrow up the front frame rails to ensure the upper link mounts/captures at the axle end (XX 4 link) in the front clear the frame on articulation. The upper link mounts will clear the frame rails on both sides on full compression/bump with some amount of 'cushion' to allow for the link/capture/joint/bolt/hardware to still clear it during articulation should my chosen link geometry pull the tire and link inboard on compression. The question I need to answer is: how much 'cushion' do I need?

Any guidance is appreciated.
The current version, v6 doesn't have articulation. V7, which is currently in development, will not either. It will show up in some manner in version 8. It may show up in a prototype form in V7.x depending on what sort of testing it needs, but probably not.

I currently don't have plans to release the full solver that I used to generate some plots for the 4 link bible. It is not written in free to use software, it takes a long time to run, and it would require lots of work to get to a state that is user friendly.

Unfortunately, I don't have anything on rules of thumb for how far in it goes.
 
The current version, v6 doesn't have articulation. V7, which is currently in development, will not either. It will show up in some manner in version 8. It may show up in a prototype form in V7.x depending on what sort of testing it needs, but probably not.

I currently don't have plans to release the full solver that I used to generate some plots for the 4 link bible. It is not written in free to use software, it takes a long time to run, and it would require lots of work to get to a state that is user friendly.

Unfortunately, I don't have anything on rules of thumb for how far in it goes.
I very much appreciate the response and am even more appreciative of the work that's already gone into the calculator as it is. Thank you.

IIRC, the geometry in my liberty buggy moves the tire outboard as it compresses up front....but I'm not sure how to get that to happen in this new chassis. Guess it's trial and error for those of us who can't model, lol.

Thanks again!
 
The chosen file/save location part is working. Now for the user interface. There's a reason that I put it off until last.
 
Update. Thanks to Austin for letting me use a color scheme similar to Irate4x4. Dark mode is much better than being flashbanged every time I open it. Link page is almost done. All that is left is a little bit of math checking, adding some what the data means labels to some of the plots, and to finish the changing of the input fields when loading a new suspension. :idea:Figured out a much easier and better way to do it as I was writing this post.

Hopefully the next few pages go faster. I had to learn and figure out a lot while doing this page. Shocks and maybe pitch should be the only two that take up a lot of time. Having the visuals working also meant I could start troubleshooting and correcting the math.

I have mixed feelings about the link page. On one hand, I got everything to fit on 1 page. On the other hand, it is rather crowded. It did require removing the plot legend from below the plots, but I think the solution of putting the color next to line selection should be a good alternative. Due to space, I am probably going to put the link end angles on the sizing page. I am very open to feedback on the interface and visuals.

1686278429080.png
 
Hi Treefrog.

I like the layout… have always been a fan of fitting as much relevant information on a single screen as possible as it provides more insights to the user. Its easier to observe the “cause and effect” for small changes in link position or angle. This is something that I really wanted to do when Dan Barcroft and I first built this calculator more than 20 years ago (really can’t believe its been that long already!) Back then the typical PC display was maybe 17” with very limited resolution. We couldn’t even get the front & rear suspension views to fit on a single screen without the whole thing looking like a crowded mess. Now, I’m sitting here in front of a 42” monitor with 4K resolution… having that kind of screen space changes everything. ;)

Good work here. Thanks for extending the capabilities of this calculator and making it even more powerful.

The only critique is that you need to correct “Role” center to “Roll” center in a few spots… that’s an easy fix.

Best Wishes,

-G
 
A couple more observations over my second cup of coffee...

Traditionally, the layout of the 4-Link Calculator has always put the "Front" of the vehicle on the left side of the screen, and "Rear" on the right.
This felt natural as being the "drivers side" view of the truck that we were originally designing for. In this V7.0 screenshot it appears that the rear suspension and details are on the left and front suspension details are on the right.

This also makes some of the scales on the graphs seem a bit odd... for example: the center views showing the vehicle in a profile view start measuring the wheelbase at 0" (rear wheel) and the actual wheelbase number is shown at the front wheel (at the right of the graph). While the number is accurate, it doesn't feel like the way that people actually conceptualize wheelbase. We think of a measurement from the front wheel center to the rear wheel center, not the other way around.

It seems like a relatively simple fix, as the two center columns of data remain in place and only the outer two columns (REAR and FRONT) get swapped visually on your screen layout.

The consistency in use of color seems good. We had always tried to make sure that colors assigned made sense (lower links were always red, upper links were always blue, panhard was always yellow). For a lot of the other calculated lines (roll centers, IC, AS) we tried to keep them in the same color families (greens, light blues, oranges) depending on what the attribute was. You have a lot of new visualizations now, so it's probably worth looking at how the colors used in all of those peripheral graphs associate with the color of the originally plotted line. I know that after a while you just run out of colors to choose from, but if you are using light green for roll center consistently in the calculator... it can be distracting to see that same color for something unrelated (like a pinion angle change graph).

A few things to consider as you continue to refine the V7.0 and as mentioned before, it does look very slick. :)


-G
 
A couple more observations over my second cup of coffee...

Traditionally, the layout of the 4-Link Calculator has always put the "Front" of the vehicle on the left side of the screen, and "Rear" on the right.
This felt natural as being the "drivers side" view of the truck that we were originally designing for. In this V7.0 screenshot it appears that the rear suspension and details are on the left and front suspension details are on the right.

This also makes some of the scales on the graphs seem a bit odd... for example: the center views showing the vehicle in a profile view start measuring the wheelbase at 0" (rear wheel) and the actual wheelbase number is shown at the front wheel (at the right of the graph). While the number is accurate, it doesn't feel like the way that people actually conceptualize wheelbase. We think of a measurement from the front wheel center to the rear wheel center, not the other way around.

It seems like a relatively simple fix, as the two center columns of data remain in place and only the outer two columns (REAR and FRONT) get swapped visually on your screen layout.

The consistency in use of color seems good. We had always tried to make sure that colors assigned made sense (lower links were always red, upper links were always blue, panhard was always yellow). For a lot of the other calculated lines (roll centers, IC, AS) we tried to keep them in the same color families (greens, light blues, oranges) depending on what the attribute was. You have a lot of new visualizations now, so it's probably worth looking at how the colors used in all of those peripheral graphs associate with the color of the originally plotted line. I know that after a while you just run out of colors to choose from, but if you are using light green for roll center consistently in the calculator... it can be distracting to see that same color for something unrelated (like a pinion angle change graph).

A few things to consider as you continue to refine the V7.0 and as mentioned before, it does look very slick. :)


-G
Thanks for the feedback.

Thanks for catching those spelling errors.

I'll take a look into the color scheme. It hadn't occurred to me to match plot and line colors. I never noticed that similar things were in the same color family. Thanks for the insight on the creation of the calculator.

Moving sections around is pretty straight forward. I don't mind doing so. But with how it's set up, positive X is going from the rear forward. It seems that is more intuitive than flipping everything around. But you've got me a bit confused. The earliest version I have been able to find only has the rear suspension and forward was to the right. When somebody on PBB added the front in v4.0, it was to the right, so I kept using that layout.
 
Thanks for the feedback.

Thanks for catching those spelling errors.

I'll take a look into the color scheme. It hadn't occurred to me to match plot and line colors. I never noticed that similar things were in the same color family. Thanks for the insight on the creation of the calculator.

Moving sections around is pretty straight forward. I don't mind doing so. But with how it's set up, positive X is going from the rear forward. It seems that is more intuitive than flipping everything around. But you've got me a bit confused. The earliest version I have been able to find only has the rear suspension and forward was to the right. When somebody on PBB added the front in v4.0, it was to the right, so I kept using that layout.


Hi Treefrog,

You are right! Looking back at some of the oldest versions of the calculator what you say is true. The development of the original 4-Link calculator started out when I was designing the rear suspension for my 1st Gen Blazer (Dan also owned a 1st Gen Blazer back then which is how we initially met - here is the event: BlazerFest 2001 in Hollister Hills, CA)

Dan Barcroft (aka Triaged) - Second from left in yellow 1st Gen
Greg Blanchette (aka Greg72) - Center in grey primer 1st Gen

5 First Gens2!.JPG


Fun trivia for those that never knew much of the original story.... :) For the first few months of its development the link calculator barely even had any graphical interface. It was literally just a page of numbers and calculations.

Here a screenshot of the very first 4-Link Calculator built on September 4th, 2003.

ExcelCAD-1 - 2023.09.04.jpg


In retrospect, it was just a matter of convenience at the time. We put the axle at 0" on the X-axis and then just started doing the geometry off that point using positive numeric values which caused all the links to be calculated and drawn over on the right-hand side. At the time, this was all just a "quick and dirty" exercise to speed up the process of figuring out a rear link suspension. Neither of us had any way of predicting what would happen to this little Excel worksheet over the next few months, years and decades....

When the 3-Link Calculator came out several months later, I used that one for my front suspension design, and it made a lot more sense visually as it was pointing the "correct" way (driver's side view) and since the rear suspension for my truck had already been finalized, it was easy to lose track of the fact that the 4-Link calculator was actually drawing everything backwards for most people.

As years went by, it always felt like it would have been great to combine both the front and rear suspension designs onto a single graphical worksheet, but the screen resolutions available at the time never allowed it to work in a practical way. You could either get the graphical image, or the input cells and the calculations.. but never both. It was just too small and hard to read... and jumping between worksheet tabs to view the rendering vs data would have been miserable to work with.

So, perhaps maybe it's not a simple matter of flipping the graphic around or relabeling a few cell values.... apologies for that. Ultimately, it feels like it would be a nicer looking layout if it were done that way but I'm not the one doing the work anymore. It's just some feedback to use at your own discretion.




-G
 
Last edited:
Hi Treefrog,

You are right! Looking back at some of the oldest versions of the calculator what you say is true. The development of the original 4-Link calculator started out when I was designing the rear suspension for my 1st Gen Blazer (Dan also owned a 1st Gen Blazer back then which is how we initially met - here is the event: BlazerFest 2001 in Hollister Hills, CA)

Dan Barcroft (aka Triaged) - Second from left in yellow 1st Gen
Greg Blanchette (aka Greg72) - Center in grey primer 1st Gen

5 First Gens2!.JPG


Fun trivia for those that never knew much of the original story.... :) For the first few months of its development the link calculator barely even had any graphical interface. It was literally just a page of numbers and calculations.
Thank you for the insight. I always wondered about how it came about.
Here a screenshot of the very first 4-Link Calculator built on September 4th, 2003.

ExcelCAD-1 - 2023.09.04.jpg


In retrospect, it was just a matter of convenience at the time. We put the axle at 0" on the X-axis and then just started doing the geometry off that point using positive numeric values which caused all the links to be calculated and drawn over on the right-hand side. At the time, this was all just a "quick and dirty" exercise to speed up the process of figuring out a rear link suspension.
Makes sense.
Neither of us had any way of predicting what would happen to this little Excel worksheet over the next few months, years and decades....
Funny how things can snowball.
When the 3-Link Calculator came out several months later, I used that one for my front suspension design, and it made a lot more sense visually as it was pointing the "correct" way (driver's side view) and since the rear suspension for my truck had already been finalized, it was easy to lose track of the fact that the 4-Link calculator was actually drawing everything backwards for most people.

As years went by, it always felt like it would have been great to combine both the front and rear suspension designs onto a single graphical worksheet, but the screen resolutions available at the time never allowed it to work in a practical way. You could either get the graphical image, or the input cells and the calculations.. but never both. It was just too small and hard to read... and jumping between worksheet tabs to view the rendering vs data would have been miserable to work with.
I'm still a bit worried about it all being readable. I've been working on larger screens and will have to check it on a smaller ones.
So, perhaps maybe it's not a simple matter of flipping the graphic around or relabeling a few cell values.... apologies for that. Ultimately, it feels like it would be a nicer looking layout if it were done that way but I'm not the one doing the work anymore. It's just some feedback to use at your own discretion.

-G
The nice thing is that it is just a matter of flipping the graphics. Flipping the signs inputs will take a bit longer, but that can be done if front left is better. I'll make a test screen and post it for comparison.

One reason I am hesitant to switch it besides it being different from the previous versions is that positive X being forward and positive z being up make sense from a driver standpoint and play well from a math standpoint. The math is easy to handle with an intermediate step, which is already done for the front.

Thanks for the feedback, I really appreciate it.

If you don't mind me asking, were there any features that you and Dan wanted to add back then? And would you be willing to share any of the early versions if you still have them? The earliest I have been able to find is 2.1
 
Thank you for the insight. I always wondered about how it came about.

Makes sense.

Funny how things can snowball.

I'm still a bit worried about it all being readable. I've been working on larger screens and will have to check it on a smaller ones.

The nice thing is that it is just a matter of flipping the graphics. Flipping the signs inputs will take a bit longer, but that can be done if front left is better. I'll make a test screen and post it for comparison.

One reason I am hesitant to switch it besides it being different from the previous versions is that positive X being forward and positive z being up make sense from a driver standpoint and play well from a math standpoint. The math is easy to handle with an intermediate step, which is already done for the front.

Thanks for the feedback, I really appreciate it.

If you don't mind me asking, were there any features that you and Dan wanted to add back then? And would you be willing to share any of the early versions if you still have them? The earliest I have been able to find is 2.1


Hi Treefrog,

Yeah, the internet stil felt pretty "new" back then and it was exciting connecting with people who shared a common hobby. This was back in the early days with Jason Payne (willyswanter) setting up online clubs for the So Cal Big Dawgs (where Dan lived) and the NorCal Big Dawgs in the Bay Area where I lived. The BlazerBash event may have first started in 2000... but I didn't have a my truck until 2001 so that was the first event that I attended. All the SoCal guys made that long drive all the way up to Hollister Hills to meet up with the guys who they'd only ever spoken to in online message forums... but it ended up being totally natural and awesome having everyone together. I kept in touch with Dan after that event and would bounce ideas off him. The catalyst for the 4-Link Calculator was an article by Fred Williams (4W&OR, July 2003)... where a lot of the "rules of thumb" were described.... (a couple of pages for historical context are inserted below)
scan0032.jpg


scan0035.jpg


It was really fascinating to read about the proportions of the upper links to the lower links, and all the ways you could figure out so much with a tape measure and a few formulas. So I basically dissected that article and put all of the math into a spreadsheet and sent it to Dan asking him what he thought. His response was something like: "Yeah, this is pretty good but you know.... there is a much better way to figure out some of this stuff". He was maybe in his sophomore year at CalPoly in this time period, but he was studying mechanical engineering so he had all sorts of cool formulas and ways to figure things out. He sent me some better formulas and suggestions and I kept messing around with it... but it wasn't until I sent him the .XLS with the side profile image of the tire, links and extrapolated lines showing AS and IC value traces that he got really excited and saw the potential of what our conversations could become. The rest, as they say, is history... :)

The idea of being able to cram a lot of data on a single page is really nice for guys with the large monitors and 4K displays... but you're right to consider that not everyone is going to be set up like that. Perhaps a simple option is to build out a few different "views" in the Excel worksheet tabs? They could be labelled as "MAX VIEW", "TYP. VIEW" and "CLASSIC VIEW" The max view would assume the user has lots of viewing space and would be laid-out with every possible graph available on a single sheet. The Typical View might have to remove some of the more esoteric items to allow for the most criical graphics to remain.. and a classic view could literally just be more of an OG-style layout with the link views and only the most critical of calculated values (AS%, Roll Center, IC, Link Lengths, etc) If you offer a few options in different tabs they really all just reference the same data so it's no extra effort in terms of the math... just filtered views of what has already been figured out.

The hardest question is "what to add next"? This ultimately is the one I'd struggle with the most. At the outset, my whole motivation for the 4-Link Calculator was to create something approachable and simple. I wasn't going to learn SolidWorks just to build a suspension model... and even that wouldn't help me understand whether it was a "good" design or a "bad" one. I think the biggest reason that people liked this calculator is because you only needed to fill in a few pieces of data in those light blue cells ("mess with these cells" as Dan liked to describe it) and you could generate a simple model showing link position, tire size and after that you could change a single value for a link position and watch all of the important attributes change... personally, being able to experiment and see which values could reduce an insanely high AS% value was the exact kind of insight that I needed to start wrapping my brain around general link suspension theory.

Obviously, as the user gets more comfortable and sophisticated, there are other attributes that they'd probably like to consider as well... but if you force the user to add dozens of required values before getting any kind of a useful model, it ruins a lot of the appeal. Especially difficult is when, as a new user, you don't even understand the interrelationships between those attributes so making guesses at values can really give you an end-result that is pretty inaccurate or useless.

Perhaps the best option is some sort of "greyed out" options that can be activated with checkboxes? Initially, the program opens up with very simple link design model, but once that is comfortably established the user can enable the spring/shock options checkbox and start filling in the relevant data there? It seems like giving a brand new user access to hundreds of options right away is a bit of a disservice and is just intimidating. Letting them ease into the program a little at a time and only make a few decisions at a time... build some confidence and understanding of the relationships between all of those link positions, angles and lengths. Then let them enable a few more features and play with some new features, and learn some more.

The most lofty goal might be some kind of feature that would examine actual designs and could highlight poor choices by changing cell shading from black (or white, depending on the background color of the worksheet) to yellow or red as the particular attribute moved further from an "ideal" or at least "acceptable" parameter....simplest example: A suspension design where the user has created a vehicle with 100% AS... but starts moving link positions and is cranking up the AS% to 200% or more.... if the program could use some simple conditional formatting for the AS% cell, it could slowly change that color from light yellow to brighter yellow to orange to red... indicating that the value probably isn't in an acceptable range anymore. The biggest problem is of course, who decides what a "good" value is... and if it is not declared anywhere in the program, the user may not even trust the feedback they are getting.

Of course, logically.... that's only a small step away from just having a button called "Optimize" and letting the program go through iteratively and try to make changes to remove any unusual values so that the entire design has been cleared of obvious design mistakes. Let the user set the optimization criteria used, or the specific items that are allowed to be changed.. certain other values could be "locked" to force the optimization to work around those non-negotiable design choices.

More than I was expecting to write when I first sat down here... but maybe some of that is interesting, or gets the discussion moving or just triggers a few new thoughts of your own.


Enjoy your weekend, thanks for listening.


-G
 
Last edited:
Hi Treefrog,

Yeah, the internet stil felt pretty "new" back then and it was exciting connecting with people who shared a common hobby. This was back in the early days with Jason Payne (willyswanter) setting up online clubs for the So Cal Big Dawgs (where Dan lived) and the NorCal Big Dawgs in the Bay Area where I lived. The BlazerBash event may have first started in 2000... but I didn't have a my truck until 2001 so that was the first event that I attended. All the SoCal guys made that long drive all the way up to Hollister Hills to meet up with the guys who they'd only ever spoken to in online message forums... but it ended up being totally natural and awesome having everyone together. I kept in touch with Dan after that event and would bounce ideas off him. The catalyst for the 4-Link Calculator was an article by Fred Williams (4W&OR, July 2003)... where a lot of the "rules of thumb" were described.... (a couple of pages for historical context are inserted below)

It was really fascinating to read about the proportions of the upper links to the lower links, and all the ways you could figure out so much with a tape measure and a few formulas. So I basically dissected that article and put all of the math into a spreadsheet and sent it to Dan asking him what he thought. His response was something like: "Yeah, this is pretty good but you know.... there is a much better way to figure out some of this stuff". He was maybe in his sophomore year at CalPoly in this time period, but he was studying mechanical engineering so he had all sorts of cool formulas and ways to figure things out. He sent me some better formulas and suggestions and I kept messing around with it... but it wasn't until I sent him the .XLS with the side profile image of the tire, links and extrapolated lines showing AS and IC value traces that he got really excited and saw the potential of what our conversations could become. The rest, as they say, is history... :)

The idea of being able to cram a lot of data on a single page is really nice for guys with the large monitors and 4K displays... but you're right to consider that not everyone is going to be set up like that. Perhaps a simple option is to build out a few different "views" in the Excel worksheet tabs? They could be labelled as "MAX VIEW", "TYP. VIEW" and "CLASSIC VIEW" The max view would assume the user has lots of viewing space and would be laid-out with every possible graph available on a single sheet. The Typical View might have to remove some of the more esoteric items to allow for the most criical graphics to remain.. and a classic view could literally just be more of an OG-style layout with the link views and only the most critical of calculated values (AS%, Roll Center, IC, Link Lengths, etc) If you offer a few options in different tabs they really all just reference the same data so it's no extra effort in terms of the math... just filtered views of what has already been figured out.
So, V7.0 is no longer in Excel. I am writing it in Python. That's why I am able to do stuff like completely rearrange the page quickly. I am aiming to have it be a standalone version that requires no other software. I think a smaller screen will be fine, its more limited on aspect ratio. But most things are 16:9. Being in Python makes simple view very doable, but I am not sure I want to. More on that below.
The hardest question is "what to add next"? This ultimately is the one I'd struggle with the most. At the outset, my whole motivation for the 4-Link Calculator was to create something approachable and simple. I wasn't going to learn SolidWorks just to build a suspension model... and even that wouldn't help me understand whether it was a "good" design or a "bad" one. I think the biggest reason that people liked this calculator is because you only needed to fill in a few pieces of data in those light blue cells ("mess with these cells" as Dan liked to describe it) and you could generate a simple model showing link position, tire size and after that you could change a single value for a link position and watch all of the important attributes change... personally, being able to experiment and see which values could reduce an insanely high AS% value was the exact kind of insight that I needed to start wrapping my brain around general link suspension theory.

Obviously, as the user gets more comfortable and sophisticated, there are other attributes that they'd probably like to consider as well... but if you force the user to add dozens of required values before getting any kind of a useful model, it ruins a lot of the appeal. Especially difficult is when, as a new user, you don't even understand the interrelationships between those attributes so making guesses at values can really give you an end-result that is pretty inaccurate or useless.

Perhaps the best option is some sort of "greyed out" options that can be activated with checkboxes? Initially, the program opens up with very simple link design model, but once that is comfortably established the user can enable the spring/shock options checkbox and start filling in the relevant data there? It seems like giving a brand new user access to hundreds of options right away is a bit of a disservice and is just intimidating. Letting them ease into the program a little at a time and only make a few decisions at a time... build some confidence and understanding of the relationships between all of those link positions, angles and lengths. Then let them enable a few more features and play with some new features, and learn some more.
I was thinking more along the lines of new pages. Like the driveshaft one that was a work in progress. I completely agree about not over complicating it. Using your shock example, it is on another page, so the user doesn't see it or use it unless they want to.

I do have some idea what is next, a 3d viewer with flex and more suspension types. I was curious if there was something that you had thought of all those years ago that never made it in.
The most lofty goal might be some kind of feature that would examine actual designs and could highlight poor choices by changing cell shading from black (or white, depending on the background color of the worksheet) to yellow or red as the particular attribute moved further from an "ideal" or at least "acceptable" parameter....simplest example: A suspension design where the user has created a vehicle with 100% AS... but starts moving link positions and is cranking up the AS% to 200% or more.... if the program could use some simple conditional formatting for the AS% cell, it could slowly change that color from light yellow to brighter yellow to orange to red... indicating that the value probably isn't in an acceptable range anymore. The biggest problem is of course, who decides what a "good" value is... and if it is not declared anywhere in the program, the user may not even trust the feedback the are getting.
It would be a cool feature, and would be pretty easy to add. Or even just having a gradient on the AS plot that gets redder the high it gets. I am hesitant with that for the exact reason you said, who or what decides what is good.
Of course, logically.... that's only a small step away from just having a button called "Optimize" and letting the program go through iteratively and try to make changes to remove any unusual values so that the entire design has been cleared of obvious design mistakes. Let the user set the optimization criteria used, or the specific items that are allowed to be changed.. certain other values could be "locked" to force the optimization to work around those non-negotiable design choices.
:stirthepot: That is a good label to put on that button... Maybe I'll have a new project before starting IFS and 3d viewing.
More than I was expecting to write when I first sat down here... but maybe some of that is interesting, or gets the discussion moving or just triggers a few new thoughts of your own.


Enjoy your weekend, thanks for listening.


-G
You too. Thank you so much for taking the time to respond. It is greatly appreciated.
 
I think switching it may cause confusion between the two versions. But I think I have a way to help make the front more obvious. I will give it a try tomorrow.
1686369088372.png
 
Hi Treefrog!

Typing this on my phone so this one won’t be my usual long-form response. ;)

I would be happy to share more of the really old historical information about the development of the 4-Link calculator so that you can put all of that into your revision notes area. If for no other reason, it shows the evolution of the design and gives a more accurate sense of the effort that was required all those years ago.

There is also a “wish list” of features that never made it into the worksheet… some because of effort required , some because of complexity and others due to the limitations of Excel itself. I will write all of that up so that you can evaluate if they belong in v7.0 or perhaps some version further down the road.

Weekend is here, time to get out into the garage and make more forward progress on the 1972 K5 Blazer. (Yes, I still have that SAME one even all these years later! It wears the link suspension that was designed using the original 3-Link and 4-link Calculators. It is a very cool ending to this long story). :)


-G
 
Last edited:
Almost through the outputs for the link sizing page. Just got done fixing some of the math for the rod ends. I think I still have a little debugging to do for the factors of safety, but I think the issue is in the force calculations, not the factor of safety calculations.

YeeP Since you asked for the feature, I'll throw this question to you. Regarding the rod ends at the axle's direction, it makes sense that up travel is positive angle at the frame, but the sign at the axle is less straight forward. Currently, it is set up to use angled up as positive. This makes it display as shown below with it not lining up with the frame side. There are 2 other options:
  1. Having both up travels with the signs meaning positive is up travel
  2. Having both up travels on the right with the sign indicating sloping up or down as it currently does
1689473877004.png
 
Just got done with the Link sizing page. I do need to make the pages for adding materials and rod ends still. I also need to fix the force calculations for the links. The tables in the center will grow if more lines are added. I may need to make those scrolling tables. Added materials and rod ends are saved to the lists of materials and rod ends, so they are remembered between different vehicles. Panhard inputs, outputs, and other link wall thicknesses have their text darkened or hidden if they are inactive. Just like the Excel version, the factors of safety change background color if they are outside the desired range.

1689999524356.png


I also updated the link calculator page some. I changed some plot colors, added a generic vehicle to the graph that shows travel, and changed the plot setting background color width. The vehicle changes size as needed. It can look a bit funky if significantly smaller tires are used in the front.
1689999673656.png

I am thinking about putting this out as an alpha soon to try compiling to an executable; I think I can do PC and Mac. But first, I need to do the settings page and want to do the about and driveshaft pages. The shock page and maybe the pitch page will take a bit of time with the amount of stuff they have. The good news is that with suspensions not being tied to the calculator, any that are done in a pre-release version don't have to be reentered in the full release.
 
I'm excited to see this moving away from Excel since I am not willing to pay Microsoft's subscription price for office. Looking good as always Treefrog :beer:

Greg72 thanks for all the historical insight! The old Petersen's article is awesome and makes me nostalgic for when there was actually good tech in Magazines :frown:
 
Another quick update. The driveshaft page is made. There seems to be an issue with some of the math and the pinion's offset from centerline is being ignored in the calculations it seems. Problem for tomorrow.
1690602541772.png
 
Quick update. The about page is done with the exception of adding the date of upload to Version 7.alpha's revision history entry. Settings page is also done. I also checked to make sure that changing settings works correctly. For example, it auto updates values when switching units systems.

I am currently working through checking the loading and saving. Had to fix some things. Imperial works as it should. Metric loading is causing issues, but I am working through them.

After that is fixed, I have one or two tweaks I want to make. Then troubleshoot the link force math.

With that as the to dos, I expect that the V7.alpha will be out very soon. Hopefully within a week.

There are some limitations that I need to work through before the full release. The biggest one is what happens when bad inputs are given. I intend to release the alpha version without input checking. A bad input will likely cause the calc to freeze/crash. The alpha also won't have the shocks or pitch pages. The goal of the alpha is so that people can give feedback and report on any issues they come across.
 
Well, the fixes and tweaks have been made. I have tried compiling it to a .exe and ran into a snag. Hopefully, I will work through that in the next few days.
 
Top Back Refresh