mikey_G - thanks for the bug report, I can see the problem and am looking into it now...
Update: I've re-uploaded the program now with, what I hope is, the bug fix. Basically I hadn't put code in place to prevent springs extending when negative load is applied, since in theory this can happen. Only for some reason it was getting completely out of control under certain circumstances that I can't really be bothered to look deeply into. I've just made added bump stops at the top of spring as well as at the bottom and it doesn't seem to crash anymore. Please let me know if that's cured the issue. I only noticed it on the UF1.
I encountered a serious error just now, when i selected "save as" and canceled it and then clicked "save" i got an error message i cant remember now but after that all my settings were gone?? Make sure you backup everything if you try this!
I have noticed that your live camber readings are incorrect.
I do not know if there are any other issues like this, but I found this one. See the images to see what the analyzer shows and what the LFS setup actually shows.
Suspension is (currently) only modelled as S1 style suspension, there is no dynamic camber with suspension movement, so the figures will not be right atm.
Once this is modelled, I can model grip loss with camber, and draw a balance diagram on the steering display.
I'm hoping I'll get around to this at some point over the summer.
another question in that style:
Spring forces calculation are made without motion ratio, isn't it? Are the springs stricly vertically modeled in the analyser?
At the moment, yes. All these things are on my big list of improvements to make over the summer.
Incidentally it would be great if anyone could give feedback on my manual, so that I can quote your comments in my project report so I have some of evidence that my manual is useful. Honest comments please.
OK I've just read the manual, it's exhaustive to learn how to use the analyzer but not how to create setups with it; however I think it's not the goal anyway..
Since this excellent tool could be used for other games and some car applications, I would like to know the LFS physics calculation limitations, like the linear undertray ground effect, aero calculation, and other physical interactions that are not calculated in LFS but present in the reality. And also the current analyzer calculations which are not yet coherent with LFS (like the suspension geometry as mentioned above)
A formula summary may be useful to know where the magic results come from, and for everyone car engineering education :nerd:
I think some technical words should be explained, a fresh noob may not know what "unsprung mass" is, and may be confused with the different meanings of "acceleration". The analyzer's manual should also invite the noob to study your excellent setup guide.
About your big list of improvement, what are they? :yummy:
Yes, the setup guide will continue to be a seperate document. The setup guide would not need to updated for every little change in the program, whereas the manual probably would.
I'm currently writing my project report, which explains ALL of the physics equations in my program. I will make this document available on my website for all to see (trust me, the manual is easy reading in comparison), although I will have to blur out some of the sections on tyre physics due to the LFS related information from Scawen that is not for public eyes.
My big list of improvements, as explained in the report:
When I re-write the setup guide (again, hopefully this summer), it will be written specifically to use this analyser (at the moment it mentions Colcob's suspension analyser and my GRC, but between them they don't quite cover everything, and are both outdated).
Bob, you're my savior... I've spent all evening mucking around with spreadsheets that just won't work all because I want to check spring frequencies, then I found this thread...
i was actually thinking about using something like in lfs
left click an arrow, value changes by .1, 1, or whatever
right click an arrow, value changes by 1, 10. whatever ten times the left click amount is.
right click the slider and you input a value.
that is a really good system. you might try fiddling with it some instead of adding more text boxes and such.
otherwise, great program.
looking foreword to figuring out how it works
I have a slight suggestion, only a minor thing really... I remember I spoke to you about this a while back. When opening a setup would it be possible for the program to remember the setup so that you can just save over it or modify it so that it doesn't replace your previous setup. Having to go and look for the setup I want to open or save amongst the hundreds of sets in my folder is very tedious!!!!
Madman - Well there is the save command that will save the current setup without you having to re-enter the filename (a setup needs to be open for that to work). I will add the ability to add the vehicle prefix and filter the sets accordingly when using save as. I will also add a recently used file list at some point. Will that be enough or will you like anything more?
Zebediah - I will add little plus and minus buttons next to each slider for sure, I'm unsure about text entry. The main problem I have, which LFS does not (really), is that this program supports configurable units. The sliders always work in SI units, so unless I convert the value you type in*, you'd have to type the number in SI units regardless of your display preferences
*this a real PITA, especially with text boxes in the form, as whenever you change the units, the values must be converted back to SI, converted to the new units, then fed back into the text box. With a pop-up entry box (a nice idea I admit) like LFS that would ease a lot of the horrible work that form text boxes involve. I'll play with the ideas over the summer and see what the best solution is.
actually I have been wondering quite a while.. say shouldn't increased rear rebound damping decrease corner-in oversteer instead of the other way around?
I'll admit to pinching the equations from Colcob's suspension spreadsheet so this app should work the same as his, I'm assuming everything is correct from there as people pulled it apart at the time.
It's also some time since I thought all this through, but if memory serves, it goes like this:
Any increase in damping results in increased roll stiffness for that end of the car when the dampers are in motion, ie. increase rear damping, rear has greater roll stiffness during transients.
Which ever end has greater roll stiffness will attract a higher proportion of the load transfer. Now due to tyre load sensitivity, a wheel pair with greater load transfer will have lower overall grip.
So basically, the stiffer end has less grip, you are probably familiar with this from basic spring tuning.
So at corner-in, you're inside rear damper will go into rebound, so if you increase the rebound damping, the roll stiffness will be greater at turn-in, taking away grip from the rear and increasing oversteer.
@Bob Smith: I downloaded and gave a glance to the app. You've done an astonishing job! I feel uneducated and kind of scared of the immense stuff in the program.
I liked that it worked well for me on a desktop with large fonts (120dpi). But unfortunately, all the window didn't fit in my 1152*864 screen. Part of the window on the right was just chopped.
I am concerning about the brake+initial turn in phrase, when the effect of roll is yet too dominant.
In my understanding, with more rear rebound, the car will have less vicious pitching effect under braking, less weight will be lifted from the rear, and less weight will be put on the front as well, so relatively the load distribution should be more even and the car should be less prone to oversteer when you turn in.
I think you might have explained what could happen in a turn-in situation without braking, when roll is the only factor, I think it is possible for a high rebound damping rate to hamper suspension movement to produce higher roll resistance, hence the oversteer. I'll have to ask my friend with actual car setup experience.
Say a car is braking at maximum strength for a good 50m. It will already have reached its maximum pitch for that rate of deceleration by that time despite any front compression and rear rebound damping - the damping merely slows this process down. Anyway, as turn in approaches a driver would lift the brakes slightly, thus decreasing the rate of deceleration. For this lower rate of deceleration, the car will tend to a lesser pitch, thus weight will begin to slowly shift towards the rear of the car again. The rate of this is limited by the front rebound and rear compression dampers. Where you're getting confused is in thinking that dampers limit weight transfer - this is incorrect, they limit the rate of weight transfer. (At least that's the way I understand the whole thing... but I'm fairly sure that's right)
I didn't confuse that, I realize higher damping means slower but not less weight transfer.
Your analysis of the situation is correct, however I think in real application you can't always use full braking power and not always brake in a straight line, depends on the flow of the track and so on. So whether "braking while turning in" means a situation where you are also inducing pitch, or simply inducing roll with your rear already lifted, would give very different outcome.
Seeing in the analyser that rear rebound damping has no effect with entry under braking, I guess it does only see it as the situation when the springs are fully loaded.
but then, why a corner entry with braking/loaded front tires, is more understeery then a corner entry without braking?
In related news, I handed in my project report today.
For the interested, or those seeking alternative cures to insomnia, I have uploaded the report for all to see, whether it be out of genuine interest, or for those looking for alternative cures to insomnia.
Note: I tried converting it to PDF (twice) but it just died (both times), so SWF it is for now. Unfortunately all the graphics are overcompressed and look a bit shitty, plus 3 or 4 images didn't load so have a lovely red x in a white box instead.
