13Jul2009, 16:26  #1  
ChocChip
Join Date: Oct 2003
Posts: 2,575

A bit of maths help...
Ok so I'm not even sure if this is possible, but I'm sure one of the trickery faithful will be able to tell me pretty quickly.
In the attached diagram, I need to know whether its possible to work out the curve of C, without measuring it. I'm assuming that there is some posh mathematical formula that will work it out, but I'm not intelligent enough to know it. The only measurement I know for definite is the length of B and the width between A and B. Oh and the diagram is not to scale... Cheers
__________________
**** 


13Jul2009, 16:29  #2  
posts for a living
Join Date: Oct 2003
Location: Piss Stinking Penny Arcade
Posts: 4,361

is that sector on the end 90 degrees?
if so it's just angle (in radians) x radius (ie your 28mm) Last edited by Bilbo Fraggins; 13Jul2009 at 16:32. 


13Jul2009, 16:48  #4  
posts for a living
Join Date: Oct 2003
Location: Piss Stinking Penny Arcade
Posts: 4,361

Gonna have to use pi sooner or later



13Jul2009, 16:51  #5  
click here to reset post count
Join Date: Oct 2003
Posts: 5,534

youve got to make some serious assumptions to come up with any answer mathematically



13Jul2009, 16:56  #6  
WARNING: May contain nuts
Join Date: Oct 2003
Location: Prestonia
Posts: 11,531

Not strong on maths but doesn't the diameter times PI (3.14...etc) give the circumference, divide that by 4 and you have the length of C?
__________________
I arose gigantic and black, I howled at the night and the night howled back.



13Jul2009, 17:03  #7  
Madmontaholic
Join Date: Nov 2003
Posts: 1,524

I dont think its possible as you dont know at what point side B starts curving....
__________________
http://www.madmont.com 


13Jul2009, 17:28  #8  
GTFOMI
Join Date: Oct 2003
Posts: 5,420

if you can assume all the radii are tangental then its simple both of em are 28mm
a would b 201 b would be 201 28  28 c would be 28 to get curve length of c itd be (2*pi*r)/4 this is all assuming the radius are tangental to B and normal to A
__________________
<Blokefish> my bro said there was this person at uni who told everyone to call him "TBag" and who wore girls clothes all the time and hit on the townies 


13Jul2009, 17:46  #9  
Bit lost
Join Date: Oct 2003
Posts: 5,949

If you had the length of B it'd be a piece of piss, otherwise I don't see htf you can do it O_o



13Jul2009, 18:18  #10  
Roboplegic Wrongcock
Join Date: Oct 2003
Posts: 15,796

He asked for the curve, not the length of the curved bit. That's why i asked, to make sure that's what he meant. The curve would be given by a fiddly equation i can't be arsed trying to type out in this text box
__________________
Please use my link for ordering Giffgaff simcards! 

13Jul2009, 18:18  #11  
i fear no midget
Join Date: Feb 2007
Posts: 10,645

37
but 37 what i ain't saying 


13Jul2009, 18:30  #12  
res ipsa loquitur
Join Date: Oct 2003
Location: on the left
Posts: 4,823

The main assumption is whether the curve is circular or not. If not it gets quite a lot more complicated and would probably be best described as an elliptic curve. You can approximate its length by integrating the equation of the curve (thereby working out the area under the curve) and deducing its length through interpolation.
http://en.wikipedia.org/wiki/Elliptic_curve As a side point, elliptic curves were used to prove Fermat's Last Theorem. 


13Jul2009, 18:34  #13  
res ipsa loquitur
Join Date: Oct 2003
Location: on the left
Posts: 4,823

So actually, forget about it and just assume it's a circle



13Jul2009, 18:59  #14  
double post spacker
Join Date: Nov 2003
Location: Up North
Posts: 6,214

mathematical anti telharsic harfatum septomin lolz!!1
__________________
Has anyone really been far even as decided to use even go want to do look more like? http://www.chickipedia.com/ 

13Jul2009, 19:14  #15  
mooop
Join Date: Jan 2005
Location: Erm... My chair
Posts: 630

If you assume it's a circle then the curve would be something like:
X^2 + Y^2 = 784 If not, it's an ellipse and you are looking at something like: 784x^2/a^2 + Y^2/784 = 1 So no, unless you magically figure out 201 = B + 2D (D being the other length) then you aren't going to figure it out tbh. 


13Jul2009, 19:36  #16  
ChocChip
Join Date: Oct 2003
Posts: 2,575

Christ didn't realize that this would spark so much debate
Bit of background, this is a car part that I'm looking to get laser cut out of sheet metal. The company have asked for exact dimentions, obviously to chuck it straight in to their CAD machine. Although I have the original, its hard to tell whether it was designed as a proper circluar curve or not. I have had a play with a protractor and am seirously thinking that its not a proper curve at all It was designed for an italian car, so that probably explains it...! Cheers for all the answers, I will go back to them and suggest that we resort to the old tracing method.
__________________
**** 


13Jul2009, 19:49  #17  
pro poster
Join Date: Aug 2006
Location: Norwich
Posts: 712

Quote:
So basically if that curve is 90 degrees, as in the diagram below, then to determine the length of that curve use the Pi formula.. We'll say for arguments sake the diameter of the circle below (B) is 56mm "diameter" multiplied by "Pi" = circumference 56mm x 3.141592653 = 175.93mm divide that by 4 to get the measurment of that curve (A) = 43.98mm
__________________
http://www.flickr.com/photos/johnmuk BF1942/DC Server: East Coast Gamers BF and DCF 82.21.107.243 < Recently Changed 


13Jul2009, 20:25  #19  
There's a sale at Penneys
Join Date: Oct 2003
Location: yo mama's house
Posts: 2,341

42.
__________________
WATCH ME http://www.twitch.tv/w0rsty It took me Forty Three years to discover that I had no talent for Gaming, but I couldn't give it up because by that time I was too famous. 


13Jul2009, 23:00  #20  
pro poster
Join Date: Aug 2006
Location: Norwich
Posts: 712

__________________
http://www.flickr.com/photos/johnmuk BF1942/DC Server: East Coast Gamers BF and DCF 82.21.107.243 < Recently Changed 


13Jul2009, 23:06  #21  
i fear no midget
Join Date: Feb 2007
Posts: 10,645

Quote:
if you had have measured the other flat side, you would still be approximating unless you knew for sure the shape of the curve. stencil is your best bet and not bothering asking these q's again your second. if it's for an italian car it will not matter as they are quirky by nature and if it doesn't match perfectly the rest of the car will have rusted to fit before too long. 


13Jul2009, 23:25  #22  
Waterborne TrailerTrash
Join Date: Oct 2003
Location: Shropshire Union Canal
Posts: 13,059

__________________
And now we rise and we are everywhere 

14Jul2009, 10:14  #23  
PostPrandial
Join Date: Oct 2003
Location: just ootside Gatesheeed
Posts: 8,690

LOL was thinking that myself.
from the CAD stuff I've done there is no way to define that curve with the information provided you dont' even know where the curve starts on the baseline....or if it's centralised. not enough information but it doesn't look like a sector of a circle
__________________
Photo Portfolio Bits 'n Pieces : General Random Gallery : All new photoblog : My Flickr ........................I read it on Wikipedia...so it must be true!........................ 





Thread Tools  

