Wheel overload measurements

poetempest

Well-Known Member
May 30, 2016
13
11
410
Just interested about weight and speed relations, and started investigations from max wheel load. Made a simple ultra-light tech with only cab in the center and venture frames for wheel mounting(2,2 gob). Loaded it with batteries for rough estimation and with gobs(GSO one blocks) for accurate. For big wheels i expanded both load platform to use GeoCorp Batteries and mounting frame to easily fit two-blocks-width wheels.

After each load increase i tried to move and if i see sparks i assume that wheel is overloaded and i start searching for the max load before sparks appear. It was trickier for large wheels cuz they don't like to spark. Than i counted load mass(2,2 gob base + 3*batteries + 24*geo batteries+gobs) and divided it by 4. As you can see, i don't take wheel weight into account(it may be a mistake).

The results are following:


  • gso
    • light trekker 3 gob
    • stabilizer 2,8 gob
    • dirt 6 gob
    • hub 5+
    • riser 7,8 gob(unexpectedly high)
  • geo
    • rugged 27+
    • workday 42+
    • jaws 48+
    • krusher 60+
    • big-boy 120+
  • venture
    • zoomer 5+
    • trotter 4+
    • landing 6,5 gob
    • springer 12+
values with + are inaccurate as max load, but fine to use.

As for summary i can say that if you want to load your tech consider using big-boy wheels. Four of them will let you fit 50+ megaton mortars=)

If you want to load a small vehicle(for example you want to carry 30 gob SCU on a small early game tech) just put more wheels inside and don't forget about stabilizers.

I gonna continue investigation in two directions, first i want to play with speedometer and different masses to understand how it works and second i want to find more reliable way to determine whether wheel is overloaded or not and by what amount.

overload.png large wheels.png SCU carrier.png
 

Olli_DXD

Räksmörgås
Dec 26, 2015
562
574
505
18
Linköping, Sweden
Just interested about weight and speed relations, and started investigations from max wheel load. Made a simple ultra-light tech with only cab in the center and venture frames for wheel mounting(2,2 gob). Loaded it with batteries for rough estimation and with gobs(GSO one blocks) for accurate. For big wheels i expanded both load platform to use GeoCorp Batteries and mounting frame to easily fit two-blocks-width wheels.

After each load increase i tried to move and if i see sparks i assume that wheel is overloaded and i start searching for the max load before sparks appear. It was trickier for large wheels cuz they don't like to spark. Than i counted load mass(2,2 gob base + 3*batteries + 24*geo batteries+gobs) and divided it by 4. As you can see, i don't take wheel weight into account(it may be a mistake).

The results are following:


  • gso
    • light trekker 3 gob
    • stabilizer 2,8 gob
    • dirt 6 gob
    • hub 5+
    • riser 7,8 gob(unexpectedly high)
  • geo
    • rugged 27+
    • workday 42+
    • jaws 48+
    • krusher 60+
    • big-boy 120+
  • venture
    • zoomer 5+
    • trotter 4+
    • landing 6,5 gob
    • springer 12+
values with + are inaccurate as max load, but fine to use.

As for summary i can say that if you want to load your tech consider using big-boy wheels. Four of them will let you fit 50+ megaton mortars=)

If you want to load a small vehicle(for example you want to carry 30 gob SCU on a small early game tech) just put more wheels inside and don't forget about stabilizers.

I gonna continue investigation in two directions, first i want to play with speedometer and different masses to understand how it works and second i want to find more reliable way to determine whether wheel is overloaded or not and by what amount.

View attachment 5986 View attachment 5987 View attachment 5988
Nice! This info could be very useful sometime later, but what does "gob" mean?
 

Andre Salvador Rainha

Well-Known Member
Sep 21, 2015
582
420
470
34
Portugal
SCIENCE!

Tough when choosing wheels you might want to use different wheels depending on the tech's goal. The BigBoy can carry lots, but you can place the double amount of krusher wheels and get the same mas load, more mobility and lower center of mass :p

Just a small tip.
Cheers
 

poetempest

Well-Known Member
May 30, 2016
13
11
410
Thanks=) infact, you don't get more mobility when stick two (EDIT: GeoCorp) wheels one to the other, because they have hard times to switch direction on their(very low) max speed. As for me, i use only two big boy wheels for all my high load(drilling stuff,megatons,geocorp batteries, SCU, power generator sorting plant, geocorp cab with armor) and mix it with three springer wheels in the back, after huge gap. Please note, that literally all mass is balanced only between big boy wheels, springer carry not more than 10 gob for all three. First of all it gives me just INSANE turning speed on one place, i can outtrack just any enemy tech by turning and shooting. Next, i have 50 mph unboosted speed(that's how 42 mph and 90 mph wheels stack). And finally i solved the problem of turning on max speed by adding venture little studs to the light(back) part of tech. It is the fastest large-scale tech i ever build and i use it as a general purpose tech=)

here is a screen http://i.imgur.com/ZxWPNoa.jpg

that was the idea from r&d chamber balance1.png
 
Last edited:
  • Like
Reactions: streak1

poetempest

Well-Known Member
May 30, 2016
13
11
410
oops o_O i suddenly found the wheel speed stacking mechanics=)
  • wut i did(experimental)
    • i took the tech i used to investigate overload and attached a speedometer to it.
    • than i tried to ride at maximum speed with minimum load(~2,5 gob) and measured max speed
    • after that i started to load the tech with batteries(small for venture wheels, geocorp for geocorp wheels) measuring max speed after each battery
    • I did it with 4 springer wheels, 4 big boys, 2 big boys and a mix of 2 big boys and 2 springers
  • wut i git(results)
    • speed function related to weight has a form of exponential decay. On large wheels it is approximated linearly quite OK.
    • overload state does not connected with speed function, it is an artificial restriction.
    • two wheels carry twice less weight than four=)))
  • wut i did than(analysis)
    • i took a geometric mean for each weight of approximation between 2 big boy wheels and 2 springer wheels(yes i had to assume that two springer wheels carry twice less weight than four) and.... the result curve matched experimental curve for 2 big boys and 2 springers. Wow.
here is a graph http://i.imgur.com/HKtCVsV.png

TL:DR - terratech has nice physics
 

ArmchairGravy

Well-Known Member
Aug 28, 2015
450
404
470
Thanks @poetempest for doing the work on getting these numbers. Some interesting results that may have me changing my designs a bit. I knew the GEO Rugged was bad, but wow, that's BAD.

I'm looking forward to @Payload releasing hard numbers for the blocks. I cannot properly min/max my tech in a completely OCD fashion until I get them!
 
  • Like
Reactions: poetempest

poetempest

Well-Known Member
May 30, 2016
13
11
410
Thanks for you attention=) continuing experiments, i gonna first proof the statement about "two wheels carry twice less weight than four". So i put 10 wheels instead of 4 in my springer wheel approximation and chose, for example, 51 gob weight and see - approximation give me 81 mph. so i go to RnD, make a dumb-simple tech meant to rige 81 .png and try....and i get 83 mph. 2,5% more than expected. So where the mistake? logarithm approximation gives me 3,7% divergence with experimental curve, what if i chose the most accurate approximation method(even if its against physics)? well quadratic polynomial approximation gives 0,5% divergence, what with results? its 84 mph. 1,25% mistake. Searching again, cubic polynomial approximation gives 0,15% divergence. And result is 83 mph, the exact experimental value. Wow.

for those who care,

y = -5*10^(-5)*x^3 + 0,0116*x^2 - 1,0487*x + 100,54

where y is speed and x is weight. It is for 4 wheels, for different number of wheels modify weight linearly(8 wheels - twice less weight)

So now we can at least predict speed and max load for at least any amount of springer wheels=)The next two steps are to get speed/weight curves for all wheels and to clarify the way of different wheel stacking(i gonna work only on second issue cuz i like only springer and big boy wheels).

TL : DR guys we need speed curves for all the wheels.
 

MidBlade

Well-Known Member
Apr 5, 2016
54
19
210
50
id try testing the Jaws an Krusher wheels as they seem to be able to carry more then the big boy wheels
 

poetempest

Well-Known Member
May 30, 2016
13
11
410
id try testing the Jaws an Krusher wheels as they seem to be able to carry more then the big boy wheels
will be grateful if you make a spreadsheet with tech max speeds matched with corresponding weight load, like this i did for 4 big boys:

speed load
43 2,45
42 50,45
41 122,45
40 218,45
39 314,45
38 386,45
37 458,45
36 506,45
35 626,45

Continuing experiments, i faced difficulties with prediction of wheel combining. so i chosen 14 springers and 6 big boys, and my approximation gave me 66 mph(for empty frame weighted 40 gob), but experiment gave 71 mph....7,5% percent divergence is inaccessibly high for science but quite OK for rough estimation.

Probably it just cant be accurate because i combine very very different wheel types and i have to start with easier combinations.

EDIT

i had to add weight distribution ratio to the speed combining formula. Now i get 69 from estimation, and the 3% mistake is much better. It can be a problem to count exact ratio(becuz its a tensor infact lol) but it still nice for rough estimation=)


EDIT 2

hahah lol experimenting with this lol wheels.png clearly shows that weight distribution DOES NOTHING to max speed. For overload i guess it does(from experince) but not for max speed.
 
Last edited:

Andre Salvador Rainha

Well-Known Member
Sep 21, 2015
582
420
470
34
Portugal
Welp...
This might not be the best time to gather data on speed tests as the wheel system is being worked to fix a couple of problems related to wheel/ground resistances. Weight tests might still be viable as the grip shouldn't interfere.

This also deppends on what version of the game you are running so... stuff.
Cheers
 
  • Like
Reactions: poetempest

poetempest

Well-Known Member
May 30, 2016
13
11
410
thanks for intel=)

as far as i understand, it never will be really proper time till release unfortunately...but right now I'm quite interested in the way my favorite wheels work=) just to play right now=) but what if they are not the best...O__o
 

Andre Salvador Rainha

Well-Known Member
Sep 21, 2015
582
420
470
34
Portugal
thanks for intel=)

as far as i understand, it never will be really proper time till release unfortunately...but right now I'm quite interested in the way my favorite wheels work=) just to play right now=) but what if they are not the best...O__o
It may not be the best time to run tests but its still fun to bug-hunt ;p
 

poetempest

Well-Known Member
May 30, 2016
13
11
410
It may not be the best time to run tests but its still fun to bug-hunt ;p
oh ye) there are so many bugs so one bugs help to workaround other bugs. For example sometimes when you try to pick SCU it disappears. It completely ruins the game, but one other bug(or probably feature) gives an opportunity to buy SCU for the cost of cab from payload terminal. Just attach SCU to cab and save it like that...scu1.png
I hope disappear bug will be fixed before SCU multiplication.