Jump to content


* * * * * 1 votes

All Gpl Mods Specs Spreadsheet


  • Please log in to reply
43 replies to this topic

#31 Lee200

Lee200

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 1,388 posts
  • Gender:Male
  • Sim interest:GPL

Posted Oct 11 2017 - 09:02 AM

Got it...thanks.

#32 Border Reiver

Border Reiver

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 960 posts
  • Gender:Male
  • Location:St Andrews, Scotland
  • Interests:GPL and GPL editing
  • Sim interest:GPL

Posted Oct 11 2017 - 10:35 AM

View PostLee200, on Oct 11 2017 - 08:36 AM, said:

I'm just guessing, but I question how accurate GPLRA is at measuring deceleration which is used as the primary input to your equation.  If it is inaccurate, then you'd get inaccurate horsepower results.

I don't know if you ever saw this.  It describes a method of testing the horsepower at the rear wheels for the '67 F1 cars.  In each case, the difference between the engine and rear wheel horsepower is about a 20% loss.

What is the 3.6 constant in your equation for?  I understand the basics where Power=Mass*Acceleration*Velocity and 9.81 is the KGs to Newtons factor and 745 is the Watts to Horsepower factor, but the 3.6 constant is a mystery.

Thanks Lee, that was a really interesting read! :)


View Posted76, on Oct 11 2017 - 07:55 AM, said:

Lee , this would be logical but the tests do not always give lower values

ex Copper T79 , my test 284 HP , friend's test 274 HP ; GPL 260 HP

From reading the article Lee posted, I think that part of the problem here could be as highlighted in the article that it is m* and not m which matters. IIRC the original 67 cars do not model the body lift of the car and so m* = m for all speeds, whereas I am pretty sure that 66 mod does include body lift at high speed and so m* in this case will be less than m, the static mass of the car which will skew the equation in the opposite direction. Presumably this would also get messed up even worse though if you tried to do it for a 69 car with wings as then the drag and lift (well, downforce and so effective mass) become a stronger function of speed and so here m* is definitely not equal to m as was being assumed.

EDIT: Same deal for 65 mod too IIRC. I think that 65 mod and 67 original do not model lift and so m* = m is a good assumption. This might also be true for F2 as well, I can't remember. For other mods I am pretty sure that aero lift and/or body downforce are coming into play and so in those cases you cannot simply use m* = m in the equations. I would assume that 65 and 67 mods will give a fairly consistent loss factor comparing measured versus stated value from GPL, although the variation seen in this was presumably then used to "balance" performance between slots to make everything come out about right in terms of relative performance on track. For all of the other mods though I am pretty sure that m* does not equal m to a good approximation at all and especially when testing at Vmax, so this must skew your results.

Rob

Edited by Border Reiver, Oct 11 2017 - 10:45 AM.


#33 Lord

Lord

    Mr. Broken Gearbox

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 127 posts
  • Gender:Male
  • Location:Latina, Italy
  • Interests:Formula 1, Football (Soccer), Grand Prix Legends and rFactor
  • Sim interest:GPL

Posted Oct 11 2017 - 12:29 PM

Mh, I'll read on Doug Nye's and Haynes's books to check if the figures are right: I can absolutely say the last word on the Lotus 49 (every chassis: fight me :P ), but for the others, Coopers and Ferraris I'll have to check better :)

#34 JonnyA

JonnyA

    Perry McCarthy

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 491 posts
  • Gender:Male
  • Location:Dorset, UK
  • Interests:Historic motor sport, sailing, astrophotography
  • Sim interest:GPL

Posted Oct 11 2017 - 01:06 PM

Ed, with your 66F2F3 carsets, my French is not good enough to understand the installation instructions. If I merge the files from your .rar file, will it write over my existing 65 mod? Do you have English instructions? Thanks.

#35 ed76

ed76

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 556 posts
  • Gender:Male
  • Location:Rouen
  • Sim interest:GPL

Posted Oct 11 2017 - 01:11 PM

View PostJonnyA, on Oct 11 2017 - 01:06 PM, said:

Ed, with your 66F2F3 carsets, my French is not good enough to understand the installation instructions. If I merge the files from your .rar file, will it write over my existing 65 mod? Do you have English instructions? Thanks.

don't merge , make a new install

#36 Lee200

Lee200

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 1,388 posts
  • Gender:Male
  • Sim interest:GPL

Posted Oct 11 2017 - 01:43 PM

Rob, I agree.  Since some of the mods model lift, the effective mass could be different.  Ed's method of testing is interesting though as I'd never thought about running up to max speed then disengaging the clutch and seeing what the change in acceleration is.  The theory is sound, but I wonder how valid the GPLRA deceleration result is...???

#37 ed76

ed76

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 556 posts
  • Gender:Male
  • Location:Rouen
  • Sim interest:GPL

Posted Oct 11 2017 - 01:51 PM

Attached File  Capture1 (1).JPG   60.95K   13 downloads

there is an other problem
I read gamma = 0,42 , but a friend tells me that the value is between 0.35 and 0.42, after him 0.37

Parce que le creux "pire" n'est pas réaliste.

Donc je prends dans la continuité de la courbe qui suit . On serait donc plutot à 0.37G je pense


Because the "worse" hollow is not realistic.

So I take in the continuity of the curve that follows. So we would be at 0.37G I think

Edited by ed76, Oct 11 2017 - 02:19 PM.


#38 JonnyA

JonnyA

    Perry McCarthy

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 491 posts
  • Gender:Male
  • Location:Dorset, UK
  • Interests:Historic motor sport, sailing, astrophotography
  • Sim interest:GPL

Posted Oct 11 2017 - 02:20 PM

Ed, Gem does not recognize the new gpl.exe file as a suitable base file for a new Install. If I manually execute the .exe it runs OK, does anyone know why Gem won't accept it?

#39 Lord

Lord

    Mr. Broken Gearbox

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 127 posts
  • Gender:Male
  • Location:Latina, Italy
  • Interests:Formula 1, Football (Soccer), Grand Prix Legends and rFactor
  • Sim interest:GPL

Posted Oct 11 2017 - 02:24 PM

Alright, straight from Colin Chapman himself, the specs for the Lotus 49 R1-R3 (1967):

Total fuel: 42 UK Gal. (16 on each sides, 10 between the driver's spine and the engine)
Camber front: from -0,25° to -0,5°
>> rear: same
Castor front: 3° fixed.
>> rear: (none)
Toe in front: 1/8 in. in total
>> rear: 1/4 in. total

Ground clearance front: 3,27 in.
>> rear: 4,13 in.

Weight: 1102 lb (495,9 kg) dry
             1562 lb (702,9 kg) laden.

---------Cosworth DFV------------

Compression ratio: 11:1
Maximum power: 408 BHP @ 9000 RPM (1967)
                             415 BHP @ 9500 RPM (1968)
                              430 BHP @10000 RPM (1969-1970)
Maximum torque: 245 lb-ft (332,17 Nm) @ 8500 RPM
Weight: 165 kg.

#40 ed76

ed76

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 556 posts
  • Gender:Male
  • Location:Rouen
  • Sim interest:GPL

Posted Oct 11 2017 - 03:01 PM

Jonny , you must had your usual GPL.exe !
(you have a MP)

#41 ed76

ed76

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 556 posts
  • Gender:Male
  • Location:Rouen
  • Sim interest:GPL

Posted Oct 11 2017 - 04:38 PM

I want to make a South African carset 1964
the cars had engines 1500cc S4 ford or Alfa 140 to 150HP at 7000rpm
I think the closest is 67F2BT



the laptime taken at Kyalami seems to correspond, + 4s compared to those of a BT11 2.7L

#42 Michkov

Michkov

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 723 posts
  • Gender:Male
  • Location:Graz
  • Sim interest:I am here to spam and wish to be deleted after registering

Posted Oct 11 2017 - 07:23 PM

View PostBorder Reiver, on Oct 11 2017 - 10:35 AM, said:


From reading the article Lee posted, I think that part of the problem here could be as highlighted in the article that it is m* and not m which matters. IIRC the original 67 cars do not model the body lift of the car and so m* = m for all speeds, whereas I am pretty sure that 66 mod does include body lift at high speed and so m* in this case will be less than m, the static mass of the car which will skew the equation in the opposite direction. Presumably this would also get messed up even worse though if you tried to do it for a 69 car with wings as then the drag and lift (well, downforce and so effective mass) become a stronger function of speed and so here m* is definitely not equal to m as was being assumed.


That seems wrong, mass should be invariant in these calculations (neglecting fuel burn). Are you confusing weight and mass here? Weight changes as it's a force, with lift acting the opposite way you get an effective weight at speed than standing still. Mass does not as its an intrinsic property of the object in question. The only way I see how weight comes into this is if you wanted to account for tyre grip.

#43 Border Reiver

Border Reiver

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 960 posts
  • Gender:Male
  • Location:St Andrews, Scotland
  • Interests:GPL and GPL editing
  • Sim interest:GPL

Posted Oct 12 2017 - 05:23 AM

My terminology regarding mass and weight was not precise, (and I am sorry), but it followed that used in the document which Lee reposted above.

My point is that the analysis here is based on assumption of balanced forces when there is no net acceleration, which when the motive force is then removed will reveal how much motive force was previously being required to maintain that equilibrium state. However, I do not think that the free body diagram of all of the forces on the vehicle for this assumption is correct and I think that weight does matter as it will affect the rolling resistance since the car will be lighter at high speed which will change the friction of the car with the road as the vehicle slows down. This means that the form of the equation used to fit the deceleration curve isn't going to be the same for mods where there is body lift modelled versus mods that do not have this and so the meaning of k1 and k2 from the fitted equations will not be the same in those two cases, which will also then make the interpretation and comparison of extracted data between different mods more difficult.

Rob

#44 ed76

ed76

    Denny Hulme

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 556 posts
  • Gender:Male
  • Location:Rouen
  • Sim interest:GPL

Posted Oct 12 2017 - 05:55 AM

what is the difference in% between m and m *?

The other causes of errors are very numerous:
is it really Vmax?
  To do this one must have perfectly done its setup, especially the gearbox  and tire pressure
heating the tires and the engine
have a straight line sufficiently long (chroma key is a bit short)
is there any certain GPLRA graphics (or GPL setup manager data)
uncertainty about gamma reading ... etc

Edited by ed76, Oct 12 2017 - 06:39 AM.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users

Sim Racing Links