Ah yea for turbos it's not really feasible. Also for dyno sheets, they are never really accurate, there's a lot of calibration and correction involved with these, it's not uncommon that you do multiple runs with like 10hp difference. Not to speak about actually using the same car on different dynos
They say you can't compare dyno to dyno, only dyno on same dyno, that is of course correct. All I know is that if I could input torque and it would end up being torque in game at max boost and max throttle, give or take a little because of temps etc. that would save hours of my time in tweaking the curve to be just about right. Of course then performance matching needs tweaking in all other resistances, aero etc. which I don't yet even know about.
There is something I found while doing research, blog has quite many dyno curves, pics of cars, engines, specs about builds etc. Worth to grab while it still is online: http://www.dynosty.com/blog/
I'm not sure if this would be related but there was an old PS2 game from Bethesda called IHRA Drag Racing 2 that let you customize almost every aspect of the engine that would affect performance; bore, stroke, cylinder count, compression ratio, throttle body style and CFM, intake manifold design, intake port(s), exhaust/header style and size, valve diameter, lifter style, type of fuel, camshaft specs and valve timing, and forced induction style. What I found interesting was that in addition to having almost any type of preset you could imagine, it would also let you make full custom engines and display a torque curve for the motor so you could visualize what each change would do in real time. Here's a link to gameplay, skip to 4:18 for engine customization I think it would be unbelievably cool if BeamNG implemented this type of depth in the powertrain system, but it may alienate some of the modders and bring the game that much further from 'game' and more towards 'hardcore simulation'.
As far as I know, IHRA 2 was only for PS2, but I run the disc through an emulator for my desktop so it's 1080p/60fps. You could probably do the same if you have a decent setup, and I imagine you would be able to find a copy on Amazon/eBay/whatever (or possibly just "acquire" an .iso for the game).
I know I have some sort of IHRA drag racing game... but seeing as I don't remember it being this in depth, your right, it probably isn't this game...
Might of been 1 instead of 2, because I remember such thing too, it was buggy, but it was quite nice because it had bit more depth than most games. There was discussion about Automobile catalog dyno curves, I think I spot some differences here, same engine:
Automobile catalog's dyno curves are notorious for showing very little torque in the lower RPMs and then having it dramatically drop off again after peak HP. Besides that, they can be very useful when someone is trying to understand how various engine features affect the performance (VVT, 2V vs 4V, etc.). Just take their info with a grain of salt, especially some of the pre-1974 domestic V8s.
Checking 288GTO and 288GTO Evoluzione torque curves can be revealing, they are identical, except other goes to bit higher rpm. However 288GTO has lower CR and boost is 0.8bar, while Evoluzione has 1.4 bar boost and higher CR, can't remember if there was even more differences (probably there is), so boost pressure has nothing to do with their formula or compression ratio. It might be that @atv_123 's tool is actually more advanced in this regard? I spend better part of yesterday in attempt to get @440cid 's spreadsheet to work with turbo setups too, I got it to where torque and hp comes out right, but it's still missing some bits because without turbo torque curve comes out funny, I would like to blame torque curve of Automobile Catalog from that though, so I have to test with other engine which I know being more accurate.
Well... mine has a few issues with boosted vehicles... mine is better at NA ones... I am not entirely sure what yet (still working that out) but with a little extra screwing around... I can still get a pretty close curve. 288 GTO 288 GTO Evoluzione Its not quite there, I get that, but then again, like I said... It has a bit of trouble dealing with boosted engines.
What I figured out is that I need to input max boost per engine rpm to get boosted curve right with 440cid's spreadsheet, then also there is efficiency of turbo in jbeam which is how many percent of maximum torque increase turbo is able to provide (ok, that needs perhaps bit more explanation). Try to ignore efficiency numbers, those are just some weird experiment. In turbocharger lua, there is a comment 1psi increases 6% of torque: So what efficiency in this context means is that efficiency of 1 would be that 6% power increase per psi of boost, 0.5 would be half of that 0 would be a turbo that does nothing I presume. Now I did peak a little of ETK turbo and Pessima turbo, their efficiency numbers are completely different, Pessima has two 1.0 values at low and high rpm, while ETK had only one at very low rpm. I have not been able to find information of how values are chosen. They can't really have relation to turbo compressor or turbine map efficiency as that has not much to do with engine rpm. Also I haven't examined if this effect for example turbo temp / exhaust temp, I know that in reality at higher rpm turbine side can be bit small, especially with oem setups and at higher rpm backpressure between turbo and exhaust valve starts to increase, which creates lot of heat and lowers efficiency because pressure build up takes energy away from spinning the turbo, it affects breathing of engine quite negatively. I don't know if this efficiency has anything to do with that. All I know is that efficiency number is how much turbo increases your power together with boost pressure. So I end up with this Torque/(C6*0.06+1)*(1-D6+1) Where Torque is JBeam Input (K -Column) in 440cid's spreadsheet (do note that I have added new column in front of RPM column!). However that is only case if there is boost, if boost is less than 0.1 value in K-Column is used. However that formula is still not really perfect, there are issues, but I'm not sure if those are because of Automobile Catalog's curve, 2.9l V8 running 0.8bar of boost probably is not making boost at 1500rpm at least not considerable amounts, I would think it would need 2500rpm to get going, I have better data from JZ2 though, torque curves of turbo and NA, also compression ratios etc. so might try that next. Your spreadsheet is however far more advanced as it creates curves from nothing but few parameters, I don't know if any of this is useful for you to improve your tool. What I need is a tool that creates torque values for JBEAM so that is what I have been creating with this formula, but I'm so dim that it's quite hard, maybe useful starting point for some brighter person
Well... mine actually can create it for Jbeam... sort of... with a little bit of work... I might be able to make a direct port page for it. Hang on a sec... be right back... Edit: Done... Copy... Paste...
That is really cool, it would save quite bit of time when one needs to do lot of testing. This is what I had in mind though, I have torque curve and boost data, which I input into spreadsheet, then it spits out numbers I can put into jbeam, I still haven't tuned exhaust power multiplier as that is needed to get correct amount of boost, so there is deviations because of that + there can be some other issues, but I think it gets quite close. This is 440cid's spreadsheet I modified to include turbo calculation, all I need to do is input torque from data and boost pressure + efficiency. I set up efficiency to 1 for all bins just for a test, if I set it lower calculation adjusts so that in game I would get torque from data which I want: Still there is that oddity of torque going down and up, maybe efficiency should be used to balance things so that torque steadily would rise with rpm? I think that after I tune boost by testing it ingame, I start tuning efficiency, hopefully I get then somewhat reasonable torque curve, I would think that my formula is correct, it is perhaps just matter of balancing other aspects to get curve nice.
Ahh... your talking about a full fledged engine with every parameter defined and what not... I gotcha... mine doesn't do that... With some work it probably could but that requires me to learn more about how to setup the turbos in game as right now I do not know how to do that.
That was actually one of the first things I added when you let me use the spreadsheet xD I did it in a slightly different way, where you copy only one column and there's no spaces. Pic: The secret formula:
I know how to make turbo, just few bits that I haven't been able to found much information, but they seem not to have huge effect of my ability to make turbo. Only things I'm not sure about are: "volumePer10kRPM" "backPressureCoef" Only thing that I don't know what other thing it affects than amount of torque increase by turbo boost: Efficiency in engineDef Most annoying part is that lot of adjusting is trial and error, that is exhaust power multiplier, efficiency and torque in main engine section, you have to balance with all three, make adjustment, test etc. Efficiency and torque in main engine section seems to be quite easy to set with formula I came up with, so that leaves now only exhaust power multiplier to be tuned in game by trial and error. Even for that educated guess is possible. So if you want to learn how, I guess I could do a little writeup at weekend or early next week?
It's crazy how hard to find are torque curves for the most common cars in stock form. Automobile catalog is a great source for mid-quality curves, but their database is not complete yet. Sometimes you may bump into a picture of the desired curve during a simple search, but that is a question of luck. Other databases are virtually impossible to find, but I found one: http://rototest-research.eu/ Also, I found some info on Volvo Modular engines: https://www.volvoclub.org.uk/tech/V40Specifications1997.pdf
