I'm having a little bit of trouble creating an engine for BeamNG. My plan was to create a replica of the engine I was going to use (a 4A-GE) in Automation (another game) in order to get a relatively realistic torque curve. I am also using the BeamNG Flywheel Torque Calculator spreadsheet. My issue is that when I input the torque figures, I end up with less HP than what Automation says my engine should make. The torque is also pretty high and flat across the graph, higher than the HP for almost the entire curve. This doesn't seem right for a 4A-GE. Now, I'm a little confused. Sometimes, you hear something along the lines of "Horsepower is non-existant, it's just a theoretical figure produced by putting the torque figures through a formula" but if this were true, then engines with the same torque would always produce the same HP. You wouldn't be able to increase one without the other. And yet, all the time you hear that a longer stroke will increase torque, or that American muscle cars typically have more torque rather than HP, or that some diesel V8 makes 150 hp also makes 1,000 Nm of torque, while another i4 might also make 150hp but only 90 Nm of torque. So I guess I have 3 main questions: Is HP really just calculated by using the formula of RPM x Torque /5252? And if so, how do we get engines with varying torque figures that seem separate from horsepower? For a 4A-GE, does this torque curve seem accurate (see below)? Are the horsepower figures shown in the BeamNG Flywheel Torque Calculator spreadsheet accurate, and how do I go about increasing the horsepower without moving torque in the JBeam? The torque curve of my 4A-GE in Automation. Thanks for any help in advance.
Few topics below this topic, there are more links that should give you lot's of information: https://www.beamng.com/threads/tips-for-creating-custom-torque-curves.52342/ HP is just calculated value, torque is measured that is in reality. In BeamNG, you have flywheel figures, but what you input is not flywheel figures, it is gross power without any losses, then you have losses like friction etc. Anyway go trough those link and you will find tools, information and more than you probably will need to get your torque curve right and maybe bit more than torque curve.
I've already looked through that topic, and as I said I'm already using the spreadsheet that accounts for losses. I'm not looking for a torque curve either, I already have one.
Too add a bit more meat to that answer, HP and Torque will always cross at 5252 RPM no matter what as long as the units are lb-ft of torque and horsepower. If you are running the units of Nm and kW, then the crossover point according to my calculator seems to be 9549 RPM (can someone confirm that? Just want to make sure I did my math right). Having said that, it is now easier to understand how we can have a 150 HP diesel and a 150 HP gas engine that make such vastly different amounts of torque. that diesel will probably be making that 1000 Nm of torque at about 1100 RPM while the little i4 will be making the same power all the way up at 12,000 RPM. Basically, Power, either horsepower or kW, is just the rate at which something can do work. If we say that the work being done is the Torque, and the rate it is being done at is the RPM, what magical number you get out of the other end is the Power. Both engines can do the same amount of work, but need to be going massively different speeds to do so. Following this logic, yes, I am saying that if you took out that caterpillar bulldozer engine (or whatever that thing is) and replaced it with that motorcycle engine and just increased the gearing ratio a lot... the vehicle would drive almost exactly the same (besides the noise of the engine doing more than 12 times the RPM it used to be to do the same job) If you wanted too, you can take that to crazy extremes. If you found an electric motor that could make 10Nm of Torque at 110,000 RPM and really reduced the gearing... it too would make that same 150 HP and it too would technically drive that Bulldozer. As for the muscle car thing... another thing to note is that both the Power peek and the Torque peek are not at the same point on the graph (Unless they magically occur at one of those two RPM's I mentioned earlier... then I guess they would). In a muscle car that makes more Torque than Power (my 84 Corvette for instance... 290lb-ft, 205hp) they are low revving engines... almost like a diesel. They have really long strokes, so they can't rev all too high or else the mean piston speed becomes to great... that and using low quality materials back in the day also played a roll in that as well. That being said, most muscle cars usually don't rev much above 5000 RPM. Not doing so magically already means that you will never have more Power than Torque. I think the highest I have ever taken my Corvette is 4800 RPM and by that point, you can tell, it is running out of steam. Little rev happy Japanese engines on the other hand almost always go above 5000 RPM... I don't really know of one that doesn't. The Japanese learned early on that the more revs you get out of an engine, the more power it will make, so you can make a little tiny engine with very little torque, make it rev to the moon and back, and still make great power... plus you usually get the benefit of better gas mileage when your not ragging on it. Edit: I think I see what's going on here. If I remember correctly, you will have losses in BeamNG due to the drivetrain. I think what you see in the little torque app in game is it calculated after those losses are taken out of it.
I'm already using the spreadsheet that accounts for drivetrain losses. I haven't even gotten to the point of putting the engine in the game yet, though. When I put the torque curve into the spreadsheet, I set all the torque values to be identical to those from automation, yet I get different horsepower figures after. Why is this? How do I get those horsepower figures higher without touching torque?
That Automation torque curve is off, it is not like real torque curve of that engine. This is only sales brochure torque curve that is lacking lot of dips in a curve, but it shows numbers, takes little work to get numbers out from it though as it is tiny and units need converting. But if you compare numbers you can see how far off that Automation curve is. Which torque app you are talking about? There is two, one that shows realtime and another that shows flywheel power? What if Automation is wrong? By 3000rpm you should have ~125Nm torque, not ~140Nm. I would not trust Automation it is just approximation like Automobile Catalog torque curves, more explanation in topics I linked. What I know is that if I put real data and account of engine's losses with 440cid's spreadsheet, I get in game torque curve app to show correct HP and torque. Now with graphs you have width and height, also units, you can get them look very peaky or very flat, in BeamNG torque curve app will make all peaks look rather flat because it is wide and not very tall, just focus the numbers, get those right and you have everything right, no matter how it looks.
I've been battling this as well coding some Subaru engines into the game. I can get the power figures right, but I'm ending up 20-30 ftlbs over on the torque. I think it's just one of those things you have to play around with until you get it right, unless you have a torque curve for a stock engine.
I'm only using Automation for a general idea of the curve shape, not for exact values. Even when using the real data of the 4A-GE, it doesn't seem to work out right in the spreadsheet. Putting in the peak torque of 148 Nm puts the horsepower at 100, not the 115-128 that it should be at (I'm aiming for a middle ground of about 120 on my engine) Also, I'm not using any torque apps, as I haven't put the engine in game yet, I'm still trying to generate torque curve data using the BeamNG Flywheel Torque Calculator spreadsheet. The real problem is that I have too much torque early on, and not enough horsepower later on. Yup, using newton metres and horse power. Also, I'm still struggling to understand how horsepower and torque are always linked. If I put the 4A-GE's max torque of 148 Nm at 5,800 RPM through the equation, I get: 148*5,800/5252 = 163 Horsepower Which is way off the 4A-GE's peak power. The other thing that I don't get is that I've previously made another engine using a torque curve graph from Engine Masters, and the horsepower matched up just fine there.
Try those values spreadsheet gives in game. Check which values that spreadsheet is showing in graph, notice how there is no HP that would go together with torque values you are supposed to input into jbeam in spreadsheet. Also jbeam input is torque in Nm not horsepower, you never input horsepower anywhere, that is just what game will calculate. I don't have your friction parameters, but I get 169Nm of torque as a result from spreadsheet that must be put into jbeam, your 163, is that in jbeam input column? That is Nm, there is no HP calculated from jbeam input, game does that. See spreadsheet I modified a bit, those HP and Torque figures graph shows are what you would get in game if you would use gross Jbeam torque as input for Jbeam, so spreadsheet calculates how much you have to add to those to get gross Jbeam torque in game and shows that in Jbeam Input: You could post what your spreadsheet shows to you of course as that would make it so much easier to figure out what is going on, but my guess is that you are thinking different units than what you are seeing here. Just input real 4A-GE torque for gross Jbeam torque and write to your Jbeam file values you see in Jbeam Input and ignore HP, then see it in game being magically correct, that is if you used same friction parameters in spreadsheet as in your Jbeam file.
Woah woah woah... slow your roll there for a second... First issue... you are using Nm in an equation for ft-lbs and HP... that alone will give you an incorrect answer. That 5252 isn't just a random number... that is a unit conversion number... it's just no one ever writes the units out. Second... You are using the Peek Torque number at the Peek HP RPM... that will also give you an incorrect answer, but at least I can see why you would have thought this would be correct. The 4A-GE engine it appears you are looking at makes 85 kW (115 HP) at 5,800 rpm and 145 Nm (101 lb·ft) at 4,800 rpm. Information acquired from here... http://webcache.googleusercontent.com/search?q=cache:http://mywikimotors.com/toyota-4a/ Alright... now... while HP and Torque are always linked because HP is calculated from Torque, the Peek power and torque numbers are not related to each other in the slightest. If you take 115 HP or 85 kW at 5,800 RPM, you will NEVER get 101 ft-lbs or 145 Nm. Same if we take that 101 lb-ft and 145 Nm at 4,800 RPM and try to get power. Reason being is because the Torque is different for an engine at different RPM's... Here is a 20V 4A-GE Torque curve... As you can see by the blue line... it isn't just straight across... it has bumps and wiggles all over the place. The peek torque happens where in the curve those wiggles are at their highest point. In this case 133.2 ft-lbs at 5274RPM. In the case of this graph, it has the HP on the Left and and the Torque on the Right, otherwise the lines would be pretty close. Ok, now lets do some math. Now if we take that number and plug it into our little formula... ((133.2 ft-lbs)*(5274 RPM))/(5252) = 133.7 HP If you look above... take a look at the red line right at the same RPM at the peek Torque... Pretty much bang on... Not impressed? Ok, let's take our peek power of 159.1 HP at 7309 RPM and calculate backwards to find Torque. ((159.1 HP)*(5252))/(7309 RPM) = 114.3 ft-lbs Take a look at the blue line at 7309 RPM... bang on the money.
So what is the correct formula for Nm? I think that would be better to work with since that is what both the spreadsheet and jbeam work in. Also, I probably should have mentioned that I'm basing my engine on the 16v version, not the 20v variant. I think I'm getting it now. I think I was forgetting that, even though torque starts to drop off slowly, the RPMs would still be quickly climbing, so the multiplication would result in a much higher value for peak power. I think I've got it now, thanks for the help. Here's a screenshot of my spreadsheet for those that wanted it:
It mentions 7700rpm, not sure if that should be your maximum rpm, text is kinda hard to read, so what that 7700rpm is really is bit unclear, but as it is not max power, it might be max rpm. But also it is 4A-GEU variant, which is bit different I guess, there are so many variants that hard to keep track of all of them.
Yeah, 7600 is the max RPM. I realised after posting that it was set to 0, because I shifted a few numbers down, but I've fixed it anyway. I'm working on the GE 16 valve variant, not the TRD GEU 20 valve variant. The 4A engines are probably among the easier engines in terms of keeping track of the different variants, and yet it's still pretty tricky. There's just so many lol.
Ah man every time I see a post and quote it, I see your ninja lol Anyways, what atv_123 said is correct, you said you were using Nm (a metric value) and putting it through the equation for lb-ft (an imperial value). lb-ft and hp always cross at 5252RPM, while Nm and hp always cross at 9549RPM. Here is a video on the topic (one of his older videos, but still a great explanation):
Thanks for the video, probably should have checked to see if he had one in the first place. I've put the engine in the game, but I'm having some issues. First is the power. I'm putting the numbers from the "Jbeam input" column of the spreadsheet into the jbeam, and the engine debug app shows 142 HP, but the spreadsheet calculated it to be 115 HP. I've made sure the spreadsheet is using the same friction and dynamicFriction values. Second is the rev limit. I've set it to 7,600 in the jbeam, but in game it seems to be limited to about 7,300. Finally, I've set the maxTorqueRating to 180, which should be plenty more torque than the engine is making, yet puling hard in 4th still blows the engine (though the debug app shows the torque hitting 200-260, but when the torque stabilises it blows at about 200-240). Perhaps something to do with gearing causing a higher torque? I can post the spreadsheet and jbeam if needed.
A picture of the jbeam and spreadsheet would help, but I think it's because you didn't enter in values all the way to 7600RPM. Also yes, th gearing is blowing the engine.
I realised the turbo was carrying over from the last engine, but my problems still haven't been solved. With both the stock intake and the sport intake (the only options on the covet), I am still making less power (about 85-103HP). I'm guessing that the intakes affect power by restricting it rather than adding to it. If this is so, then I'd guess I also have to create a new intake with no restriction in order to achieve the power I entered. Removing the turbo also stops the engine from blowing, but pulling hard in 2nd shows an over-torque risk on the damage app, and I think I feel a small power drop at this point, too. Perhaps this is the cause of the lack of power? Also, removing the turbo seems to have dropped the rev limit even lower to about 6,600. I don't have any more time for testing tonight, but I should be able to test tomorrow. Screenshots are here: Spoiler Jbeam (part 1): Jbeam (part 2): Spreadsheet:
Are you calculating burn efficiency? Also, the stock intake on all cars doesn't have any effect on the car, besides adding the header mesh.
That is odd input, your maximum torque is at 6600rpm? Check your engine file with search function of notepad++, search for maxRPM, there might be some other part setting max rpm. Also make sure you don't have another file with same engine name, which might get used instead of your current file, that is not actually rare to happen. Your max torque is set to 180, but I think you should have that at 300 or so. If your intake or some other part changes torque values (NOS?) then that would explain why you get over torque, but then again you are so close that it might issue a warning, not sure of that. Most likely explanation that comes to my mind is that your torque curve is not one from the file you think it is from.
