(post deleted by author)
This is another example of benefiting from flow work. Take a look at your first pic. See how the exhaust is going red while the rest of the āflow benchā is mostly green? You can tune the flow of the various parts to shift torque and power around, as well as open the engine up more or choke it down strategically. The constant is the valve flow, once the head and cams are chosen. Or you can change those up too and try to optimize. Bore and stroke variables change the valve flow, as do intake and exhaust decisions. Try it with this first engine and see what you think!
Thanks for making the car/engine Ray but itās not importing it into Automation for some reason. I added it to the CarSaveImport folder but nothing is showing up in Automation under the cars or engine, only the ones I have created.
I have been struggling finding out specific details on this engine. It seems a lot of the information is based on the Wikipedia article and the two best sources I could find were those sites I linked to, the official Toyota specs for the model I am interested in released in 2003-2005 and the Toyota Club link which shows some details on the variants of the engine.
There are also some conflicting details about which materials were used for the engine block and cylinder head, the Toyota Club article suggests cast iron for the block and aluminium for the head whereas on the Toyota site, they say aluminium for both. I couldnāt find any details about the manifold size or exhaust size other than a forum post suggesting the exhaust is 1 and a half inches in diameter so I converted that to mm and got 31.8. I am not sure if the EFI is single or per cylinder, I think per cylinder.
And I am also not sure about the catalytic converter as there seems to be either a version where it fits directly to the exhaust manifold and one that fits further down the exhaust pipe and Iām also not sure if itās 3 way or high flow 3 way. I also donāt know whether the exhaust headers are cast low or mid but there is a big difference in performance depending which you choose.
Perhaps I am restricting the ability of the game to reach my desired torque and HP goals in trying to stick so close to the real world specs. I also donāt know anything about tuning or engines in general lol.
IĀ“ll write you via pm as soon as I have timeā¦
for the rest: I am a dinosaur and know 80s cars quite good, and yes, SPI was rare in Europe, I cant recall an example for that, except the Lada Samara that had this afaik. Electronic carburetors or MFI were the things most German manufactors used until 1990. The SPI, however, simulates those electronic carbs quite realistic. Most cars had a Pierburg 2E2, like the Volkswagen Golf or Opel Kadett.
When most Japanese already used MPI systems in the mid-80s, a Bosch KE Jetronic was the peak in Mercedes, BMW and VAGā¦
What do you mean by MFI? Iād assume Mechanical and Iām aware the Germans used that a fair bit going from the mid 50ās Gullwing to at least as recent as the MK I Golf in the late 70ās. To be clearer as to why Iām asking, itās because āorā instead of āandā to me implies the latter is an alternative name for the former.
This is further backed up, as in it seems youāre using the āMā in āMFIā to stand for somthing other than āMechanicalā since youāre saying although SPI simulates the electronic carbs well, theyāre not the same, making me curious what this maybe other MFI alt name for the carbs is! āMonoā is about all I can think of.
Oh, and you specified a Lada as the only Euro SPI you recall, but maybe you meant outside of Germany?
Most modernish standard cars from the discussed era are multi-point ignition, single intake butterfly.
For engine creation I found it easier to mix imperial and metric, which you can do in the settings. So. Cylinder measurements in inches but capacity in cc/liters. I guess it just depends. You can use the Excel sheet exporter program to help get specs recorded between cars, and thatās what I use to do competitions.
How soon should the torque peak be, then? How about 3300-3800 rpm?
Ah, I didnāt realise I had the exhaust measurements the wrong way round and it should have been 38.1 not 31.8 lol.
This is turning into quite the challenge trying to get the Automation engine to match the real-worldās. With regard to the EFI system, from my understanding of the diagrams and write-up in that Toyota Club article, I think it is multi-point and the configuration is per cylinder as it looks like thereās one injector going into each cylinder but I may be wrong and each injector has a 10-point nozzle rather than just 4-points like in the K3-VE variant of the engine which the 2SZ-FE is similar to.
The 2SZ-FEās crankshaft also has 4 balance weights so in the Balancing Mass section I set it to Balance Shafts and from 50 to 4 (+0.7kg). I donāt know if that makes any difference or if I am doing it completely wrong.
I am not sure if the VVT setting should be set to intake only or all cams. From reading the Wikipedia page on the VVT-i system VVT-i - Wikipedia (first section), it just says āVVT-i varies the timing of the intake valvesā¦ā which I took to mean it just effects the intake rather than all cams.
Ok I managed to get within 0.1 of the correct numbers although itās probably pretty unrealistic. I got 122.1 NM of torque at 4200 rpm and 86 HP at 6000 rpm after a lot of tuning and setting the fuel mixture all the way to lean.
I chose to use balance shafts after reading up about them and the role of harmonic dampers in general. According to Wikipedia, balance shafts are used in inline 4 engines but I am not sure if mine does or not, I know it has balance weights but I donāt know what that means and whether or not there is a balance shaft so I just added one anyway and left it at the default 50.
I had to set the cam profile toward sport and up the quality. The area where I had to do the most fiddling was with the fuel and manifolds. I had to keep going backwards and forwards to gain a point here and lose a point there. In the end, I could have penny pinched from various areas and arrived at much the same conclusion and I didnāt know which are was better than any other.
I think I have been able to manage to keep most of the engine within the published specifications and areas where I couldnāt find any information and due to my own lack of mechanical or engineering knowledge, I had to take some liberties in chasing the numbers.
What does everyone think? Have I completely messed up and created something that is nothing like the real thing?
The problem with recreating real engines.
Automation donāt have ignition and injection settings adjustable for EVERY rpm.
IRL a EFI might have settings for every 100-250 rpm point, making it possible to fine tune the torque curve.
In some systems the torque/power can be adjusted for each gear as well.
This would add a complexity that would be to much for most gamers, maybe as a option for the more nerdy players among us. (Yes, I would activate such a setting )
Exporter_-_Yaris.car (13.7 KB)
Hereās my attempt. That engine, according to the wiki, has an 11:1 and not 10:1 compression ratio. What I did is lean out the engine a lot (maybe too much, look at that tiny throttle response stat) so that the torque curve naturally curves down due to too throttle retarding. The flow profile here is also a whole bit more reasonable than in the OP.
Either way, lean burn is a good way to make your engine fall off in the top end. your cams were pretty remarkably HRRRRGH (and mine still are fwiw, wtf Toyota)
The timing setting actually just means that the engine is tuned to run at a lower octane than what the manual recommends. This is a common adjustment done in order to prevent failures (or the need for excessively retarded timing) if you get a bad batch of gas. In Automation, it really just increases your reliability.
The reason the octane rating is removed from the left stats tab is because the engineās timing is automatically tuned to the octane number given by (Octane Rating - Ignition Timing Map), which is the one in the testing tab. This means that I donāt actually āhaveā 6 octane left over, Iām just tuning the engine for 84 and selling it with a manual that says āyou should put 90 in itā.
If you open the timing map (stopwatch icon above the power graph), you will see a map of the ignition timing (color) in a plot of RPM vs throttle position. The top of this map corresponds to the power graph (full-throttle operation).
Bright green on the graph means you have exactly the right octane to run optimum timing; red means that the octane is too low, so your timing is retarded from the optimum position; and blue means you have optimum timing, but you could have still had it on lower octane.
If a point on the timing map is blue, you will increase power/efficiency there by increasing compression, kind of like in the old designer when you had āoctane left overā; if the point is red, you will increase power by decreasing compression, because the engine wonāt have to retard timing as much to avoid knock. And the top right of the timing map (full throttle, max rpms) is very red in the engine I provided, meaning it has retarded timing and thus less power.
I didnāt ignore the information you provided Ray and I really do appreciate you taking the time to help me with this, I just didnāt understand it. By my own admission, I know nothing about cars, engines or the game itself and so a lot of the information being discussed goes over my head. I do try my best to take it on board and I read lots of articles about different things but I do miss certain pieces of information or just plain donāt understand it. People process information in different ways and something that makes sense to you, doesnāt necessarily make sense to me in the same way or at all.
I am trying to learn some completely new topics in a relatively short period of time and all I really wanted to do was recreate my friendās car (not mine, I am just a passenger in it) and export it into BeamNG for messing about jumping over ramps for a silly video. After watching some YouTube videos of Automation and Beam, I thought it would be a simple process of just getting the carās specs and inputting them into Automation to create the engine and then just make a rough body that looks somewhat similar to the real thing, how wrong I was.
I didnāt realise that Automation works in itās own way and converting real-world information into Automation language might be different as the gameās mechanics might not allow for more realistic engines, certainly with regard to this budget economy vehicle. I honestly didnāt foresee me having to go into this much depth to make the engine and suddenly having to scour the internet for information and try and get my head round mechanics and how engines work.
Thatās my fault though and perhaps Automation isnāt the game for me or at least for this fun little project, I should either just let the gameās AI auto-tune the engine for me after inputting some of the specs or just accept whatever I can get out of it.
Again, I appreciate everyoneās help with this and anytime you have had enough then I donāt expect replies or anyone to help, itās not that big of a deal. Maybe I should have explained all this at the start.
Donāt despair just because Automationās a little overcomplicated at times. Itās definitely not a game you pick up for ājust one little projectā, but as you will learn you can replicate or modify all sorts of engines in a fun manner.
Automation cannot literally create a copy of a real-life engine because if it could, then half of the worldās automotive engineers would have been out of a job by now. Engines in real life are complex things. There are simplifications and there are approximations, both to make the interface slightly less labyrinthine and to avoid literally turning into a hyper-complex thermodynamics simulator.
But Iām sure that if youāve shown this much interest over a Yaris engine, thereās more fun to be had!
I think itās just a familiarity thing. I assume you didnāt play around with this update when it was in its open beta; I did, and Iāve had plenty of time to get used to it; I barely even think of octane anymore.
Youāre absolutely right about the whole ātuning with the mapā bit. That map (and the efficiency map) are incredibly important to tuning now, because economy engines can deliberately be up-tuned so that they make the best use of octane at part-throttle and low to medium RPM (which is where youāll usually be when driving efficiently) as opposed to the peak power point.
Ope, sorry, wrong word. I meant they have huge compression ratios. Like thereās an actual 87 AKI-running Toyota engine out right now that gets a compression ratio of 13:1; the 91 AKI version has 14:1 compression. Mazda has had similar ratios for a decade by now, too.
Itās definitely not āthe devs faultā that something doesnāt match real world. Thatās not the right way to think about what Automation is doing. It isnāt supposed to replicate real world conditions beyond what is considered an acceptable simulation. These engines are showing you outputs, but is it explained that thereās x amount of parasitic loss due to alternator, water pump, etc? There are friction charts which are important to watch, but itās not like you can pick the viscosity of the oil and simulate a change. Instead itās a game where things are āclose enoughā and it is addictingly fun, like Satisfactory (for many of the same basic engineering-minded dopamine hit reasons.)
A note on balance: this simulation does not show primary and secondary vibrations so you really are using balance methods to achieve a rev goal, smoothness, power, or reliability. Youāll see each decision modifies these for or against. Anyway, the Mitsubishi-licensed balance shaft technology came first and was most common on large displacement 4 cylinder engines, like the 2.6L Starion unit, or up to the Porsche 944 S2 3.0L one. In a 4 cylinder with modest power goals below 2.4L (ish), crank weights should be fine below a certain rev point.
Enjoy the game and I hope to see your designs in some of the competitions we have. Thatās the best way to learn: friendly competition.
Barrels for carbs, luckily pretty well explained in the game. Dcoe stuff is not as clear.
I think what makes things less than 100% clear is that it should infer something like āinduction typeā , so I think itās more word salad and it might even be a regional thing. Automation also deserves some tooltip elaboration. Could all of this be better? Sure I agree on that.