Let me know if you have suggestions on how to improve my engine building skills! Definitely need some tips on better timing and how to control engine power through timing adjustments!

@cat_lord5149

8 ай бұрын

Maybe put the sensors on the shaft I see that in a lot of really good engines, tho idk what that does better

@SenkoSan-kl3xo

8 ай бұрын

make a rotary gnome engine!

@akostorok5054

8 ай бұрын

Kan, i love you videos and if you want a piston engine with "gearbox" (simple timer piston engine without mod) u can see my build. (Toilet pump, piston engine with gearbox)

@idiotbutransrights

8 ай бұрын

I don't know how it would affect the pistons but I would like to see a "driveshaft" built like one of those old hand drills with the curved handle for spinning the drill

@T0xIcNigHtMarE

8 ай бұрын

Just out of curiosity could you please try replacing the timing wheel with pipe pieces?

Now that you have built a v8, there should be a multiplayer drag race with the challenge being building a piston engine powered dragster. I’ve heard a lot of people recommend this and now that you got a v8 it would be so sick

@The_Calvary

8 ай бұрын

Especially if they had to build pro-street/pro-mod style cars. Every time they do drag racing it’s always just thruster powered cars based off the Top Fuel dragster designs. Its not bad but it would be nice to see them build actual cars that they’re drag racing. Maybe even they all start with the same frame to see what they can all do

@xxxepicskaterman420xxx8

8 ай бұрын

@@The_Calvary yeah that would be cool

@s.s.j5046

8 ай бұрын

Aw, that would be cool, and racing with boats with these engines as well!

@alondjeckto

8 ай бұрын

V16🤑

@cooloof9007

8 ай бұрын

@@alondjecktow16

Like you said at 13:30, this engine is basically 2 inline-4 engines bolted together in the shape of a V8 engine. Now for scrap mechanic this is probably not a bad thing, because a true V8 engine needs a timing wheel with 8 sensors. This introduces a lot of extra complexity. Its definitely worth to try making a V8 with 8 separate strokes. Timing wheel wise, the first thing I can say is ditch the wedges. Everyone in scrap mechanic knows how bad the collision on wedges is if you've ever made a door. This is even worse for a piston engine. Another personal preference I have is putting the sensors on the crank instead of the color wheel, which is done to slightly compact the engine. As you've already stated in the video, losing the logic gates will improve the engine a ton. Every engine should have an angle on which it simply will not turn because all pistons are canceling eachother out. For this reason I personally always hook my timing wheels up with a 90 degree offset, ensuring that the engine starts in a non-rotating state right after spawning without logic gates. Finally I have a remark about the frame you build for the engine. One thing you want to make sure to minimize is potential collision points. Because you have rotating pieces right next to full blocks there is a chance these will collide, costing power. This is especially true when the potential collision point involves a wedge, due to their previously mentioned terrible hitbox. Again, getting rid of the wedges should solve this, but I also suggest building most of the frame out of pipe pieces. Thats part personal eastetics, part way to reduce (visual) size and part solution to any potential power loss due to collisions. Plus it allows for the use of 0-width blocks, which is a supprise tool I'll teach when the time is right. Expect way better passthrough, massive space savings on the crank and more! Edit: For those who wanna know how the 90 piston works, in the blueprint you'll see sth like: "xaxisA": 1, "xaxisB": 1, "zaxisA": 3, "zaxisB": 3 The value you want to change as kAN said is "xaxisB". The value can be either -3, -2, -1, 1, 2 or 3, but you need to keep in mind that: - "xaxisB" cannot be the same as "zaxisB" or its negative. (in this case not -3 or 3) - negating the original "xaxisB" gives a 180 degree rotation - using one of the remaining values gives a 90 degree rotation (in this case -2 or 2), which is clockwise or counterclockwise depending on which you choose Edit 2: Idk if anyone noticed, but kAN wired the pistons wrong, one side fires 90 degrees out of sequence if my analysis of the 10 seconds visible wiring is correct.

@ig-rd5923

8 ай бұрын

This engine specifically only has four power strokes because of where the crankshaft connects to the pistons. look at 0:31, you can clearly see, on the third set of pistons, they will need to fire at the same time. This engine is pretty much a boxer at an angle; "opposing" pistons aka pistons at a 90-degree offset in this case, fire at the same time.

@outandabout259

8 ай бұрын

about the 8 stroke 8 sensor V8, I tried building one and asked some people why it was terrible. Turns out that sensors in the corners of the sensor wheel are significantly off from where they should be, like 10-15 degrees, so the engine would not get a signal every 45 degrees but with uneven intervals. This is terrible for power and efficiency. So even though a V8 with 8 power strokes would theoretically be more efficient, that doesn't really translate into scrap mechanic.

@loneside4912

8 ай бұрын

@@outandabout259 I can attest to that. I have seen people use the sensors on grid and their engine works great. Me personally, I just have two sets of sensors (look at the top of the ws as of today) and that also works great and correctly. The issue with proper 45 offset timing is more so with anchor position of the pistons. If it isn't at the right point, timing is also effected. Anyways, in my testing with just 4 sensors, I had terrible power and torque ratings due to the offset timing of the physical pistons. With 8 sensors, the engine gets insane torque over the whole range (it isn't trying to fight against itself) and also results in more HP as a result. The reason why some older designs, such as those from Alstrak in the early days, worked so well with 4 sensors, is because those are with the old creative pistons. The piston that always fired early could not overcome the leverage of the piston firing on-time or after TDC, and as a result, would allow for good rpm and torque, thus, good HP ratings. With the newer, higher power pistons, we have to account for the opposing force, and can no longer do that offset timing with just 4 sensors- slows the engine rpm, and destroys power performance. So more or less, everyone building good piston engines now must work hard to tune them for best efficiency. Kein is an absolute madlad with how much attention to detail he puts, but is very rewarding as he has built some super impressive stuff that seem impossible. Edit: just noticed Vegard replied above your comment lol, he has a pretty good v8 using 8 sensors on-grid.

@blueflame_sm

8 ай бұрын

@@outandabout259 I did take that into account when building mine. I used 2 timing wheels offset by 45 degrees instead of a single 8 sensor wheel. Such an engine gets close to the power a boxer-8 like engine gets, but thats it. It idsn't all that much smoother either.

@casparroist2920

8 ай бұрын

comment bump, I hope kAN sees this

17:20 I have found that the collisions of different parts can collide. It's not really the friction aspect. In my own testing, I have found that sticking the sensors in place of the timing wheel is better, as the sensors can point to solid square mesh blocks. Then for timing, just paint different sections of the square mesh.

@paulkanja

8 ай бұрын

A common practice to compact this further is to replace 2 of the mesh blocks with the controller. I am not actually sure whether this impacts engine perfomance in anyway

@ame7165

8 ай бұрын

yeah i concur. there doesn't seem to be friction but the parts can flex under load, like the crankshaft, and then when it's even one degree off axis, the edges of sections of the timing wheel can collide with edges of the sensors and such. that said, the new pass through bearing and crankshaft he showed in this video seem to almost completely remove the flexing you can get with bearings, so maybe it's not an issue in his model

@itchytrack

8 ай бұрын

he does not need a block of room, the wedges are the problem. He just has to switch to using 9 of the same block so that SM sees it as one shape, not 7 with what he has

@RealPanzer999

8 ай бұрын

In my experience with moving parts on a bearing past other parts friction does apply, however I have no idea how much of a difference it makes.

Would recommend a mod like "Bingo's Block and Part Editor" which lets you blueprint edit stuff without having to spawn it on the lift and go to the files.

@maxgaming4627

8 ай бұрын

When the next video about crash lander

@jakewynn

8 ай бұрын

Thanks

@nikkiofthevalley

8 ай бұрын

Wait, what!? I've been using a custom blueprint editor for quite a while now, there's a mod for this!? That's helpful!

@Ben_Bingo

8 ай бұрын

Ayyy free mod advertisement

@maxgaming4627

8 ай бұрын

@@Ben_Bingo u no just tryin hype it up

So you've made an inline 4 and a V8 now ya just need to make the boxster and the rotary engine. Love these builds 🤘

@ktmmechanicsandaudio6803

8 ай бұрын

rotary would be crazy to see!

@outandabout259

8 ай бұрын

rotary or radial? Rotary engine can't really be made in scrap mechanic because it doesn't have any pistons.

@thejwilliams333

8 ай бұрын

​@outandabout259 unless you set the pistons up to act like the firing chamber in the rotary engine nit sure if it was spin center though

@djcarrotking

8 ай бұрын

@@thejwilliams333 I feel like that would essentially end up being a pretty crap ratchet system or just a boxer/inline

@timehunter9467

8 ай бұрын

A rotary engine would be scary and hard to see working, the cylinders spinning around would look weird. They’re scary in real life too.

I don't know much about piston engines but i do think that you should increase the mass of the timing wheel so it functions more like a flywheel. increasing the radius of the timing wheel would potentially allow for greater fine tuning I think.

I've never played Scrap Mechanic but still love watching the progress kAN has made with piston engines. Keep it up!

Love these piston engine vids man.

Really mindblowing techniques! Love that, i gonna try it at some point.

Played Scrap Mechanic ever sense it’s first year and I thought I’ve seen everything. Apparently I haven’t. Props to the folks who find out about all this crazy stuff. 👍

I've had weird issues in the past with wedges causing friction, especially while building tank turrets, so it may have an effect

its pretty cool that the v8 vibrations are still there

Found your channel from the first big wheel challenge video you made, I watch every engine video you make now. I know you beat the challenge recently but I'm hoping one day I'll see you absolutely destroy it

love these new piston engine videos

Nice! that passthrough will make drift cars much easier. Will act more like a true welded diff.

The Passthrough concept there is quite interesting. might be handy for some joints, maybe even ones with gears for transferring the rotation from a piston engine. Wonder if it could help with a trammel gear too

@crimzonplays1134

8 ай бұрын

Also works for an extra connection point from the body to the crankshaft to reduce the amount of flex pistons put on it.

I wanna see this engine, but 4 pistons on all 4 sides. That’d be funny and really epic to see.

16:46 I think you could test this by making two near identical engines with the difference being a space between the encoding wheel and sensors or not and hooking both up to the dyno to see if there is a difference in output.

I dont know why but this piston engine series is so entertaining

Man, that glitch welding method is smart. I used to just rotate things with controllers out of position, and then placing blocks there, but with the lift its much simpler

Definitely want to learn how to make those blueprint edited shafts and how to make variable timing wheels

He balanced the crank 🫡

Please look up opposed engines I’ve always wondered if they could come out on the torque side. Curious to see what you come up with.

would be interesting if you could experiment with some modpack angle sensor and number stuff. also w16 engine. timers as a vanilla alternative maybe to eliminate sensor jank

Can't wait for the W36

For the grid, could also do a boxer engine (Twin Sideways I3)

I'm not a fan of blueprint editing but I love that passthrough thing so much

Please make those concrete block using pipes it would look so cool xD

the passthrough bearing works fine using just blocks as the center and sides doesn't look as cool as using pipes, but works

Your learning alot about piston engines, but I want to see the training arc montage of glitch-welding. It's a skill you put-off for far too long and, no, you don't need mods nor file edits. You already done one of the techniques with the pass-through, but there are so many more. The anime of kan needs this in order for the plot to progress.

Find a way to incorporate those logic blocks as indicators if they are not part of the logic for making the engine work. They are a great mechanic to indicate firing order, like those spark plug indicators you could get for normal ICE engines when diagnosing, makes it a lot more interesting as a viewer to see.

Yeah, the particle generator seems to be working overtime on the timing wheel side, I don't think it would hurt to give a bit of a gap between it & the sensors. I know you can speed controllers up by editing the blueprint files. Might be worth looking within the file for a logic block to see if an edit might alter its response time.??

if you wanted to do a sort of load based timing you could try to use spirngs on the output shaft (like a sprung clutch) and if the spring is compressed the engine is applying more torque so you could change the timing based on that

It would be awesome if kan did a piston engine olympics sometime! Imagine this: 20 minutes to build the best poston engine you can and then you gotta put it through 2 different trials. Trial 1: Weight challenge, has to be able to turn a wheel with lots of weight attatched to it within a time limit, whoever spins the wheel the fastest wins the trial. Trial 2: speed challenge, attatch the motor to a basic car, the fastest to do 1 lap around a certain track wins the trial. Technically yes, all of this could just be measured with the dyno but I think putting it through trials would be more fun

It's been a while since I've played, but I think the big thing is using the logic gates. 1 tick is like 20 milliseconds? So the piston firing will always be 20 ms behind when it should be firing. 100 rpm being late by 20ms isn't much but 1000 rpm, 20ms is a much larger percentage (affecting where the wheel is when the piston fires) Removing that latency may be all you need. I'd definitely work on that before spending time on a dynamic timing wheel.

Now what you should try to develop is a vvt-i sistem using lots of logic gates and timers and maybe some controllers and pistons ina combination to detect load, speed, etc etc.

for variable timing, you could have the encoder wheel being larger, with the pattern of on/off at each block radius being different, and to change the timing you could use pistons to push the sensors out to the next bock radius, and thus timing pattern

@rccookie6202

8 ай бұрын

There’s no point in doing so if you only have 4 strokes - each stroke will be at exactly 90 degrees offset anyways, and you rotate the timing wheel to apply an offset to all of them

12:18 Family friendly torquing. 😂

I'm not sure how I feel about the save editing. I mean, like you said, part of what makes these cool is being able to build them in vanilla and you couldn't use that technique in survival or challenge mode. As for the variable timing, could you use another encoder wheel and a timer as a sort of accumulator setup that'll trigger things once the wheel's signals become frequent enough?

nice video

The separation of the piston head and base coordinates is probably due to the fact that they can extend so it's probably easier to represent their locations then just connect the 2 with the piston shaft instead of calculating the entire piston

@Poldovico

8 ай бұрын

In terms of blueprint data they're always contracted, though. In theory, you only need to save what one end is doing, and the other end can be calculated from that. Good that this isn't the case, though, or the blueprint edit wouldn't be possible.

the pass through perfectly works with a block and you only need one bearing for it

I find it funny that I pointed out you needed to get rid of the bearings on the shaft... and you needed a passthrough bearing. I did not know if you needed mods, glitch welding, or whatever, but it was plain you needed these things. So while I was not a master builder I had the concepts right. This should smooth things out a lot... and having it balanced too. So much better. Can you upload a save for the driveshaft and passthrough bearing?

Hey kAN, V8 Engines are possible with with a bearing between the crankpins, however some time ago the mechanical community of SM had figured out ways to also make short crank V8s using the the same technique like the one you are using. Here is a way to improve your V8 even further: Piston stretch. Apprently I had discovered that a while ago with my Vulcan engine (made it around 2018) and didn't even know about it. Some people noticed the difference between power I had compared to their V8s and it turned out that if you pull the pistons a bit further they are capable of even more power. Credit to Ben Bingo for finding that out with me. Also very important thing I learned is that you wan't your pistons to be set at a distance of 3 and not 2. SM physics makes it so you have more torque that way. This was told to me by Biggycheese. Anyways, like I have said previously I am happy that you are making engines now. I hope you continue and I hope we do a challenge one day. That would be awesome and I think many people would be happy if you try your knowledge against the OG piston engine maker. Cheers! :D

Can't wait for kan to bust out all these new techniques he's learning when they finally do a piston engine challenge in multiplayer monday

I haven't done piston stuff in a long time, but why use a timing wheel at all? pistons I made were always sensors on the pistons, if the piston was in position to fire the sensor would see and make the piston push. I made a v8 big rig with it, it was fast and powerful but I ran into lag issues with all the bearing and sloppy gearing, it didn't have any of the fancy bearing tricks shown in this video either, kinda want to try revisit it now haha

I’m curious to see if you could build something like an impact driver? And attach it to the piston engine. Idk how scrap mechanic physics would like it but who knows.

Could put the bearing attached to like W/S so as the engine gets load you can manual change the timing?

I'd love to see a gear shifter using piston engines

You'd need to build some sort of PID system to measure the speed of the wheel and change the position of the timing wheel to do something like that, but it'd require some crazy logic gating to stay vanilla. Maybe the flywheel will even out the jumps enough that a single timing will do fine.

You could make wacky vtec using weights attached by suspension, high rpm will push the wights out and activate the second timing, i think thats possible in vanilla.

im thinking about another encoder wheel but with only one coloured block and attach it to the bearing of the main encoder wheel, so if it goes fast it wont have time to change the angle of the main encoder wheel but if it starts to get slower, then the bearing will start to advance or decreese (depending of which one makes more torque) the angle of the main encoder wheel. please try this and lend me know if it works! greetings from Argentina!

I think the adjustable timing wheel could just be based off of rpm. Not sure how you would do that in vanilla, probably just by separating the piston firing timing itself from the driveshaft location sensors.

@jakewynn

8 ай бұрын

You do it by making the flywheel increase the angle on the switch, using the controller. Then the switch can slow down / speed up the engine. kan would have to do this manually

I think the next big challenge should be to make a V12 piston engine, and maybe make a super car chassis to put it in.

WOOOOOOOOOO NEW UPLOAD

18:30 I think you would just count the ticks between one sensor activating and the second one. You have a video on making memory bits in vanilla I think. 17:20 Vehicles have collisions with themselves. Look at any catapult, they only work because blocks can hit each other when part of the same creation. You dont need friction for a block to get in the way of another block.

What you can do for changing timing dependent on rpm is to have a sensor linked up to a timer circuit such that above a certain number of on/off cycles per second you get an output. You can do this by XOR'ing the output of the sensor and a short timer connected to it. This way, whenever there is a change in state, the XOR pulses. From there, you can mechanically measure the rate of triggering by having something push on a weak spring every time it triggers, you can also use something like a flip-flop to count every other time it triggers or something like that. I cannot however think of an easy way to do the measurement purely with logic and timers.

Try use this blue pistons from modpack. Torque will be insane

7:32 just how much force is that if you try to torture it with the ridiculous heavy block from the the modpacks?

Is there any way to create a passthrough in challenge mode? I have been trying to beat the hardest setting of the torque piston challenges, but I have been having lots of trouble creating trammel gears stable enough to actually work. The bearings just have too much give at such high torque and lock up after they bend out of alignment. I tried the other similar passthrough method shown in kAN's previous videos, but it isn't stable enough to deal with the torque it seems like.

You could try out this game called besiege. It's great for building actual engines and even has steam mechanics! Deffinetly recomend

I have 0 knowledge on how scrap mechanic building works but as far as an RPM dependent encoder wheel, could you use suspension to create an encoder wheel that changes what color the sensor sees as the rpm increases? The idea is that the suspension is either on the inside or the outside of the encoder wheel and as the wheel spins faster the suspension either stretches or compresses revealing basically a second encoder wheel for higher rpm.

use better lift, you can weld stuff to the ground and still pick up the creation makes it faster and just overall better. also lets you get creations that run away from you.

You know the little lever action thing on the side of steam locomotive wheels? That could be a good way to build an engine, you wouldn't have to pass anything through the axle.

I think you can make your V8 smaller by glitch-welding your pistons directly to eachother thus avoiding the need for the block in-between them. No idea what effect his might have if any, but might be useful to compact it incase you want to put it in a vehicle.

Also, what is stopping you from putting pistons on all 4 parts of that thing like with the magnet engine?

could you get that rotating shaft on the workshop?

Have a sensor on when 1 rotation has compleated to make a controller advance the timing and reverse the timing if it takes too long to compleat the rotation? This may work.

Concept: a two layer timing wheel. The layer closest to the sensors only has the white blocks, and the rest are missing. The next layer is half and half, and can be rotated independently to adjust the length of the activation. Instead of a piston that is on half the time and off half the time, you could adjust it in tiny increments.

I haven't tried it, but i bet you could use a single timing arm on the shaft, triggering a single piston to extend slowly. It would auto retract between cycles. You could use a block and sensors to tell what is position is, as a sort of rough speed gauge.

*GASP!* You lied to me! Where is the radial engine you said would be in the next video? LOL

What about aome kind of mechanical spring to control the advance on timeing.

fun fact: Wedges have bigger collision than blocks and that's the part of the reason why your engines create a smoke effect

I think using mods in the engine control is ok. Would be really cool to see

Can you try to make a star engine like in airplanes with four il-line fours please kan PS: awesome videoes I learned so much keep it going !

Could you make the timing wheel larger in diameter giving you greater control of timing

I think you need to build a standard vehicle that you can put the different engines into and do a hill climb or maybe a tractor pull to get "real world " results 😮

make an aircraft circular piston engine, and you could also do some kind of gears instead of pass-through, as the power transfer, does not have to be on the same axle, well you can do double sided pistons, dc turbine motors

i think because your taking power straight out of the engine you need to change the timing while its running. but alternatively you could build a simple transission which should maybe help

"Look at how smooth the shaft is" -kAN 2023

19:00 why can't you just track the revolutions? Like a bearing is tilted 5 degrees for every revolution, then as it reaches certain points, it triggers sensors to make the timing change. I think that would work.

For the love of everything that is good and holy, place the control block for the encoder wheel at the output of the engine and add a button to restart the engine so you wouldn’t have to run around to tweak it

Cool

cool

awesome lol

perhaps take the logic from the dyno, attach it to your piston engine and play with logic to apply degree changes to the timing wheel based on power out puts in the dyno logic?

@Ekipsogel

8 ай бұрын

Dyno uses mods

Welcome to the wonderful world of advancing and retarding an engines timing! Normally in an older style engine with cap and rotor it will advance or retard an engines timing via centripetal force and spring tension. There's also the use of vacuum as well. Not sure how you can translate that to scrap mechanic though.

@ducewags

8 ай бұрын

@Indy509 The old saying, people are like distributors. Some are advanced and some are retarded. Weights and springs come into play, more then vacuum, or the lack of vacuum on some engines. Fun fact, more then I can count, the gear on the shaft is on the wrong side of the timing line tooth. The roll pin should be on the half tooth of the gear, not on the full tooth timing mark when lined up to the pointer on the rotor.

your actually correct, having different tuning for load is what real world engines do with advancing the timing of the spark plugs. I think the main requirement to develop different tuning will be a tachometer, which obviously won't be easy in sm. or though it might be possible to instead of having a tacho, to actually develop a variable throttle, that could operate in the same way. for instance a simple version could be to use the steering ability from a steering wheel to control a lever that then has different points that sensors pick up on. or alternatively having a multi stage throttle, with more limited control.

When are you making a car with a piston engine. It would be super sick i think

you could have a little calculator that calculates the period of the wheel and use that to calculate the speed of the drive shaft qnd theoreticly it could all be done in scrap mechanic

YES FINALLY :D

so when are you going to host a piston engine contest?

Even in scrap mechanic, the LSX takes over

Maybe a survivor series but you can only drive something IR fly something with a piston engine

I was the first to like the video hell yeah love your content pick up the good work ❤

It'd be super bulky and Idrk know Scrap Mechanic enough to know if it'd work or if I'll explain what I'm thinking well enough but, for a variable encoder wheel, could you do a large wheel, with sensor rings, coloring each layer as needed for the higher speed. Then use a couple timed logic inputs to verify the wheel speed to determine which ring of sensors to use as the encoder. So like the inner most ring will run if no other ring is running, the next closest ring will run if two inputs receiver input within say 3 seconds, the next ring will active if they both fire within 2 seconds and so on.

We used to have a volvo penta boat engine wich adjusted its timing to accelerate, dunno if that helps but i just wanted to say it

is the passthrough on the steam workshop?

I can verify that there is zero or significantly less friction between parts of the same creation. I have a tree harvester that pushes logs into sawblades and if I weld the log to the pusher plate the pistons move just fine but if I weld the tree to the plate and then delete block to drop the log onto the chute the pistons struggle more and more with larger trees. So clearly there's a significant reduction in friction if it's all the same creation, even if they're separated by pistons.

@ame7165

8 ай бұрын

the problem is the wheel is made from wedge pieces and most crankshafts use bearings and those are very flexible under load, so the edges of the wedges that make up the wheel can catch and have collisions with sensors they're directly against (hence all the metal banging sounds we've all heard in this game). if things didn't flex at all, it would be completely friction free, but when things have collisions it defeats the whole friction free part lol. but it's nothing a gap cant fix