Calculating Rotor and Housing Shapes
Added 2022-12-05 04:28:26 +0000 UTCHey guys, check out this short demo which will appear in my next video (scheduled release: December 11). This shows the algorithm I developed to calculate both the rotor housing and rotor shape. As you may have noticed, I like to make things as realistic as possible so not only do the rotor and housing look correct, they are correct and have the ideal theoretical shape that a real engine would use. The algorithm can calculate these shapes using just the throw of the eccentric shaft and the rotor radius. I probably could have faked it and drawn a triangle with some arcs attached or used a pre-made asset, but where's the fun in that??
Comments
This is awesome, good job Ange. :)
2022-12-05 08:09:47 +0000 UTCA Reuleaux triangle is actually just the intersection of 3 circles and isn't what's shown in the video (it also isn't the ideal rotor shape). The ideal rotor shape maximizes the compression ratio while not touching the housing at any point other than the apex seals. The shape can be flattened from here but I'm not sure how common this modification is. Typically the compression ratio is altered via a depression in the rotor face (not shown in the video but nonetheless accounted for) which ensures that the combustion chamber never has 0 volume. Unfortunately I don't have access to a rotor to measure the exact profile Mazda used so it's entirely possible that they deviated slightly from the theoretically ideal shape.
AngeTheGreat
2022-12-05 06:32:13 +0000 UTCIf I recall correctly, the reuleaux triangle is the ideal shape being calculated here, which Mazda flattened to have a functional combustion volume (if the ideal shape is followed, there will be no volume at full compression).
dung
2022-12-05 06:02:59 +0000 UTCI love the amount of attention to detail in this sim. I'm grateful for that and I commend you for it
larsloveslegos
2022-12-05 04:45:52 +0000 UTCAppreciate the dedication to real life accuracy over the easy way to do it
Tyson Makovec
2022-12-05 04:31:16 +0000 UTC
