Ultra narrow 2WD differential for standard three-gear system. gear type: 48 pitch, pressure angle 25 degrees gear width: 9mm gear tooth count: 51 (typical Team Assoc. 2WD 1/10th scale electric off-road) diff ball diameter: 7/64 inch, x10 diff ball center placement diameter: 19.5mm thrust bearing ball diameter: 5/64 inch, x8 inner diff bearings: 10x15x3mm, x2 outdrive outer diameter: 13.0mm (outdrive bearing 13x19x4mm) dogbone ball diameter: < 8.0mm dogbone pin diameter < 2.7mm dogbone shaft diameter shown: 3.7mm dogbone pin protrude shown: 2.3mm inner axle length: 11.0mm (touch-to-touch on dogbone ball extremity) typical dogbone ball-ctr to ball-ctr: 21.0mm (as shown, 1.0mm cushion per side) outdrive taper angle, dogbone "gamma" angle shown: 22 degrees outdrive bearing inner to inner: 13.2mm (configurable, can be extended) DESIGN NOTES: Losi XX-4 had similar outdrive dimensions, namely 12.7mm outer and 8.0mm inner. By employing this thicker outdrive cross section, a dogbone pin that protrudes past the 8.0mm dogbone ball even 2.3mm still clears a perfect 13.0mm diameter no matter what the angle of dogbone (assuming the pin diameter is not excessive). Therefore the dogbone ball can be "inside" or "within" the 13x19x4mm outdrive bearing. The bearing can offer strength to the outdrive, holding it together if it's made from a weaker material, much like the XX-4 had "outer ring strengtheners" for its moulded outdrive variant. The inner axle design is such that the 11.0mm long inner axle can be made of titanium or steel, and the outdrives (or "diff halves") are preferably made out of 7075 aluminum. The steel threaded inner axle will be glued into the right diff half using red threadlock compound. It may be possible to use moulded plastic material for the outdrives, but then there is the issue of how to keep the inner threaded axle in place. These are designs that I happen to stumble upon in my quest for the ultimate 4WD drivetrain. By bringing the dogbones close together, lengthening them, and doing something similar to the rear suspension arms will increase the overall performance of rear suspension. In this particular design it's possible to use universal-jointed dogbones or it's possible to use the spring-loaded Roger Curtis design that was made popular in the original RC-10, by adding a ~1mm thick plastic stopper which has a 1.2mm diameter pin that inserts into both sides of inner diff axle (1.2mm hole drilled but very shallow). I have many more ideas for 2WD, such as underbody aerodynamics and front simplified suspension and steering geometry, but my focus is really on 4WD. Thank you for stopping by. I hope you check out my other designs, e.g. the 4WD road differential which is a 2-in-1 diff (right to left and front to rear).

See also original version: https://odysee.com/@Neri_Engineering/4wd-road-diff-assy The major changes include diff tube cutouts for maximizing diff stub outer diameter, and better engagement between diff rings and diff halves. Music added. Hope you enjoy.

I want to specify the angles of various pieces because these angles are rather difficult to come up with, mathematically. I could see two methods of coming up with these formulas - the first method is to use projections of angles in 3D onto 2D planes, and the other method is by analyzing what certain points of the system do by passing them through a composite of 4x4 matrices, and then seeing how the positions of those points are altered by said matrix. Anyhow, here are the formulas that I came up with, by using a direct mathematical computation method involving projection of angles onto planes. Define 'theta' as the angle of rotation made by white shaft. This white shaft will be our "input shaft" and it's rotating at constant angular velocity in the animation. The angle theta will usually be changing rapidly, as the input shaft turns. Orient the white yoke at theta angle zero such that the incident angle of green shaft (described below) causes the green shaft to bent towards one of the white yoke pivot points, and away from the other white yoke pivot point in a symmetrical manner. Define 'gamma' as the angle of incident between the two shafts (white and green shafts to be exact). An angle of zero would correspond to the shafts being perfectly aligned and an angle of 90 would correspond to the shafts being perpendicular to each other (an impossibility in practice). In practice on a universal joint this angle gamma rarely exceeds 45 degrees. In the animation, I believe, the chosen angle gamma was set to 46 degrees or so. In the animation the angle gamma is constant, meaning that the angle of incident between shafts is not changing for the duration of animation. Define 'alpha' to be the angle of rotation made by green output shaft. If the incident angle gamma is zero then the axis along which white shaft is rotating is the same as the axis along which the green shaft is rotating, and theta will always match alpha. - alpha = atan2(sin(theta), cos(theta) x cos(gamma)) Now the red inner cross can be drawn as rotating with the white shaft, but it will be swiveling in the white yoke to point towards the green shaft on its other axis. Call this swivel angle 'phi'. For example with an angle phi of zero, the red inner cross sits perfectly square in the white yoke. When the angle of incident angle gamma is 30 degrees (for example), then the maximum swivel angle phi of the red inner cross will likewise be 30 degrees. The swivel angle of the red cross (phi) oscillates between zero and the angle gamma (but technically it reaches a value of -gamma as well, which is represented by the formula). - phi = -asin(sin(gamma) x sin(alpha)) Hope that helps!

This is the previous animation with a few parts made translucent, a few lines added (to show projected angles), and the camera rotated and following the frame of reference of white shaft. In other words, the incident angle between white shaft & blue ring, together with incident angle between green shaft & red minor pivot pin (angle measured from its own frame of reference, NOT the frame of reference of camera) are equal, as they were in previous animation. The yaw angle is increased and decreased to show movement, as before. I have drawn some projected angles, overlayed on top of the piece. - The yellow angle is stationary and represents the pitch angle of white shaft. - The orange angle represents the angle of the blue ring as viewed from this perspective (pivot of white & blue poking into our camera). Again, the angle made between yellow & orange is equal to angle made between green & its red minor pivot pin, which is on a different plane and is not drawn visually. - The brown angle is twice the angle away from yellow as compared to orange. In other words orange is the bisector of yellow-to-brown. The brown angle is something that we will be striving to measure mechanically, but it's nearly impossible to arrive at without over-complicating the mechanical design. If we were able to attain the brown angle somehow, then halving it to arrive at the correct orange angle would be simple. Halving angles in mechanical design is a simple operation. - The pink ball and resulting pink line represent a hypothetical angle made by the "lion gear" (a half-ring joining white shaft pivot ponts) if that lion gear had a groove to track a hypothetical cylindrical protrusion at the spot where pink ball is currently drawn. - The purple ball and resulting purple line represent a hypothetical angle made by the lion gear if it had a cylindrical protrusion at halfway around its ring that tracked inside of a groove that was etched into the green shaft/globe, where circumference is marked with black circle. We see that neither the pink nor the purple lines give the desired brown angle. Taking the average of pink and purple in some manner may give a very close result to brown, or perhaps even a perfect result, depending on how we perform the average. However that would complicate the design. It seems that the pink and purple give the same amount of error. I think we would prefer using the purple tracking method becuase it would be less strain on parts (less movement and less forces).