First, a little bit of application background. Recently I came across a shipment of DC-powered pumps (like these, it says diaphragm but in reality they are more like plunger ones, i.e. they require a "slow" driving motor), all of which turned up dead after running for 6-72 hours.
Their DC-motors were to blame - all of them seized to start and pulled excessive current (even more than if you were to stall the rotor of a "still working" one). So, I figured either the commutator or the rotor itself shorted, and disassembled some of them to see what's up. And here's what I saw (please excuse the shaky hands):
See the radial surface of the brushes? Its axis is perpendicular to the rotation axis (don't mind the little "correctly" oriented grooves, the commutator grinded the brushes a bit)! That's how all of them came from the factory. Is it just me, or is the radial surface shape actually supposed to maximize brush-commutator contact area and slow down brush wear? And thus the aforementioned axes should coincide? And the reason why they all died is a shorted commutator due to brush material being grinded away too fast and clogging up grooves in the commutator?
At first I thought they messed up brush placement, but no, there's no way to rotate brushes 90 degrees, the springs are specifically designed to enforce this exact orientation.
Fine. Forget about pumps. I ordered a bunch of (very) different RUICHI DC-motor models, and all of them use the exact same brush housing design (at least the ones with a plastic back) and possess the same geometrical issue. However, this time, none of them died during my testing. Also I noticed that commutator sparks had different color (white instead of yellow), so I'm thinking different brush materials are used.
Well, I am starting to doubt myself. Is this counterintuitive brush configuration OK? Why is it used?