Comment by cpgxiii
1 day ago
I wouldn't say excellent. Motor current->torque is well correlated for motors in isolation, or with very low gear ratio. Not so with higher gear ratios. Can the current value be used to make control decisions for the motor? Yes. Does the current give you torque in Nm? No.
In the robotics world this is sometimes distinguished between actuators that report "effort" (i.e. a current-derived estimate) and actuators that report torque (i.e. actual torque sensing or direct-drive with current sensing). Both can be useful, but "effort" is not torque.
Current will never directly give you torque in Nm which is a straw man argument - you will always have to know the motor constant, input voltage, coil resistance, and gear ratio to back out torque. That's characterizing the motor. Yeah, that's a lot of work for a hobbyist, but it's not at all unreasonable to consider a motor an excellent torque transducer - that's what it does.
If the current->torque behavior of motors with gearboxes could be generally well characterized, the entire distinct market sector of "cobots" would not exist because every industrial robotics vendor would have long had good torque modelling on their actuators.
As it stands, Universal Robots (and likely their clones) do use current->torque characterization for their actuators (which, amusingly, is then stored on a robot-specific USB drive or SD card), and their torque sensing is shit. Shit enough that for any useful force/torque application you still need a separate force/torque sensor. Schunk, for some of their electric parallel grippers with "force" feedback, only characterizes them at a single velocity and there is significant error in the force estimate at any other speed. Good current->torque characterization of a complete actuator is so difficult that approximately no vendors in the automation space are willing to do it.
The main keywords here are high gear ratio and cheap parts. I am not making universal claims about all use cases. For most sane designs, sure electric motors are great at converting energy into torque. Normally you would want most of the electric energy to be turned into useful work. Which implies that electric power directly correlates with mechanical power or torque*speed. The same can't always be said about gearboxes, especially for some of those targeted at hobby/toy use cases with crazy gear ratios like 1000:1 or higher. Gear ratio can be so high that you will strip the gears or make the shaft connection slip long before you can apply sufficient torque to slow down the dc motor which is almost free spinning and spending most of it's energy to overcome friction in first few gearbox stages instead of doing useful work. In a toy or hobby project when you want something to spin slowly it might not matter that the gearbox is <10% efficient if it allows you to reuse same cheap brushed DC motor as hundreds of other toys. Increased torque is partially a side effect of slowing down the tiny motor not the primary goal, although at those slower speeds you probably want slightly higher torque but nowhere near as much as what the gear ratio gives in theory. Even some non toy use cases like cheap home appliances might occasionally use crappy inefficient gearboxes and dumbest electronics possible, especially if the 1-3W motor isn't main consumer of power. There might not even be motor controller or MCU to monitor the current.
It's not always question of lot of work for hobbyist, as it is result of using cheap of the shelf parts and modules which are optimized with different goal in mind and give very poor signal to noise ratio. Doesn't matter how much characterization you do if the change in temperature, grease viscosity and distribution, plastic flex produces higher variance in motor load than any force you can apply to final gearbox stage. I guess the more careful choice of suitable combination of parts from more specialized stores can be considered "lot of work for hobbyist" compared to picking first result on amazon or whatever you found in your junk bin so your argument still stands.
Of course high gear ratio or slow speed doesn't always mean inefficient gearboxes. There are solutions for slow rotation with or without high gear ratio which are reasonably efficient thus allowing to use motor current for estimating torque. And any serious or well designed equipment will use them. But that usually means more complex gearbox, motor controller or purpose built electric motors all of which is either more expensive or require high MOQ orders from manufacturers.