Why Hasn't first gone Brushless?

[Goldeye]

The cost-based argument against brushless motors is quite compelling, but how about the possibility of a single brushless motor (and appropriate control devices). Might there be some brushless motor that would fill a nice niche for FIRST robots; perhaps higher torque applications without needing as much gearing?

[Tom Bottiglieri]

Originally Posted by Jared341

Stepper motors would be SWEET. Position control without having to install yet another encoder and write yet another PID loop?

Sign me up.

Being able to use the CAN port on the new motor controllers should help *a little bit*

[Jared Russell]

Originally Posted by Al Skierkiewicz

Jared,
Stepping motors and feedback are not mutually exclusive. You have to know where you are in order to control where you want to be.

True, but for small loads, counting steps from a known reference position is often just as good as using feedback.

[Lil' Lavery]

Originally Posted by Goldeye

The cost-based argument against brushless motors is quite compelling, but how about the possibility of a single brushless motor (and appropriate control devices). Might there be some brushless motor that would fill a nice niche for FIRST robots; perhaps higher torque applications without needing as much gearing?

Brushless motors actually perform better under low loads (they don't have huge advantages over brushed motors for high load tasks), so a high torque application with low gearing is the opposite of what you'd want to use them for. That's why you typically see brushless motors in high-speed, lower-load applications such as hard drives, cd players, or RC airplane props.

The ideal applications on an FRC bot for brushless motors would be shooting wheels (best application), intake rollers, and some drive systems (or similar).

[AdamHeard]

I know how you could be solely responsible for getting brushless in FRC.

Simply get a brushless motor you would like to use, the speed controller for it that is compatible with our system, then buy, oh, 10,000 or so. Donate to FIRST, it may happen.

Other than that, sorry.

Separate from that, I've used brushless motors in combat robots before. FIRST shouldn't have brushless for two reasons; one, the drive is plenty, plenty powerful enough, and secondly teams would be destroying them constantly.

[Jeff K.]

One good thing about brushless is that they can come sensored, and extra encoders/sensors are not needed as the motor, tied with programming can precisely know its rotation or degree.

The brushed motors seen in these robots do not go through as many revolution in their lifetime as their similar counterparts in RC cars or in factory machines. These motors do not get to the point of maintenance that brushed motors usually require like new brushes or having the commutator cut. The maintenance of brushless motors in this case is unnecessary then.

Like Adam said, the motors we are given already are more than enough powerwise and efficiency wise for 2 minute matches, if designed properly. Races usually go at least 5 minutes up to 10 or even 20 minutes require brushless motors to last the longest and lipo batteries with huge capacities that do not start dipping until the last 1 or 2 minutes of the race.

These are smaller reasons of why they are not necessarily crucial to being in the KOP, but also because of costs. Brushed speed controller vs brushless and having 10,000+ motors. That would be a very generous donation as each motor would be at least twice the cost of a cim motor.

But to give brushless a fighting chance, motors and speed controllers have been getting cheaper. They could someday be cheap enough to make their way into the kits.

[Adam Y.]

Originally Posted by Jeff K.

One good thing about brushless is that they can come sensored, and extra encoders/sensors are not needed as the motor, tied with programming can precisely know its rotation or degree.

Most brushless motors are sensor less and that leads to another reason why not to use them. Sensorless motors are harder to start.

[Andrew Y.]

Originally Posted by Adam Y.

Most brushless motors are sensor less and that leads to another reason why not to use them. Sensorless motors are harder to start.

Cogging is what your thinking about, and this goes to cost again...low cost motors and controllers tend to have SEVERE cogging issues at low throttle

NEU 6pole motor with a MGM or Caste Creation MM or MMM can compete with many sensored setups. If i can get a 1500kv sensor less BL motor to go 0.2 mph on a crawler....Completely unrelated...i honstly believe sensorless BL motors will be the way...they are already proving to kick more butt on the tracks

[Adam Y.]

Originally Posted by Andrew Y.

Cogging is what your thinking about, and this goes to cost again...low cost motors and controllers tend to have SEVERE cogging issues at low throttle

That isn't what I was thinking of. Sensorless motor controllers work by measuring zero crossing point of the back EMF. If the motor isn't moving there is no back EMF which means you are essentially driving blind as to what position the motor to the windings.

[kamocat]

Personally, what I'd like is 3-phase motors. If you chose a motor within the correct speed and torque range, you need no gearing whatsoever, and can actually direct drive every wheel with its own motor. (crab drive with traction control, anyone?) There would be no need for a CVT.
However, there are some good reasons we haven't done this yet:

Luminary micro doesn't make a polyphase motor controller. While we could theoretically use 3 AC motor controllers to control torque, we would still require a variable-frequency 3-phase inverter to control speed.

AC signals can't be treated the same as DC signals. Calculating wattage and resistance is different, and you have to account for inductive reactance. Also, FIRST wiring standards would have to increase greatly to deal with 240v power.

I haven't investigated it, but I'm pretty sure 3-phase motors of a usable scale to us are mainly used on milling machines, and thus are upwards of $150 for the motor alone. This definitely isn't something FIRST can get for free.


Anyways, I figured I'd add my thoughts to the wish-list.

[Jared Russell]

Originally Posted by kamocat

Personally, what I'd like is 3-phase motors.

...

Unless everybody drags 120VAC extension cords all over the place, I don't think we'll be seeing 3-phase any time soon.

[Al Skierkiewicz]

Marshall,
Three phase (or polyphase) motors operate in relation to the way they are wound and the input frequency. Their greatest torque is at the speed for which they are designed. There is a distinct relationship between the number of windings/poles and the operation of the motor. When these motors are manipulated for speed, the controllers are adjusting frequency. The controller is essentially a large power amplifier. And they don't change speed quickly. In general when a person speaks of three phase motors, they are referring to power line operated devices used in locations where three phase power is available.

[Jimmy Cao]

There are so many problems with using 3-phase motors on FRC robots. Not the least of which is safety, having 120/240V around. I'm sure we'd all love to have kids wiring a high voltage system, which could rather easily kill them. I'm also not too familiar with inverters, but we would either need something that converts from 12VDC or have some way to store higher voltages (and then convert those to AC) on the robot, or have power lines leading to the robot (can you say "accident waiting to happen"?)

In my opinion, the safest, and most logical thing to do for FRC robots is to use some form of DC motors. Right now, brushed motors are being used due to logistics, and availability/price. Maybe, sometime in the (hopefully near) future, brushless motors could be used.

[Jared Russell]

While we're on the topic of allowable actuators, I would really like if we were allowed to use electric solenoid actuators. There are so many applications (dog-style shifters, brakes, clutches, latches, etc.) where I'd love to be able to eschew a full-size motor or piston, and where a servo just doesn't cut it.

[Al Skierkiewicz]

Originally Posted by Jared341

While we're on the topic of allowable actuators, I would really like if we were allowed to use electric solenoid actuators.

I wholeheartedly agree. There are a variety of great solenoids that eams could put to great use if allowed. Additionally, electrically operated clutches would be nice.