FRC Blogged: - Motor Controllers

[techhelpbb]

The issue of lifting the current limit doesn't have to come down to whether someone designs a single mechanism that exceeds 40A continuously.

If you place 8-10 of these Jaguars on a robot and occasionally they draw more than 40A from the power system via multiple Jaguars they can collectively draw enough power to adversely effect the robot power.

That's an interaction that is a little more subtle. The larger point of interest for me is whether they upped the wattage of the current sense resistor. In the old beige/gray Jaguars it was possible to heat that component so much it would reflow that resistor and open the circuit which would leave the attached motor dead in the water.

So I'm a bit curious to see how they'll handle that. Maybe they'll increase the wattage of that resistor or decrease it's resistance to handle the higher currents the existing MOSFETs can easily handle. Maybe they'll short it out but then you can't measure current like that.

In any event this action makes good FIRST assurance they'll make sure these are available. That's great because it makes things much more clear for everyone involved. Plus now LinuxBoy's CAN terminators are for an active product with an audience.

[roystur44]

Originally Posted by JesseK

The real solution would be for teams to stop designing mechanisms that consistently hit the 40A + mark for extended periods -- which means more teams need to do the math.

Our 2010 robot Storm was geared to 17 ft/sec and we could hit top speed then we tried to stuff balls into the goals or crash into walls to pick balls up. No matter what we tried we reset those Jags(Can) every match till we replaced the drive controllers with Victors and ran PWM.

The following year our 2Can enable Jags just reset when they felt like it. We switched to PWM Jags and pushed on for the 2011 season. We put a lot of resources behind finding out the problems with the Jags and how we could program to use the Jag like we wanted.

2012 we decided not to lose matches because our jags were resetting so we used Victors.

So the moral of the story is if you want to make it fast run the motor till it almost blows up

[s1900ahon]

Originally Posted by andreboos

The post mentions a conformal coating - am I right to hope that this would reduce their vulnerability to stray metal shavings?

You are right to hope. Conformal coating covers the PCB and components insulating it from swarf (debris).

Ideally, you don't put swarf in the motor controller to begin with.. but..

[s1900ahon]

Originally Posted by 1075guy

Perhaps IFI used some of the knowledge gained in purchasing the rights to Jaguars to improve the 884's into the 888's.

IFI didn't purchase the rights to Jaguars. TI donated the rights to FIRST, and FIRST contracted IFI to produce them for 2013.

IFI independently improved the 888.

[Clinton Bolinger]

Has anyone else noticed the Part Number on IFI's Web Site, for the Jag?

Part #: 217-3367

Looks like it should be called the "Hot-Killer-Chicken" (217, 33, and 67)

Interesting.

-Clinton-

[Paul Copioli]

Originally Posted by Clinton Bolinger

Has anyone else noticed the Part Number on IFI's Web Site, for the Jag?

Part #: 217-3367

Looks like it should be called the "Hot-Killer-Chicken" (217, 33, and 67)

Interesting.

-Clinton-

VEX's director of product development, John V-Neun, has a strange sense of humor.

[Clinton Bolinger]

Originally Posted by Paul Copioli

VEX's director of product development, John V-Neun, has a strange sense of humor.

I like his style and sense of humor.

-Clinton-

[IndySam]

Originally Posted by Paul Copioli

VEX's director of product development, John V-Neun, is strange.

FYP

[Phyrxes]

*cheer* now to wait and see what the official ruling on our new options will be.

Thank you FIRST, TI, and IFI for doing this!

[rachelholladay]

Although the specifics on motor controllers won't be released in till kickoff (3 months, 1.5 days!) does anyone have comparison data on Talon vs the Victor vs Hot-Killer-Chicken (endorsement of JVN's humor) ? I'm looking ahead to motor controllers for next season and I'm trying to figure out the pros and cons of each for different subsystems.

[AlexH]

Originally Posted by rachelholladay

Although the specifics on motor controllers won't be released in till kickoff (3 months, 1.5 days!) does anyone have comparison data on Talon vs the Victor vs
Hot-Killer-Chicken (endorsement of JVN's humor) ? I'm looking ahead to motor controllers for next season and I'm trying to figure out the pros and cons of each for different subsystems.

(well assuming the continuous current of the talon was taken at 28v it should be good for 140a at 12v. vs the victor which will do 40a at 12v

you could pop the main breaker with a single talon.) ignore, incorrect information

also... THE TALON LOOKS LIKE A VICTOR WITH A HEATSINK!!!1!1!!!

[EricVanWyk]

Originally Posted by AlexH

well assuming the continuous current of the talon was taken at 28v it should be good for 140a at 12v. vs the victor which will do 40a at 12v

you could pop the main breaker with a single talon.


also... THE TALON LOOKS LIKE A VICTOR WITH A HEATSINK!!!1!1!!!

That is not at all how that math works. An H-bridge's current rating is very nearly flat across its voltage range.

Additionally, the victor will handle more than 40A at 12V.

[AlexH]

i thought it was (voltage1*amps1=total watts=voltage2*amps2)

but seeing as that doesn't apply to h-bridges....

[Ether]

Originally Posted by AlexH

i thought it was (voltage1*amps1=total watts=voltage2*amps2)

You were using the voltage delivered to the output (the motor) times the current delivered to the output. The power dissipated in the motor controller is not 28*60=1680 watts. It would fry.

[EricVanWyk]

Originally Posted by AlexH

i thought it was (voltage1*amps1=total watts=voltage2*amps2)

but seeing as that doesn't apply to h-bridges....

That formula is close to what you want, but you have it in a form usually used for power conversion.

Total watts is still volts * amps, but you are thinking of the wrong volts. The voltage that matters is the voltage developed across the bridge elements themselves, not the total voltage applied across the entire bridge.

The switching elements (FETs) have a small resistance when they are conducting. The voltage is then the product of the current times that resistance. You end up with Power = (Current)^2 * (Resistance).

This is only dependent on the motor current, and is unrelated to the applied voltage.

There are a few other heating effects related to that applied voltage, but they are largely dominated by the I^2R losses and can be ignored for a first pass glance at the numbers.