blog; Motor Controller Options for 2015

[Jon Stratis]

I have to say... opening up the beta box and seeing these last week was pretty exciting! It's a nice, small, integrated component that hopefully is just as reliable and durable as the previous speed controllers we've had to work with. I'm definitely looking forward to trying these out once we put the beta hardware on last year's robot in a few weeks (the robot is unfortunately sequestered at the State Fair until then).

[pastelpony]

Going to love using these. We used mostly Talon SRs on our competition bot this year because the Jaguars we were using were too big. These new ones make the Talon SRs look giant.

[s1900ahon]

Originally Posted by qnetjoe

My only question is what is the pinout for feedback pins on the SRX?

I only counted 10 pins on the feedback connector, but the Jag has 12 ( 5 - encoder, 3 Analog Input, 2x2 Limit Switches). I am sure a few pins are doubled up but I wonder which ones?

Yeah, but of the 12 pins on the Jaguar, 4 of those were grounds. This was done to try to make each interface easy to wire. Using a unified connector accomplishes the same thing (and an interesting alternative).

Let's speculate.. that's fun. What I suspect is we'll find there is a certain amount of pin multiplexing (peripheral signals that share the same pins; the firmware connecting a single peripheral to a pin based on configured usage). Very common on MCUs.

Assuming that a connection can be established across either the CAN, USART, or SPI connection for control, that needs 6 of the 10 pins on the connector be distinct. Add 2 more for power and ground. That leaves 2 signals for a sensor.

End speculation.

However it ends up, it will be well thought out I'm sure.

[sanddrag]

Well, this will certainly change the way we do wiring a little bit, but I'm sure we'll still come up with a way to make it look nice. I like that the mounting holes are in a line parallel with the case and a nice round 2.000" apart, rather than at opposite corners across a diagonal. I also like that we can pack a bajillion of these into a small area, and do not need to design our chassis around the width of a row of speed controls.

Also, I like that I don't have to practically write a whole grant just to put some speed controls on our robot.

Thank you VEX and CTRE!

[Nemo]

I like it. I hope these are the controllers for a nice string of years so we can stock up and reuse them for multiple years.

It will be a tough call when we decide what controllers to buy this year and how many. Cost of new stuff vs space savings.

[AK Koaster]

The only thing I'm slightly worried about on these is the heat dissipation: I'm not exactly the most knowledgeable on the subject, but looking at the size, I'm a bit worried whether or not there will be enough surface area on the controllers to adequately cool them down. In the documentation, it states that the fins will take care of passive cooling, but we've traditionally put fans on our speed controllers, even the Talons (we had a bunch of extra fans of that size, so we figured why not). Would like to talk to some teams early in the season after they've started working with them, see what their findings are. Active cooling for these might be a challenge, aside from putting a big computer fan over a network just for a little cooling, I don't see an extremely efficient way of doing it.

However, I'm still really excited, as these seem to be pretty well suited for FIRST in terms of price point, size (haven't used Jaguars in years specifically for that reason) and reliability during competition. Not working with screw terminals will be great as well, I always seem to lose those screws, and they're really hard to find once they roll away on the shop floor.

[AdamHeard]

Originally Posted by AK Koaster

The only thing I'm slightly worried about on these is the heat dissipation: I'm not exactly the most knowledgeable on the subject, but looking at the size, I'm a bit worried whether or not there will be enough surface area on the controllers to adequately cool them down. In the documentation, it states that the fins will take care of passive cooling, but we've traditionally put fans on our speed controllers, even the Talons (we had a bunch of extra fans of that size, so we figured why not). Would like to talk to some teams early in the season after they've started working with them, see what their findings are. Active cooling for these might be a challenge, aside from putting a big computer fan over a network just for a little cooling, I don't see an extremely efficient way of doing it.

However, I'm still really excited, as these seem to be pretty well suited for FIRST in terms of price point, size (haven't used Jaguars in years specifically for that reason) and reliability during competition. Not working with screw terminals will be great as well, I always seem to lose those screws, and they're really hard to find once they roll away on the shop floor.

I trust the parties involved when they say they are fine as is.

The talon worked fine without a fan and the thermal mass on it appeared to be much smaller for reference.

[Ether]

The body is made of aluminum, and it is isolated from the electronics inside. If you're concerned about cooling, rub some thermal paste on it and bolt it to an aluminum plate.

[cgmv123]

Originally Posted by Ether

If you're concerned about cooling, rub some thermal paste on it and bolt it to an aluminum plate.

Or don't risk violating the rule against modifying electronics and add a fan instead.

[Ether]

Originally Posted by cgmv123

Or don't risk violating the rule against modifying electronics and add a fan instead.

What rule did you have in mind, that prohibits mounting a motor controller using the provided mounting holes?

The thermal paste could be omitted, if that is the concern.

[cgmv123]

Originally Posted by Ether

The thermal paste could be omitted, if that is the concern.

Yes, that would be the concern.

[philso]

Originally Posted by AllenGregoryIV

I'm more worried about the PWM cables being attached. We've had far more PWM wires go bad over the years than 12AWG wire. It's going to be a little more annoying to troubleshoot since you can't just swap the PWM wires around at the speed controller. However I am glad that they all come with wires and I would hope they are nice high quality cable and connectors.

One of my concerns is that the PWM/CAN wires enters the enclosure within a mm or two of one of the input power wires in both the Talon SRX and the Victor SP. It will likely lead a lot of people to route the PWM/CAN wires adjacent to the input power wires for some distance. This arrangement makes it much more likely that the PWM/CAN wires pick up the noise from the switching action of the output transistors. It is likely that they have been able to use a smaller enclosure by using newer MOSFETs that have lower on-resistance (for lower conduction losses) and/or shorter rise and fall times (for lower switching losses). Shorter rise and fall times are likely to cause stronger inductive coupling of the switching noise into signal wires running in parallel with the power wires.

The integrated PWM wires will probably mean that more teams will have to purchase the pins and crimpers to replace/repair the connectors when they inevitably wear out or get damaged.

I am hoping that the power wires are truly a "high strand count" type (28 or 30 AWG strands instead of 25 AWG strands) with the soft silicone insulation. It would make it so much easier to do a neat job of running the power wires (VEXPro, are you listening?)

[RyanCahoon]

Originally Posted by philso

One of my concerns is that the PWM/CAN wires enters the enclosure within a mm or two of one of the input power wires in both the Talon SRX and the Victor SP.

The SRX interface drawing lists the CAN leads as being twisted pair, which should help some (though I'm not sure how much).

---

I hope there's an allowance in the rules for year-to-year reuse of motor controllers that have had connectors attached to them.

[donkehote]

Originally Posted by cgmv123

Yes, that would be the concern.

What rule disallows thermal paste? Sticking chemicals on the outside of electronics is modification? Would the same rule disallow Velcro attachment, or labels being applied to speed controllers? Both of those use chemical compounds applied to the heat sink causing different operating conditions.
Your post reminds me of the bumper pool noodle tape issue from last year.

Some things don't need regulating IMHO.

[mman1506]

Originally Posted by philso

The integrated PWM wires will probably mean that more teams will have to purchase the pins and crimpers to replace/repair the connectors when they inevitably wear out or get damaged.

Or cut an PWM extension in half and solder a new connector on. All teams should have soldering supplies.

Integrated PWM wires are very very common on RC multi-rotors, planes, cars .etc that experience quite a lot of wear. Multi-rotors particularly depend on high speed control loops where noise would be noticeable.