New Speed Controller Announced

[artdutra04]

Originally Posted by Cory

It won't be so awesome for the teams who will have thousands of dollars in effectively useless Victors once FIRST transitions fully into the new control system.

eBay! Just because Victor 884s might not be allowed in FRC in future seasons would not mean they've lost all use. They are really good speed controllers, and there are always other people (especially Battlebots community) who would buy them.

Originally Posted by Cory

I'm extremely concerned with the size of all the components of the new control system. All these pieces together are absolutely massive compared to what we've been using.

Gone are the days of giving the electrical and controls people a cubic foot of space and telling them to make it work...

[ChrisH]

Originally Posted by artdutra04

Gone are the days of giving the electrical and controls people a cubic foot of space and telling them to make it work...

Obviously your mechanical people have been spoiling you. They have been giving you far too much volume

[dpraedan]

I would be inclined to say that these controllers use the standard R/C PWM wiring scheme the Victors employ, seeing as a) the digital sidecar for the new system has standard PWM outputs and b) the PDF explicitly states it uses the standard R/C PWM servo interface.

[tdlrali]

Since when is big the new small? First the radios, then the maxi breaker panels & distribution blocks, then the new control system, and now the speed controllers...

[Al Skierkiewicz]

I have been estimating 6 lbs and about 1/2 cubic ft. for the control system alone. You can back into the size from the picture knowing that the pin spacing on the header is 0.1 inch or judging the two RJ 45 jacks. I am guessing that it is at least 1/2" wider and 1" longer than Victors.
16kHz is still audible to students and I can tell you from experience that 15,734 Hz is annoying to anyone under 50 and some over 50. I question the choice of switching frequency vs. the inductance of the motor windings though. Testing will confirm this.

[R.C.]

Maybe more teams will use the old Victors, they are still available.

[Joe Johnson]

Originally Posted by Al Skierkiewicz

16kHz is still audible to students and I can tell you from experience that 15,734 Hz is annoying to anyone under 50 and some over 50. I question the choice of switching frequency vs. the inductance of the motor windings though. Testing will confirm this.

The flyer (pdf) I read said 20kHz. Am I mistaken?

As to claims of linearity, that depends on what you mean. The output may actually be very linear, but the system need not respond in a linear way.

My experience in automotive systems is that when you go from low PWM freq (100-400Hz) to higher PWM freq (10-20kHz), the motors do not respond the same, especially at low duty cycles.

For example, if, in order to get a system moving, we needed say 20% duty cycle at 100Hz, we would need to bump the duty cycle up to 30-40% at 10kHz PWM freq.

This was very puzzling but we eventually attributed this our mechanical system's time constant. At 100Hz, the mechanical system's time constant was such that the system had time to react during a single PWM pulse -- These "full on" pulses provided a kick to get the motors and gears turning. At 10kHz, the system can only react to the average not the individual pulses, so it required more on time to get things rolling.

It does not say in the flyer but I hope that the H-bridge is configurable in ways that are feedback loop friendly. What I mean is that you can drive motors via an H bridge in several ways.

1. ON = 12V, OFF = Open Circuit
2. ON = 12V, OFF = MOTOR LEADS SHORTED
3. ON = 12V, OFF = -12V

#2 is very nice for feedback control of large arms for example.
#3 sounds very strange at first, but it can provide some very nice control features as well (thought it can be tough on the electronics) in this case

• "OFF" for the motor is accomplished by outputting 50% 12V & 50% -12V.
• "50% Forward" is accomplished by outputting 75% 12V & 25% -12V.
• "25% Reverse" is accomplished by outputting 38% 12V & 42% -12V.

As I said, it seems strange but it has some very nice behaviors. This type of control is called locked anti-phase motor control and I discuss this issue at length here

Once they implement CAN, I hope that they also consider some other control friendly strategies:

• using current messurements to allow for velocity (and even position) control without the use of encoders or other sensors. I discuss this here: What Dr Joe Wants for Christmas... (Man, I love referencing myself ;-).
• implementing PID control onboard the speed controller (send it your gains and from then on just give it a desired position -- let IT figure out what the PWM outputs should be)
• Diagnostics, diagnostics, diagnostics. At a minimum teams will know if a speed controller has become unplugged -- I guess that feature alone would have changed the outcome of about 20% of the FIRST tournaments ever held. But open your mind a bit and you can imagine diagnostics that tell you if your motor is overheating, if your mechanism is binding up, if your wheel treads need to be replaced, etc. etc. The mind reels at the possibilities.

It is only August, but it is looking more and more like Christmas to me...

Joe J.

[dlavery]

I notice that this press release comes from Luminary Micro, and not from FIRST. They are bold enough to announce that they "will be included in the FIRST Robotics Competition's (FRC) official Kit of Parts (KoP) and distributed to over 1,700 teams in the upcoming 2009 international FRC." This "announcement" has not yet been acknowledged or confirmed by FIRST. But I have heard rumors that the content for the 2009 KOP, and the associated rules about what is legal to use on the robots, have not yet been finalized. Therefore, I would surmise that this announcement may be a bit premature.

I have heard tales of some members of the GDC that are devious enough to pull an item out of the KOP specifically because the community got unauthorized advance knowledge of a part, and started to do pre-design and pre-build work before the season officially started. After all, wouldn't it be interesting if the entire community got all spun up about a rumored new "capability" that was then never actually included in the KOP? That would be an excellent way to divert their attention away from what was actually going to happen.

Nah, something like that would never happen.

-dave



.

[AdamHeard]

I weary of the new system just because it is new, but a lot of these features are very interesting to me. I'm really interested to know more about the sensor feedback.

Originally Posted by dlavery

I notice that this press release comes from Luminary Micro, and not from FIRST. They are bold enough to announce that they "will be included in the FIRST Robotics Competition's (FRC) official Kit of Parts (KoP) and distributed to over 1,700 teams in the upcoming 2009 international FRC." This "announcement" has not yet been acknowledged or confirmed by FIRST. But I have heard rumors that the content for the 2009 KOP, and the associated rules about what is legal to use on the robots, have not yet been finalized. Therefore, I would surmise that this announcement may be a bit premature.

I have heard tales of some members of the GDC that are devious enough to pull an item out of the KOP specifically because the community got unauthorized advance knowledge of a part, and started to do pre-design and pre-build work before the season officially started. After all, wouldn't it be interesting if the entire community got all spun up about a rumored new "capability" that was then never actually included in the KOP? That would be an excellent way to divert their attention away from what was actually going to happen.

Nah, something like that would never happen.

-dave



.

Brandon, can you ban his account? Please....?

[Al Skierkiewicz]

Joe,
I was referencing a previous post with the 16 kHz. At this and higher frequencies, the combined inductance and series resistance will produce a low pass filter that will serve to limit rise times and produce triangle waves instead of the the expected square waves. Glad to see you back on the forum.
Perhaps Richard, if you are reading this, can give us a handle on motor winding inductnce. I would think it would be pretty high, on the order of 100 mH. I won't be able to do any measurements for several days.

[sanddrag]

While the new control system (including these speed controls) is more capable than the old one, the increase in physical size is a lot. We would not be able to comfortably fit these components on any of our previous four robots.

[Joe Ross]

Originally Posted by Joe Johnson

As to claims of linearity, that depends on what you mean. The output may actually be very linear, but the system need not respond in a linear way.

My experience in automotive systems is that when you go from low PWM freq (100-400Hz) to higher PWM freq (10-20kHz), the motors do not respond the same, especially at low duty cycles.

For example, if, in order to get a system moving, we needed say 20% duty cycle at 100Hz, we would need to bump the duty cycle up to 30-40% at 10kHz PWM freq.

This was very puzzling but we eventually attributed this our mechanical system's time constant. At 100Hz, the mechanical system's time constant was such that the system had time to react during a single PWM pulse -- These "full on" pulses provided a kick to get the motors and gears turning. At 10kHz, the system can only react to the average not the individual pulses, so it required more on time to get things rolling.

Through the private beta test channels, I asked these types of questions. They just released some test data publicly, which show a very nice linear response without any slow starting when driving a CIM motor. http://forums.usfirst.org/showthread.php?t=10182

[Qbranch]

When they announced the '09 controller, I didn't mind letting somebody else fiddle with it.

But these motor controllers... if they do everything they say they do well, they should be a knock out. I've been working with/around CANbus for about three years now and it's always been very stable and VERY robust (you can, afterall, short the bus completely out and nothing will smoke).

They have this unit at the NI Week convention in texas. I asked my remote observation unit (aka my dad) to go take a look at this new speed controller and see what he could find... it ended up he handed the phone to the guy from luminary so, just two minutes after I first read about it here, here I was talking to the guy who makes em! Funny ol' world, isn't it.

Anyhow, they don't have all the specs nailed down yet, but so far they do know the speed controllers take an input of up to FOUR MEGAHERTZ (meaning, if you put a 1024 count encoder on a CIM motor's shaft, it could run at it's free speed and the processor wouldn't bat an eyelash) on the quadrature encoder input.

Also, the chop rate on the drive's output is 16KHz so no more angry buzzing first robots, now we'll have a nice whine that's way up at the top of we human's audible range (could be annoying, but probably will be barely detectable. 32-bit resolution over the duty cycle range from full on to full off, linear accross the entire range with 1-bit resolution (meaning minimum throttle is 0.00000002% instead of 10%).

Oh how I wish we had these speed controllers this year! Our autonomous would have been SO much smoother!

-q

[RyanN]

When I was at the Championship, I talked with the people extensively about the new control system. They said that there was a very good chance that we will still be able to use Victors on our robot.