Re: Banebots RS-775 Case Short
 

Joe,
Anything less than 10K is likely going to get the Crio upset at some point. Don't forget that the voltage polarity on the case will reverse when the motor direction reverses. It will also have the speed controller switching component and any brush noise that might be induced as well. You monitoring needs to compensate for high noise and 15kHz for the Jag and 150Hz for the Victors.

Re: Banebots RS-775 Case Short
 

While this is true for the relative polarity between the two motor leads, it is not so for the can. The can will be shorted to one or other of the brushes*. The voltage on the can will then be pulled hard to Vbatt when that bit of the winding touches whatever side the Jaguar is holding to Vbatt and it will be pulled hard to Gnd when the motor turns such that that same portion of the commutator is touching the side the Jaguar is holding to Gnd.

As to noise and Jag PWMing, I think this the least of our concerns. A simple filter should make that bit go away.

As to the cRio getting upset with lower than 10K, I don't see how this comes into the picture.

Some ascii art:
Vbatt
...|
100 Ohm
...|
Motor Case -- RC filter (band pass) -- cRio Analog Input
...|
100 Ohm
...|
Gnd


The band pass should remove the DC offset of 1/2 Vbatt, allow the rail to rail swings (say 1200-18000 RPM motor speed so allow 20-300Hz or so), and remove the Jaguar PWM and other junk (1000Hz and above).

I am not a EE but it seems that there may be some room for a solution in that space.

Joe J.


*well, it will be shorted to both because the motor windings connect the two, but odds are the short will happen with the wire length to one brush being much shorter than the other so it will act like a short to that one brush.

Re: Banebots RS-775 Case Short
 

Why worry about it when the robot is in action? You can't stop the game and do anything about it anyway.

Monitor the resistance between matches, change them out if it reaches a value bothers you.

Re: Banebots RS-775 Case Short
 

If power to the failing motor is cut then it may save a cRIO reboot and thus the robot will remain mobile for the match instead of being dead in the water for 30-60s.

Re: Banebots RS-775 Case Short
 

If the 775 is completely electrically isolated (via both the case and output shaft) through plastic mounting plates and either acetal gears or timing belts, will the cRIO ever actually reboot if a motor starts to go bad?

Re: Banebots RS-775 Case Short
 

I recall somewhere way back in this thread that at least one team had an issue, in rare instances, with a pair of 775s shorting together , instead of through the frame. The current draw would tank the robot's voltage and cause a reset. This may not be an issue depending on your exact setup.

Joe said "I think that isolating the case (from anything, including other BB775s) will limit the damage to that one motor." which might be an issue if multiple motors are failing/failed and short together.

I guess I skipped a few details in my thought process :p

Re: Banebots RS-775 Case Short
 

Joe,
The Crio chassis is tied to the negative lead of the battery. The power lead to the Crio is not sufficiently low impedance to prevent (nor is the boost regulator capable of overcoming) a disturbance in the power to the Crio. If anything on the Crio finds a frame fault including the Crio chassis, considerable current (not full motor current of course) will then flow on that negative wire. It appears that in the motors opened last year, the majority of shorts occurred between motor windings and the armature. How close to one end of the winding that fault occurred is anyone's guess. A few appeared to be stray solder that migrated between a brush assy and the case. Another reason for concern is the possibility of two robots with electrical faults becoming engaged with current flowing between frames.
While it appears that the Crio can operate at less than 19 volts, the spec is to operate above 19 volts input.

Re: Banebots RS-775 Case Short
 

We are not using the Banebots 775 motors.

Re: Banebots RS-775 Case Short
 

I thought the chassis of the cRio was tied to 24V.

Even if on one robot the chassis is connected to 12v and on the other it is connected to gnd, nothing would happen if these two robots touched each other. Sure, you'd be able to read 24v between the (-) terminal of the first robot and the (+) of the second if you stopped the match and walked out there with a voltmeter, but there is no complete circuit for current to flow through. Both the (+) and (-) of the same battery would have to be connected the different parts of the chassis and/or frame for any current to flow.

Re: Banebots RS-775 Case Short
 

If you don't mind me asking, who is "we"? If 254 isn't using them, that really says something, because I know they do their homework when it comes to motors.

Re: Banebots RS-775 Case Short
 

254. I'm pretty good about updating my current team info.

Re: Banebots RS-775 Case Short
 

This is actually one of most common myths in FRC, perpetuated in a blue box in the rules each year. The real reason we isolate our frame is to allow the system to survive single-fault wiring failures.

This is similar to the way the PD's prototype big brother detected chassis faults, except we pushed an active signal instead of using a resistor divider. The control system would be fine with 100 Ohms, but that'll burn three fourths of a watt. I'd bump up your impedances into the 5-20k range. You're signal quality will be unaffected but it can be smaller and safer.


No. A fully isolated but faulty 775 will not have an adverse impact on any portion of the control system electronics. I do not have any information as to whether or not this fault will have an adverse impact on the motor itself.


To the negative lead through a fuse and a self-resetting fuse. If bonus current is sent through this path, the two race to see who will blow first. The single-use fuse rarely wins, but can go with repeated abuse.

True. This is another myth that used to be in a blue box.

Re: Banebots RS-775 Case Short
 

Erik,
I have to disagree. While the fuse internal to the Crio is meant to protect parts of the internal Crio electronics, electrical faults on the robot can cause real voltage drops across the #18 wiring used to feed power to the Crio. This is well documented over the past two years irrespective of motor issues. With these faults current merely flows through the chassis of the Crio to the negative lead and out the power connector.
Isolated (electrically) frames have been part of the rules to prevent robot to robot electrical faults from causing sparks and fire on the playing field. The potential in this scenario is the ability to provide up to 24 volts at high current, through parts of the frame and any wiring that might be in the circuit. Anyone who has been around more than a few years remembers the spectacular displays, smoke, hot (incandescent) wiring and yes even some flame that occurred prior to the inclusion of this rule.
Jason, the common myth is that robot to robot contact usually includes not only the robot frame but appendages or robot parts when tipped over, which have reached inside another robot. This is the second or third path that most people forget. In some cases, current will continue to flow even after the robot is disabled and the main breaker has been opened. It is why inspectors are so critical when checking the insulation on all electrical connections.

Re: Banebots RS-775 Case Short
 

Al, Eric's post was rather long, and it's unclear which part of it you are disagreeing with. Would you please quote the portion you are disagreeing with?

Re: Banebots RS-775 Case Short
 

I could only disagree with Eric's response to my quote and answer Jason's question.