Keeping the main breaker from tripping

[Aren_Hill]

Originally Posted by Cory

Generally breakers have 2 methods of protection. If you have a slow increase of current until you get to a continuous amount that will trip the breaker, the trip occurs via a piece of metal bending as it heats up until it no longer makes contact with the circuit.

If you were to have a short you'd see a huge sudden surge of current, which would cause a solenoid to pull the contacts apart.

I assume the one we use features both methods, but I've never looked inside one before.

The internal construction is actually a spring steel plate slightly warped to naturally press against a contact point. The off button springs this up (the plate bending is the resistance you feel), and lets the lever you flip to close the breaker swing between the plate and its contact.

So its performance is entirely thermal, warm that plate up and it will trip, do it fast or slow, doesn't matter, hit a certain temp and it pops (buckles) and when it buckles the "on" lever isolates the plate and its contact, leaving your robot off.

(Opened one up a few weeks ago to satisfy my curiosity...)

-Aren

[Al Skierkiewicz]

Guys,
FIRST hasn't ruled against the canned air because it really doesn't do that much. The thermal device is relatively isolated from the sealed case. You cool the case down but it doesn't reduce the temperature of the bimetal strip enough to change the characteristics. As Aren has pointed out, this is a simple and small piece of metal. It will heat or cool rapidly without intervention from the outside world. That is what it is designed to do. The current flowing through the contacts have the greatest effect. To a lesser extent, poor connections to the main breaker cause heat to be generated and conducted to the contacts through the metal carrying currents in the breaker. However, the thermal mass of the wiring and contact terminals on the breaker are sufficiently high that one would have to cool down the entire main electrical system to have the desired effect. No matter how you cut it, six CIMs are still drawing the maximum current from the battery in stall at full throttle. Teams that claim they are not tripping the main breaker have learned to ramp up to full throttle or prevent drivers from doing so in a match. In addition to the dangers of inhalation, the rapid cooling in a humid environment produces moisture on electrical parts and the chemicals tend to wash out lubricants. Your mileage will surely vary.

[FrankJ]

The copper lugs going into the breaker are great conductors of heat. Chill them & it will affect the internal workings of the breaker. But when you stop & the breaker will quickly go back to working as designed. So it really is not an issue unless you have someone chasing the robot on the field with canned of air.

On a non FRC but related subject, the NEMA thermal motor overloads work on the same principle heating an element. The overload element is actually called a heater. In some styles the element actually melts in a controlled way. Anyway to properly size them for the general case you need to take into account the ambient temperature of the controller relative to the motor & they still need short circuit protection.

[Chadfrom308]

For current readings, we are using the IC from this board:
http://www.panucatt.com/Current_Sens..._p/cs-100a.htm

I don't use this for FIRST, but on the Formula SAE team I am on, we are using this to read the alternator current and the total current output by the PDB.

I don't know exactly if it is legal to put this in between the board and breaker, in fact, I'm pretty sure it's not. Also, we use either a 150a or 200a version of this, I cannot remember. It is like $17 minimum order for these of digikey (I think, I didn't order them)

[cgmv123]

Originally Posted by Chadfrom308

For current readings, we are using the IC from this board:
http://www.panucatt.com/Current_Sens..._p/cs-100a.htm

I don't use this for FIRST, but on the Formula SAE team I am on, we are using this to read the alternator current and the total current output by the PDB.

I don't know exactly if it is legal to put this in between the board and breaker, in fact, I'm pretty sure it's not. Also, we use either a 150a or 200a version of this, I cannot remember. It is like $17 minimum order for these of digikey (I think, I didn't order them)

Current monitoring like this seems to be what the second sentence of 2014 R53 has in mind.

[mman1506]

Originally Posted by Al Skierkiewicz

Guys,
FIRST hasn't ruled against the canned air because it really doesn't do that much. The thermal device is relatively isolated from the sealed case. You cool the case down but it doesn't reduce the temperature of the bimetal strip enough to change the characteristics. As Aren has pointed out, this is a simple and small piece of metal. It will heat or cool rapidly without intervention from the outside world. That is what it is designed to do. The current flowing through the contacts have the greatest effect. To a lesser extent, poor connections to the main breaker cause heat to be generated and conducted to the contacts through the metal carrying currents in the breaker. However, the thermal mass of the wiring and contact terminals on the breaker are sufficiently high that one would have to cool down the entire main electrical system to have the desired effect. No matter how you cut it, six CIMs are still drawing the maximum current from the battery in stall at full throttle. Teams that claim they are not tripping the main breaker have learned to ramp up to full throttle or prevent drivers from doing so in a match. In addition to the dangers of inhalation, the rapid cooling in a humid environment produces moisture on electrical parts and the chemicals tend to wash out lubricants. Your mileage will surely vary.

Originally Posted by Cory

To clarify, I don't think anyone cools their breaker to keep alive...they (or at least we) just do so in order to not start with an elevated temperature due to short match turnaround. An elevated temperature means your breaker will trip under lower current draw than a room temperature breaker.

Teams that claim they are not tripping the main breaker have learned to ramp up to full throttle or prevent drivers from doing so in a match.

We don't, I'd bet 254 doesn't either.

[Chadfrom308]

Originally Posted by cgmv123

Current monitoring like this seems to be what the second sentence of 2014 R53 has in mind.

Sweet! I didn't have the rules handy, at the time. I will definitely think about putting these on the robot next year

[Joe Ross]

Originally Posted by Chadfrom308

For current readings, we are using the IC from this board:
http://www.panucatt.com/Current_Sens..._p/cs-100a.htm

I think it's likely that you would exceed 200 amps for short periods of time. We measured total currents in excess of 300 amps with the 2015 PDP. http://www.chiefdelphi.com/forums/sh...&postcount=153

[Jared Russell]

Keep in mind that not all main breakers are created equal. The spec sheet has pretty large error bars on tripping current vs. time and tripping current vs. temperature (close to a 20% swing). This could easily be the difference between tripping a breaker and not tripping a breaker in a match. Many teams opt to replace a main breaker once it has been tripped. There are two reasons to do so:

1) Anecdotally, teams claim to have observed that a breaker may fatigue after repeated tripping (though this is not that likely until there have been dozens of cycles). Would love to see some actual data behind this.

2) A breaker that has tripped is statistically more likely to be from the less forgiving end of the main breaker distribution. By replacing a breaker every time it trips, you are likely to end up with a more robust breaker in your robot for most of your matches.

A few other quick thoughts on this topic:

* When driven fluidly (whether through practice or software help), your drive draws less current and preserves more momentum when maneuvering.

* Well known teams with 6 CIM drives still occasionally pop breakers. Maximizing performance is all about balancing on a knife edge, and occasionally things go the wrong way. Attention to every little detail (wiring, mechanical efficiency, software, pre- and post-match maintenance) is necessary to minimize the frequency of problems, and most teams would be better off being more conservative in their gearing and motor allocation.

* At some point, browning out your control system becomes the limiting factor in power draw. The RoboRIO doesn't like temporary low voltage spikes any more than the cRIO did, and in some ways it is less forgiving (ex. disabling PWM outputs at around 6.5V).

[hrench]

I agree with the people saying that canned-air cooling wouldn't last, but I wonder if anyone's ever considered mounting the breaker in a 'refrigerator' made with a 12volt peltier cooler? If you don't know what these are, here's one:

http://www.adafruit.com/products/133...FYKPMgodZhwAiQ


I'm nearly sure there will be some rule that negates this possibility and giving up 5 amps of battery isn't so good either, but you could shut it off under high driving current-draws.

[Monochron]

Originally Posted by hrench

I agree with the people saying that canned-air cooling wouldn't last, but I wonder if anyone's ever considered mounting the breaker in a 'refrigerator' made with a 12volt peltier cooler? If you don't know what these are, here's one:

http://www.adafruit.com/products/133...FYKPMgodZhwAiQ


I'm nearly sure there will be some rule that negates this possibility and giving up 5 amps of battery isn't so good either, but you could shut it off under high driving current-draws.

I think that would be actually something dangerous to do because it is altering the breakers operating parameters rather than just resetting it to normal operation after a match.

[Jared]

Originally Posted by Cory

By cooling the breaker you merely restore it to its rated current level. You don't get "bonus" current. You just aren't de-rating it due to it being hot (since it is a thermal switch).

By extension, you are against fans on motors?

I am not against fans on motors. I am also not against cooling the breaker. Our team cooled off our breaker with canned air too. I don't think it's any more dangerous than a lot of the other stuff we do. By my interpretation, the breaker spray is illegal, but FIRST doesn't think it is illegal. I have no problem with spraying the breaker, so I am more than happy to go with FIRST's interpretation.

The fans on motors comment isn't a good comparison. Motors don't operate based off of heat like the breakers do and fans are specifically called out as being allowed. Fans also don't involve spraying a dangerous liquid chemical on electronics, and don't involve dangerously cold temperatures.

[brycen66]

Originally Posted by inkling16

Why not just sum the currents at the outputs to get the current at the input?

Personal preference. Both approaches are valid I just wanted to bring up that the input approach does exist and is what I would use.

[mman1506]

Originally Posted by hrench

I agree with the people saying that canned-air cooling wouldn't last, but I wonder if anyone's ever considered mounting the breaker in a 'refrigerator' made with a 12volt peltier cooler? If you don't know what these are, here's one:

http://www.adafruit.com/products/133...FYKPMgodZhwAiQ


I'm nearly sure there will be some rule that negates this possibility and giving up 5 amps of battery isn't so good either, but you could shut it off under high driving current-draws.

Considering the efficiency of a peltier (5-6%) it won't do much.

[brycen66]

Originally Posted by Ether

Please post details?

Place a steel washer or ring(any ferrous metal works) around one of your main wires. Then cut a slot in a steel washer and place a linear hall sensor in the slot. We epoxied ours in to keep it intact. The hall sensor outputs voltage which is directly related to the total robot current, and it can be plugged directly into the analog breakout. This is the system we have implemented into our practice robot and the results so far look good.
http://www.jameco.com/webapp/wcs/sto...001_1915940_-1