Circuit Breakers

[Al Skierkiewicz]

No, all circuit breakers exhibit this phenomena. Consider yourself really lucky...so far. Remember though, the circuit breakers are most sensitive to side impact, i.e. where the wires enter and exit.
Good Luck

[Ian Mackenzie]

Originally posted by Al Skierkiewicz
On the subject of the 60 amp breakers, I do not believe that any team can trip this breaker on current overload.

Unfortunately, I think it is quite clear that our robot DID trip our breaker multiple times solely on current overload. We were running six-motor drive, and we continued to trip the breaker even after gearing down, providing better ventilation to the breakers, fixing some faulty wiring, and allowing the breaker to be supported in mid-air by its own wires only (and cushioned by surgical tubing in case it moved too far).

Also, our breaker tripped mainly during sustained pushing matches, such as continuous low current being used to keep two goals stationary in our scoring zone. We eventually had to make the painful decision to disengage the drills from our mobility gearboxes, which has so far solved the tripping problem but has noticeably reduced speed and torque.

What seemed to happen sometimes is that one motor would trip its own 30 amp breaker, and while that motor was not receiving power the other two would be driving it, increasing the load on them and exacerbating the problem.

-Ian Mackenzie
Team 188
Woburn Robotics

[Al Skierkiewicz]

Ian,
Are you coming to Midwest regional? I would really like to take a look at your robot and talk to you about the tripping. You kind of imply that one of the 30 amp circuit breakers tripped out and stayed that way. The 30 amp breakers reset pretty quickly and on a sustained fault have been known to buzz as they repeatedly trip and reset. You did not mention if your controller was rebooting, did that happen as well? Let me know.
Good Luck All

[Greg McCoy]

Yeah, and which motors are tripping the 30A breakers?

[Ian Mackenzie]

Al,

First of all, no, we will not be at the Midwest regional, although we will be at the nationals (and by that time we will have competed in the Canadian Regional, during which we might discover something else). I confess that I don't know which motors were tripping the 30 amp breakers; in the heat of competition, we believed that it was better to make a drastic change (taking out two motors) that would definitely solve the problem than to take our time investigating and fine-tuning individual breakers and speed controllers. The latter seemed to be, as we say, "building to tolerance". Actually, I couldn't confidently say that the breakers were tripping at all; we never really got conclusive evidence of it, so we may very well have been grasping at straws.

I don't believe our controller lost power, because I never remember the light turning off temporarily (I have seen the controller shut off during practice back in Canada, usually when our battery was running low, so I think I would have recognized it if it happened in competition).

-Ian Mackenzie
Team 188
Woburn Robotics

[Al Skierkiewicz]

Ian,
Good luck at the regional and let us know if you do find anything there. Otherwise, see you in Florida. Stop by and say HI
Al

[Greg McCoy]

Ditto

[Ian Mackenzie]

I seem to remember seeing someone post something about large spikes of power tripping the breaker magnetically, not thermally...does anyone know more about this? And if so, since we are allowed unlimited permanent magnets, could a few magnets be strategically placed around the breaker to alleviate the problem?

-Ian Mackenzie
Woburn Robotics
www.team188.com

[Al Skierkiewicz]

Ian,
It is doubtful that magnets will help with the trip charachteristics. Effectively there is a little electromagnet inside the breaker. When the current gets to the trip point the magnet attracts an armature that is connected to a release mechanism, which lets go and the spring in the breaker pulls it open. Adding magnets may even make it trip at a lower current. Remember, these breakers are designed to be stacked up next to each other inside an electrical panel so stray magnetic fields are present all the time.
Good Luck

[Mike Rozar]

With this years game being a contact sport, we prepared for this. We obtained a piece of 35mm DIN rail (legal - extruded aluminum) which the breaker snaps onto. We took two of the shock mounts which were supposed to be used for the compressor to mount the DIN rail to the robot frame. The #6 wires are not tied down near the breaker. This is very forgiving.

We have been hit hard and only have dents but no breaker trips.

Hope this helps, (looking for pics)
Mike
See our robot at: http://www.chiefdelphi.com/forums/sh...&threadid=3479

[Al Skierkiewicz]

With all of the input on this subject and observing robots firsthand, I still believe that the majority of breaker trips are impact but I now think overcurrent trips are a possibility with the other variables in place. Although the spec sheets provided in FIRST documentation do not show trip derating curves, the Square D website does at...
http://www.squared.com/us/products/c...5661300555d9d/$FILE/2496.pdf
The trip current is predictable and decreases with high current over time. In looking at the curves, a 60 amp breaker can be expected to withstand six times it's rated current for a few seconds but that rapidly falls if the overcurrent remains. This is due to heating of the thermal parts within the breaker. If a robot design is such that the breaker receives heat during operation, you would expect that the trip point would decrease rapidly as well. In many of the robots that I have observed, the breakers were enclosed within the body of the robot, wiring to and from the breaker was very short or the breaker was mounted near heat generating components such as motors, transmissions, speed controllers and circuit breaker panels. In the majority of teams that reported tripping the 60 amp breaker they were also tripping the 30 and 20 amp breakers. These breakers (burned fingers to prove it) get very hot if repeatedly tripped. Add to that heat generated in loose contacts (such as the screw terminals on the breaker or the battery connector) and a serious problem occurs. Fudging the numbers and looking at the curves will give you a headache but you can guess that a breaker running about 30C and 180 amps non-continuous current will trip anywhere from 50 to 75 seconds into a match without drawing the battery below 8 volts. Some teams have reported freeze spray lets them run a longer period before trip. The reality is anything you can do to weigh the variables of temp, vibration and current on your side will help. Check the temp of the breaker, the #6 wire, and area surrounding these components when the match ends. Try using software to slow down the motors that draw high current. By changing the duty cycle of the heating current, either in software or driver training, you may be able to get a full two minutes of match time with only a small loss in speed or power. Check out the other posts about efficiencies in transmission design and drive system losses.
Please post any other input you might have on this subject.
Good Luck All and See You in Florida.

[Stephen Sexton]

I did a little analysis on the circuit breaker. Please refer to attached x-ray analysis. The top picture is a control sample in the the "on" position prior to an inpact. Main areas of interest are a) on and off switch, b) thermal unit and c) breaks contacts. Not labeled on the photo is the moving component at 1 oclock on the yellow circle.

The sample was then subjected to an impact which popped the breaker.

It appears the "thermal unit" moves when the impact happens, which is followed by moving component swinging down to the right and pop............... end of match.

Conclusion

Mount the circuit breaker with "ON" position down and "OFF" position upwards. With this oreintation the thermal unit's gravity works in our favor.

Stevo

[Al Skierkiewicz]

Stephan,
Great Photo, it helps a lot. I had a team open one for me and I think what you refer to as the thermal unit is actually the magnetic unit. I think the thermal is at the top of that assembly, there is a factory adjustment screw at the top accessable behind a label on the breaker. (There was no identification of the parts on the Square D site so I am guessing here.) It is easy to see that the two parts visible in the yellow circle will be forced together with impact on one side, tripping the breaker. Hitting on the opposite side keeps them apart.
Please note that any heat conducted into the breaker through the wires, (especially the left connector) is tied directly to the thermal unit through the internal metal parts.
BTW, we mount the circuit breaker horizontal, the same orientation it was designed for with the #6 wires used to shock mount from the frame.

Hope this helps give teams a clearer view.
Good Luck All

[Lloyd Burns]

We took a similar 40 A Square D breaker apart to examine the innards: the "thermal unit" in the Xray has a frame (in a 'J' shape) (call it 'J') attached to the wire terminal, and the thick piece (call it 'A'), just beside 'J' going into the "C"-shaped bottom of the 'J', carries a recess into which the long piece from the handle fits. 'A' is springy where is it welded to the end of the 'J' near the terminal, and swings toward 'J' with any force applied. When the thin loop hanging from the 'C' piece, which is bimetallic, starts bending (at its end near the 'C') it bends, and because it is attached to 'A' it pulls on the 'C', which pulls 'A' and the recess away from the handle-piece.

When rapped smartly along the long axis of the body, the cold breaker snaps open.

The connection from the breaking contact on the side near the other terminal, is made by a piece of thick stranded wire welded at both ends. We saw no evidence of any magnetic coils, nor of any mechanism to trip magnetically on theflow through the conductors as arranged.