Our team has encountered a most unusual problem: during a match, only when plugged into a competition switch, one side of our wheel base stops working. It does not happen during all matches. As far as I can tell, it happens randomly- but ONLY when plugged into a field or competition switch. It isn’t an overheating problem; our wheel base consists of four brand-new high-power motors with internal torque gearing and we have intentionally strained the robot when not plugged into a field and it had no such issue.
Our code is the standard Competition Template, and the software experts in Michigan State University’s robotics program have told us they don’t know of any code problem that could stop only two motors. Our lift motors have never done this.
Our wheel motors are wired through a power expander. We know there is no problem with the motors or their wires, because we had the problem at our first tournament with older motors and before our second tournament we replaced them with new ones. We think the power expander may be to blame.
Have any other teams had this problem? If so, what was the cause, and how did you fix it?
It would help if you could post more comprehensive details as I outlined in this post yesterday.
It’s almost impossible that the field control or competition switch can affect only two motors. Have you run the robot on the field tiles outside of a competition? On several occasions I have seen a design that when faced with the different friction of the field tiles, as compared to say carpet, had issues with the drive. This was usually due to lack of rigidity so that when the robot was turning the motors were working to hard. Do you have any pictures of your drive you could post?
After a match when it stops, we can unplug it from the switch and drive it off of the field. Also we have had numerous practice meetings where it preformed perfectly on the tiles without a competition switch.
Perhaps the breakers in the extender are, ahem, breaking, but unplugging the cord gives your robot just enough time to cool down and begin working again. We had a similar problem in the finals matches…
I believe a power expander has 4A max capacity,
whereas a Cortex has 4A for ports 1-5 and 4A for ports 6-10.
So putting all 4 wheels on one power expander doesn’t maximize this current bottleneck.
If you think the power expander is the problem, have you tried not using it?
You don’t think the motors themselves are the problem, even though they are new (and untested)?
Correlation is not perfectly linked to Causation;
there may be other common factors that you don’t recognize and list.
I would try using the scientific method, (form a testable hypothesis…) and see if that helps you debug the problem. There are lots of other email threads with similar problems, suggestions, and results of experimenting, if you know how to use the search features of the forum.
Drive the robot around at your meeting for upwards of 15 minutes straight (or similar large number of minutes) without stopping once. This will help determine if the motors are burning out…
Thank you all for your suggestions. We will experiment with bypassing or maybe replacing the power expander.
This is just so infuriating because in the finals we couldn’t do anything but drive in circles.
I have no pictures of the drive base to post, so I will try to explain it as best I can:
There are four high-powered motors geared for torque; two on each side.
On each side, there are three wheels driven with five gears (similar to the protobot, where two motors drive four wheels), with the motors driving the gears that are between the wheels. This puts more motor power to the ground.
I will post more updates when we do some experiments at our meeting tomorrow. Thanks again for your suggestions!
Exactly what kind of wheels? 2.75"? 4"? Omni vs hi trac vs bare plastic?
Turning circles on vex foam,
at a competition where anti-static spray has been used on the foam,
requires much more torque from the motors than straight driving on hard-tile.
The six wheels are 4". On each side, the front wheel and rear wheel are omni-directional, but the middle wheel on each side is a regular 4" wheel to prevent other robots from pushing us sideways. The omnis help us turn more nimbly.
We have experimented with driving our robot into a wall and found that the circut breaker in the power expander is the cause. We have programmed button 7U on our controller to cut our motor power for 5 seconds to reset the circut breaker. We have found that, like the time when the robot is disabled at the end of a match, this rest period successfully destalls the motors and allows us to continue driving normally.
Thank you all for your help! Your suggestions have been instrumental in finding the issue. If anyone has better ideas than our time-costly reset, we’re all-ears.
Here is our new issue: the distribution of power. The power expander has a 4-amp circuit breaker. The cortex has two; one for ports 1-5, another for 6-10. The stall current for the 393s we use on our wheel base is 3.6 amps, so no matter where we connect them, we will have stalling. If two motors are in ports 1-5, and the other two in 6-10, with our scissor-lift motors in the power expander, then encountering an opposing force (such as another robot) will quickly raise the amps above the 4 amps on each circuit breaker, stalling the whole wheel base. We would like an insight into how we can arrange the motors to avoid this. To help I will post our motors and ports:
Port 1: Intake conveyor 1 (269)
Port 2: Left-front drive 393 (routed through power expander)
Port 3: Left-rear drive 393 (routed through power expander)
Port 4: Scissor Lift motor 1 (269)
Port 5: Scissor Lift motor 2 (269)
Port 6: Scissor Lift motor 3 (269)
Port 7: Scissor Lift motor 4 (269)
Port 8: Right-front drive 393 (routed through power expander)
Port 9: Right-rear drive 393 (routed through power expander)
Port 10: Intake conveyor 2 (269)
Suggestion to try: Don’t put all your 393s (drive motors) on the power expander. The drive motors tend to draw the most current of any motors. I’d put the intake conveyor motors and two of the scissor motors on the power expander, and the drive motors spread on the two cortex breakers.
That or something like it is what we may do. But one thing nags at me.
The problem is that even two 393s on a circuit breaker could trip the breaker. The stall current per 393 is 3.6 amps, so two motors approaching their stall current would trip the breaker. Although, I suppose that would be a lot better then four on one breaker.
Under what conditions does a 393 reach its stall current?
That’s a good thought… I hadn’t thought of that…
The stall current is reached when the motor cannot spin.
One idea would be to put one 393 from each side of the drive on each breaker. That way, you’d have to be stalled on both sides of the drive to trip the breaker.
Also, it seems to take a few seconds to trip a breaker, also… (it has to heat up before tripping)
This would work for two breakers, but as there are four motors and three breakers we would have to double up one of them. Which breaker do you think would be best to have two in (or does it not matter)? The Power Expander may be a good choice for two motors because it has its own battery.
You said that the motor reaches stall current when it cannot spin. So, if we were pushing an opponent that had a less powerful wheel base (say, four 269s), they would not stall?
One thing to understand is that the PTC in the power expander will not trip immeadiately when the current reaches 4A, I need to lookup the datasheet to find the exact spec, but it should be able to tolerate more current for a short amount of time. Secondly, your motors should not be working near stall but rather in the middle of their range at perhaps 1.5A. Here’s is a post from a similar thread that explained how the PTC’s work, in this case it was for one in a motor but the principle’s the same.
Ok, so the PTC in the cortex is an HR16-400. I assume it is the same in the power expander.
This means that it’s minimum hold current, the current at which it will not trip at 25 deg C, is 4A
The minimum current at which it will always trip is 6.8A, according to this graph it will take around 20 seconds.
The graph also implies it can withstand 10A for perhaps 5 seconds. All these numbers need to be derated as the PTC heats up but even at 50 deg C it can still hold 3.2 A according to the datasheet.
So, I am given to understand that curve A is the Power Expander PTC. If this is wrong, do correct me. But in that case, it seems that the power expander is the best spot for two drive motors, since curve A trips with higher current.
