Micromouse interesting problem

Thread Starter

mik3

Joined Feb 4, 2008
4,843
Hello guys,

We have built a micromouse as a project at university.

The problem is that when is runs in the maze and someone touches on any metallic part of it, the microcontroller resets.

The maze floor is made of plastic (I think it is PVC) and the wheels of the micromouse are aluminium with rubber around them.
The chassis is aluminium but it is not connected to the circuit signal ground.
It is powered by a LiPo battery.

I believe that the problem is static electricity. As the mouse moves around charge builds up on the chassis due to the friction of the rubber of the wheels and the plastic floor. When someone touches a metallic part, it discharges quickly to earth and the current pulse resets the microcontroller. I connected the chassis to earth via a wire and the problem was solved.

Any ideas of what is going on?

Thank you.
 

hgmjr

Joined Jan 28, 2005
9,027
You may want to consider using what are essentially carbon fiber brushs that are attached to the chassis of the micromouse so that the bristles touch the surface of the floor. This should prevent any build up of static charge on the micromouse.

hgmjr
 

Thread Starter

mik3

Joined Feb 4, 2008
4,843
You may want to consider using what are essentially carbon fiber brushs that are attached to the chassis of the micromouse so that the bristles touch the surface of the floor. This should prevent any build up of static charge on the micromouse.

hgmjr
How does this prevents static charge build up?

The mouse has two metallic ball castors for mechanical stability. One of them is always touching the floor.
 

hgmjr

Joined Jan 28, 2005
9,027
That's a good question Mik3. I would have imagined that the metal castors would provide a discharge path to any buildup of ESD.

Would you be able to post a picture of your micromouse so that we can see what it looks like?


hgmjr
 

Thread Starter

mik3

Joined Feb 4, 2008
4,843
That's a good question Mik3. I would have imagined that the metal castors would provide a discharge path to any buildup of ESD.

Would you be able to post a picture of your micromouse so that we can see what it looks like?


hgmjr
I don't have a photo of mine now but this is the chassis used:

http://www.active-robots.com/products/platforms/microrobot-chassis.shtml

Also, it looks like this:

http://www.robotstorehk.com/micromouse/RS-AIRAT2.html


How does a metal touching the ground prevents the static charge build up since the floor is not conductive?
 

THE_RB

Joined Feb 11, 2008
5,438
1. Connect the alimunium chassis of the robot to the main 5v) ground. It's best to use a 270 ohm resistor as the connection, in case of accidents.

2. Check the reset or MCLR pin of the microcontroller, some boards use a very high value resistor to hold the reset pin in RUN mode, like 47k. You might need to reduce the value of that resistor, generally 10k will still work well and provide much better immunity against noise-caused reset.
 

Thread Starter

mik3

Joined Feb 4, 2008
4,843
1. Connect the alimunium chassis of the robot to the main 5v) ground. It's best to use a 270 ohm resistor as the connection, in case of accidents.

2. Check the reset or MCLR pin of the microcontroller, some boards use a very high value resistor to hold the reset pin in RUN mode, like 47k. You might need to reduce the value of that resistor, generally 10k will still work well and provide much better immunity against noise-caused reset.
1. The robot is power by batteries.

2. The MCLR is not used, it is used as an I/O.
 

hgmjr

Joined Jan 28, 2005
9,027
You may be able to reduce the ESD problem by spraying your bot with a light coating of anti-static spray.

hgmjr
 

THE_RB

Joined Feb 11, 2008
5,438
1. The robot is power by batteries.

2. The MCLR is not used, it is used as an I/O.
It is still relevant. Wherever the 5v ground is (the gnd pin of the 5v regulator) attach that to the metal chassis through a 270 ohm resistor. If it does not have a 5v regulator then connect the ground plane of the micro PCb (or gnd pin of the micro) to the chassis through the resistor.

MCLR pins used as inputs have lots of spurious reset problems. Tie the pin to Vdd (5v) through a 10k resistor (and maybe add a cap from that pin to ground), or use it as an output pin instead.
 

Thread Starter

mik3

Joined Feb 4, 2008
4,843
It is still relevant. Wherever the 5v ground is (the gnd pin of the 5v regulator) attach that to the metal chassis through a 270 ohm resistor. If it does not have a 5v regulator then connect the ground plane of the micro PCb (or gnd pin of the micro) to the chassis through the resistor.

MCLR pins used as inputs have lots of spurious reset problems. Tie the pin to Vdd (5v) through a 10k resistor (and maybe add a cap from that pin to ground), or use it as an output pin instead.
I will try that with the 270 ohm resistor. I have a 10K resistor between 5V and MCLR. I will try to put the capacitor too.

Can you explain what is the purpose of the 270 ohm resistor?
Why not to use a wire?
 

THE_RB

Joined Feb 11, 2008
5,438
The 270 ohms is low enough to function well for grounding.

But it adds a safety factor, since we can assume the electronics on this robot can be touched by humans (ie PCB is exposed) then the robot will have some protection if a user accidentally touched a metal screwdriver between metal chassis and some powered part of the robot (like a 5v rail or a live output pin). In that case the 270 ohm resistor will limit current to about 18mA so nothing is destroyed.
 
Top