possible solution to the static problem

[engunneer]

I used to work at iRobot on Roomba, and some of the very early testing models had static issues (we were trying to cause them, we kept a special testing room at 10% humidity or less). As we insulated more and mor, and protected the processor more and more, the static would build up higher and higher until it eventually caused permanent damage when it did zap (it will always zap eventually). The adding static bleeders really helps the situation.

to the earlier comment that you might not be in contact with the carpet long enough before you hit the field wall - Electricity is pretty fast. If you have a good ground tail, it should bleed off the charge almost immediately.

I hope things are better this upcoming week. At least Seattle and Portland don't have issues with low humidity.

When I was working at iRobot to troubleshoot this, we picked up a simple static meter to measure the static on objects. It's very odd waving a box at your target and getting a static reading. The sensor on the front of the box was just a PCB with a giant copper pad that you had to point at your target. It was pretty cool.

[writchie]

Originally Posted by Tom Bottiglieri

The FRP is non conducting, so there is no path from the robot to the ground while it is driving around. If you are driving full speed into a wall or airlock, there may not be enough contact with the carpet to discharge the build up into the ground before it contacts the field frame.

Yes the FRP is non-conducting but the use of a few static discharge wicks in light contact with the FRP floor may equalize the potential between the floor and the robot. This is common practice in aircraft design where static discharge wicks attached to control surfaces and trailing edges help prevent buildup by discharging to the air. It is also good practice to bond any metal parts that may be floating from the main chassis, except of course for the camera and cRIO.

In Florida, we usually have such high humidity that static is rarely a problem. But this year, with the unusual cold, many days here remind me more of the static environment of the Northeast or Canada. Our students have been getting zapped left and right, much to their surprise. We haven't had any resulting electrical problems (yet).

Depending on the exact mechanism involved in the static related resets, it could be that a static drain resistor of 100K - 1Meg between chassis ground and electrical ground might afford some protection. This would at least prevent very high static voltages from building up between the chassis ground and the electrical system without interfering with the safety aspects of the electrical isolation. In fact, it would be pretty easy to have a jumper with a resistor that could be installed (or not) to see if it makes a difference. Testing in a low humidity environment is certainly needed and perhaps this could be done in practice matches next week.

We had ESD problem a few decades ago with a communication device that output to line printers. The issue only cropped up in a few locations in the winter and IIRC we solved it with a drain wire on the paper basket and a couple of discharge wicks in the paper path. Although the technology was older (Z80's), the problems were similar - mysterious unexplained resets.

I think FIRST need a "Tiger Team" on this to get control of the issue by next week. Just my $0.02

[DonRotolo]

Originally Posted by Tom Bottiglieri

The FRP is non conducting, so there is no path from the robot to the ground while it is driving around.

Perhaps, but as writchie notes, a grounding tail will allow static to discharge readily. We used some finely-stranded #10 wire, about 3 inches long, one end into a ring terminal and the other end fanned out into a good imitation of a paintbrush.

This was purely precautionary, as we have not experienced any issues where we might suspect ESD or static.

Originally Posted by Wayne C.

We wear Static Guard like cologne.

Ah, that explains that

Folks should also note that just the spark of an ESD event is not the only chance for damage - the spark creates an electromagnetic field which can cause damaging voltages to be induced into near-by conductors. So it's not enough to control when & where it sparks; you need to control the sparks themselves. Team 25, with their anti-static wrist straps and mats, are on the right track, This isn't rocket science people.

Originally Posted by David Brinza

One of the tricky issues with ESD is that it can cause "latent" damage. A piece of equipment might continue to work after a "zap", but in reality some internal damage may have occurred. The next zap - even a tiny one, can cause a permanent failure.

Such latently damaged items are also called "walking wounded" and are especially problematic - the wound may also cause unintended operations, but it will be so intermittent it becomes difficult to find. And, in a high-reliability application (such as a spacecraft) it is a disaster.

I will ask our coach to ask Q&A to 'approve' a 1 MOhm resistor between components (frame, power, and cRio) for ESD bleed.

Don

[jerry w]

Originally Posted by David Brinza

There are some of us in FIRST who do understand the importance of ESD control.

Those out there with real-world experience in ESD control should post recommendations for teams. FIRST needs to take the appropriate measures with the field (some of which are already in place, like the earth ground at the alliance stations).

Team-79 Antistatic solution
This should be tied to a bare-metal location on the robot. Allow about 2 inches of chain to drag on the floor.
As electrostatic charges are accumulated during motion, they are simultaneously drained from the robot. Thus dangerous levels never occur.

CHAIN BEADED 12" BRASS
SKU: 34733

77012 WESTINGHOUSE
$1.15

<http://www.acehardwareoutlet.com/(wo...e=chain+beaded>

[writchie]

Originally Posted by jerry w

Team-79 Antistatic solution
This should be tied to a bare-metal location on the robot. Allow about 2 inches of chain to drag on the floor.
As electrostatic charges are accumulated during motion, they are simultaneously drained from the robot. Thus dangerous levels never occur.

CHAIN BEADED 12" BRASS
SKU: 34733

77012 WESTINGHOUSE
$1.15

<http://www.acehardwareoutlet.com/(wo...e=chain+beaded>

Unfortunately, this solution would not appear to be permitted at present.

See the Q&A

http://forums.usfirst.org/showthread.php?t=12228

Hopefully there will be an officially sanctioned solution soon.

[Tom Line]

This problem has now directly affected us and our regional.

Both the ethernet ports on our DS were fried. The first occurred the first day of the regional. We began using the second port at that time.

The second occurred AS the drive team was connecting to the field controls for our first elimination match. We had just tested the robot tethered in the pits. The control panel went directly to the driver station - and they couldn't get connection.

We were forced to remove the DS from our board and remove (12) PWM cables, fasten the new DS and reattach all 12 PWM cables in the time it took the robots to reboot. As a result, autonomous did not work. Two of the PWM cables did not make good connections. This allowed their human player to have their way with us. We lost by two points.

In the second match, the only thing that worked was the drivetrain joysticks. The turret joystick did not work, and none of the analog or digital inputs were sending signals. As a result we went the entire match without being able to score.

We will be grounding the robot from now forward AT ALL TIME if we are going to tether. Period. I strongly suggest other teams do the same.

We were told by field personnel that they had seen another (or more?) DS that had dead ethernet ports. Their guess was that the static discharge is being transferred down the ethernet cable from the robot while tethered.

Be careful and do everything you can to prevent it from happening to you.

[jerry w]

Originally Posted by writchie

Unfortunately, this solution would not appear to be permitted at present.

See the Q&A

http://forums.usfirst.org/showthread.php?t=12228

Hopefully there will be an officially sanctioned solution soon.

actually this is a mechanical solution.
The rule that allows the chain is R05

<R05> Exterior or exposed surfaces on the ROBOT shall not present undue hazards to the team members, event staff or GAME PIECES.

We are not affecting traction as required by R06

<R06> ROBOTs must use ROVER WHEELS (as supplied in the 2009 Kit Of Parts and/or their equivalent as provided by the supplying vendor) to provide traction between the ROBOT and the ARENA....
No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted.

The chain is a mechanical device fastened to the robot frame and does not violate R41

<R41> All wiring and electrical devices, including all control system components, shall be electrically isolated from the ROBOT frame. The ROBOT frame must not be used to carry electrical current.

By adding a brass chain to the robot frame we have eliminated the buildup of electrical charges. This is a safe and legal method for protecting humans from electrical shock. It also protects the electrical components on the robot.

[Ioski]

We experimented with a makeshift grounding cord made out of a shoelace that was attached to the frame of our bot. It dragged along the floor under the robot, to try to eliminate some of the static. It worked well... seeing in how we never fried our driver station.

[Viper37]

Originally Posted by writchie

Unfortunately, this solution would not appear to be permitted at present.

See the Q&A

http://forums.usfirst.org/showthread.php?t=12228

Hopefully there will be an officially sanctioned solution soon.

It depends on where you are. Our CRIO as well as two other team's were cooked over the course of the regional. FTA/ Field Coordination authorized all teams at the regional to use ground wires dragging on the field under the robot.

After everyone started putting these wires on, not a single CRIO failed.

[writchie]

Originally Posted by jerry w

actually this is a mechanical solution.
The rule that allows the chain is R05

<R05> Exterior or exposed surfaces on the ROBOT shall not present undue hazards to the team members, event staff or GAME PIECES.

We are not affecting traction as required by R06

<R06> ROBOTs must use ROVER WHEELS (as supplied in the 2009 Kit Of Parts and/or their equivalent as provided by the supplying vendor) to provide traction between the ROBOT and the ARENA....
No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted.

The chain is a mechanical device fastened to the robot frame and does not violate R41

<R41> All wiring and electrical devices, including all control system components, shall be electrically isolated from the ROBOT frame. The ROBOT frame must not be used to carry electrical current.

By adding a brass chain to the robot frame we have eliminated the buildup of electrical charges. This is a safe and legal method for protecting humans from electrical shock. It also protects the electrical components on the robot.

The rules govern. The Q&A "interpretations" of the rules are official. The "lead inspector" is the final authority at each competition. That is how I understand it.

Re: <R05> this rule is prohibitive - not permissive. It doesn't permit you to install something that would not otherwise be permitted.

Re: <R06> The lack of friction in the discharge "wire" was irrelevant to the Q&A answer. The question presumed the traction was immaterial.

Re: <R41> The Q&A questions was specifically in this context.

IMHO attempts to classify a brass chain, or wire, or metalized string, or whatever as "mechanical" and not "electrical" in order to escape an "electrical" rule are likely to carry zero weight. The actual use of the part makes it electrical, mechanical, pneumatic, or in this case electrostatic.

The referenced Q&A answer prohibits a bleed resistor between electrical ground and the mechanical frame and it prohibits a static wick between mechanical frame and the floor. The meaning of "no , and no" is pretty clear, even in a world where the meaning of "is" may be disputed. The GDC could have answered, yes and yes, no and yes, or yes and no, or even "maybe". Instead they said "At this time, no and no, as these would be violations of Rule <R41>." Since a "static discharge wick" is presently prohibited, IMHO the same would apply to drag chain.

I fully agree with you that a chain (or two) would prevent the electrical buildup. So would a static wick, a wire tail, a foil tail, and perhaps other methods. In fact, your chain idea MAY be the best. I've certainly seen it used successfully in the past. Perhap FIRST installing a drag chain on each trailer (which it controls and could easily supply) and allowing teams to install Static wicks made from off the shelf stranded wire will be the solution.

The GDC has determined that for the time being it wants to maintain floating electrical, mechanical, and surface grounds without electrical connection, although it left this open to later revision. While I am surprised by this ruling (especially regarding wicks), I am sympathetic as are most who have been in similar situations with mysterious ESD issues and a large number of units already in the field. At this stage, I think the GDC is trying to minimize the variables in order to reach the quickest possible solution. We should let them do their job. Perhaps, a couple of fried DS and cRIO's are worth it if it quickly leads to a permanent solution. Maybe with week 2 now completed, they will promulgate a fix and/or rule change that is proven not to cause more harm than good.

Personally, I think there may be multiple problems. One results in a dead Ethernet port on the DS. The other results in a cRIO reset, and sometimes a fried cRIO. Since both are proprietary components, we don't know what protection they have internally - we can only speculate. But since the DS doesn't even have a pull up on the digital inputs, it is likely quite vulnerable. I doubt it has Ferrite Beads and TVS diodes on the Ethernet ports and it is possible that the actual damage comes from induced surges.

IMHO, the best course of action for teams is to have only ESD trained team members handle the Ethernet ports and insure that everything is discharged before plugging things in. I would also recommend that teams be prepared to install static wicks (or chains) if these are ruled as permitted. When operating tethered, I would recommend always using a hub or switch with known ESD protection and only connecting/unconnecting the DS cable in the pits.

As for good news, the forcast for the Florida Regional is 81F/61F with humidity of 40% - 50%. With the return of warm nights, the conditions for static should must less than they have recently been, although Murphy makes sure that ESD can get you anywhere.

[Bryan Herbst]

Our team experienced this, though not at a regional. While I was away for other things, our driver and a few members were driving the robot on gymnastics mats. They quickly found me to inform me that there had been a huge static spark, and one of the jags had been fried.

Totally our fault.

In any case, to those talking about the old system being better- I am no expert, but I see no reason why ESD would not affect the old system (mostly victors). The actual controller would be safe, yes, but everything else would be just as at risk.

[jason701802]

Although we did not have any problems at the portland regional i did notice while watching the matches most of the "field problems" happened to the same teams. whenever they had to do a rematch of a rematch, which happened quite a few times, it was the same teams that had problems in the original matches and the rematches.
someone mentioned that portland was lucky, although we may not have been in the worst shape, we weren't lucky. the high humidity in the valley may have helped in the pits but the field was built on an ice rink and in ice rinks they keep the air dry to reduce condensation. cold and dry, the perfect breeding ground for static. another mistake they made was to turn off the house lights during lunch on Saturday, cooling down the field even more, so it took more than an hour an a half to finish the first four quarterfinal matches.

[Greg McKaskle]

It depends on where you are. Our CRIO as well as two other team's were cooked over the course of the regional. FTA/ Field Coordination authorized all teams at the regional to use ground wires dragging on the field under the robot.

Do you mean the gray chasis with modules on the robot, or do you mean the blue driver station? I'm asking because to my knowledge I hadn't heard of static being an issue with failed Jags or cRIOs. It can cause a reboot, and we have seen some internal fuses blow, but thus far not associated with static, that we know of. Please clarify.

Greg McKaskle

[Mark McLeod]

We had a case in NYC of a robot rebooting itself when it ran it's ball conveyor. It could be reproduced in the pits and the problem went away when the conveyor was sprayed with anti-static solution.

Mark Black can give you a report.

[Greg McKaskle]

We had a case in NYC ...

The reboot of the cRIO due to static buildup is not ideal, but not unexpected given that the chassis is forced to be floating. Cooking the cRIO is not expected, and that is my real question. Were cRIOs cooked at the event? Mark Black is not back from travel yet.