RoboRIO WAGO breakout board with built-in BO055 Gyro mount

As an off season project we designed an MXP breakout board that incorporates a mount for the Adafruit BNO055 gyro. It has push in to lock wago connectors that hold wires in more securely.

What are the rules for using custom built boards? And what are the rules for needing to release designs?

Also, if I offer to sell these as a COTS does that help us with those pre-season development rules?

Lastly, ordered 10 of these. Only probably need a few, is there any interest out there to buying the extra from us?

We also have a full RIO version that covers all of the Relay, AO and PWM and DIO channels.





Here is the full RIO version.

20181029_101747.jpg

We don’t know the 2019 rules yet, but 2018’s R75 has remained pretty constant in the roboRIO era. I can’t tell whether your board is an active device or not (if so, you probably have an uphill climb), but that should get you started. :slight_smile:

R75 definitely comes into play, if you want to use the board to control any motors. Likely it would fall under part B, “via a network of PASSIVE CONDUCTORS used to extend the PWM pins”, but as an LRI having the Gyro mounted in there does present at least some level of concern. The Gyro is, undoubtedly, an active device. This would fall under part of the blue box:

The “network of PASSIVE CONDUCTORS” only applies to the pins being used for PWM output to motors or servos. This means that connecting an ACTIVE DEVICE, such as a sensor to one MXP pin does not prevent other MXP pins from being used in accordance with R75-B.

And so you would need to be able to show that to the satisfaction of your LRI - are the traces exposed on the backside? Is it easy to tell where everything is connected to?

You would also need to comply with R14, “Physical ROBOT elements created before Kickoff are not permitted” and R15, “Software and mechanical/electrical designs created before Kickoff are only permitted if the source files (complete information sufficient to produce the design) are available publicly prior to Kickoff.” So that means posting the designs publicly before kickoff, and not ordering/building the board you use on the competition robot until after kickoff. Which would make the 10 you already made illegal, and force you to make a new one each and every year you want to use it.

If you became a VENDOR, R14 and R15 don’t apply. But that isn’t as simple as offering to sell them. You need a Federal Tax Identification number, can’t be wholly owned by a team, must be able to meet general shipping guidelines (meaning you have to maintain stock), and a few other things - see page 65 of the game manual (the second page of the Robot Rules section) for full details.

The better solution would be to partner with one of the current VENDORS, like AndyMark or Vex. They could satisfy the VENDOR rules for you and help you get the part out to all the teams that want it.

Everything is completely passive and the BO055 is not actually part of the board. It’s something someone can buy from adafruit and add themselves.

Thanks for the feedback. We just wanted to make sure we didn’t have wires falling off our RIO. We thought this would be an easy solution.

The components are off the shelf. Board we have custom made in china. What if we don’t solder anything together first and just have the parts ahead of time?

The key is “Board we have custom made in china.” The first page of the Robot Rules defines COTS as:

Many rules in this section reference Commercial-Off-The-Shelf (COTS) items. A COTS item must be a standard (i.e. not custom order) part commonly available from a VENDOR for all teams for purchase

emphasis mine.

You would hare to order the board after the season starts, based on design files that were publicly posted prior to kickoff. It qualifies as a COMPONENT or MECHANISM, but not as COTS.

What if they didn’t make the design files available but instead made some tweaks to the design prior to having it fabricated at the start of build season? It’s a new design as far as I can tell. Rules don’t seem to specify anything about iterations.

Also, what if they posted the design on something like OSHPark and then clicked “order” at the start of build season? Or clicked order BEFORE build season? JIT fabrication isn’t currently covered either.

If the circuit board is the hang up, release the CAD pre kick off, order 3+ week turn PCB post kick off. Use these PCB assemblies on the comp bot. Would that satisfy the rules?

What if you ordered it prior to kickoff and it arrived after kickoff? Wouldn’t that satisfy the rule? You have no control over the manufacturing process and when it is actually fabricated - just so long as it shows up after kickoff, doesn’t that mean it was in compliance with the rules?

By the way - we’re all talking about rules that we don’t know exist yet. R14 and R15 (or whatever they end up being) can’t be verified until after we get the manual… unless of course someone at HQ wants to write a blog post saying otherwise.

There is certainly some room here for interpretation. If it were my team, I would want to be able to show the LRI the following:

  • clearly time/date-stamped public posting of a design, if there was any possible way (like a discussion thread on CD) anyone could think it was designed prior to kickoff. That would, at minimum, show the pre-kickoff design, even if changes were made post kickoff. That’s my proof that the team followed the design rules.
  • A clearly dated receipt or invoice for the order. That date would be my proof that my team followed the fabrication rules, and I would want all dates on that invoice to be post-kickoff so there was no room for argument.

Where you get into some gray area… What if I place the order with an overseas supplier the day before kickoff? Everyone would know that there was no possible way I would have the part in hand prior to kickoff, and likely the supplier wouldn’t make it until after kickoff anyways. What if the supplier lists the purchase date as well as a guaranteed ship date? Which date do you use?

Looking at this from the perspective of “what would I want my team to do?”, I would definitely be erring on the side of caution, and on being able to provide as much proof as possible that the rules were followed. Looking at this from the perspective of an LRI… I would *really *hope questions were asked on the Q&A to help clarify this section of the rules. Of course, the rules may change for next year, but this section has been fairly stable lately, so until kickoff I would go with the assumption that these ones would not change.

Is this Jon Stratis the poster on CD or Jon Stratis the LRI speaking?

Because I’ve never gotten the vibe that anything other than the team’s attestation that they followed the construction rules was needed. If you’re speaking in the latter capacity I have extensive concerns and would need to see some backing from the rules/q&a before I comply with ANY [L]RI request receipts.

This is what I, as a mentor, would want my own team to do. In areas where we can provide documentation that we’re following the rules, I want my team to have that documentation ready and available. It helps to avoid any potential headache at competition. For example, last year we had a collection of spec sheets for *all *of our pneumatic components, with the necessary spec’s highlighted. None of the components were questionable - pretty standard Clippard tanks, Automation Direct components, Bimba Cylinders, etc. But we had the documentation for them all the same.

Looking beyond this specific rule, as an LRI I have rarely found a team to actually have documentation ready when I need it. Wire gauge and pneumatic tubing ratings are a very common one - if it’s not printed right on the wire or tubing, then I need to see something to convince me it’s correct. Even when it is printed on the wire or tubing, if I can’t get to a position where I can read it (the only place the wire is used in buried in the middle of the robot) there’s an issue. Teams that don’t come prepared with documentation can have delays in passing inspection as we try to verify it. Last year it was teams using linear “automotive motors” that were unable to show they were actually automotive motors per the rules.

There are quite a few “quick turn PCB manufacturers” that will manufacture small quantities of PCB’s to a professional quality standard in the space of a few days for a pretty reasonable cost. The cost might be a bit higher than having them done overseas. The reality is that the cost would still be a small fraction of your overall annual budget. They are generally pretty reliable since they serve professionals who are under tight time constraints. One that quite a few of my colleagues used is PCB Express.

Nice board! I made something similar last year. Unfortunately we never got a straight answer regarding the legality of our board, even though it was passive, like yours. We even went as far as taking a sample to HQ during FTA training, and still no definitive answer. The other issue with making custom PCB’s is you have to get them made during build season in order to comply with the rules. It’s frustrating and in the end we gave up on the idea.

Aren’t those Weidmuller LSFs (like on the PCM and VRM) not Wagos (like on the PDB breakers)?

I would have chosen a different connector. Even though we use ferrules and proper crimpers, we’re still experiencing problems ensuring that the weidmullers’ never come loose. Almost cost us a match at Ruckus.

“Publicly posted” can be pretty vague.

But the plans were on display…”
“On display? I eventually had to go down to the cellar to find them.”
“That’s the display department.”
“With a flashlight.”
“Ah, well, the lights had probably gone.”
“So had the stairs.”
“But look, you found the notice, didn’t you?”
“Yes,” said Arthur, “yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard.”

I have always understood that when questioned the team had to document to the RIs satisfaction that a particular component is within the rules. I remember particular silliness in locomotion documenting that the minibot switch was a “light switch” and not some random electrical switch that looked exactly the same as a “light switch”

I have developed a similar board that uses screw terminals, but it only covers the DIO ports, the I2C port, the RS-232 port and the RSL. This way we can have a solid connection for sensors. I’m not as concerned with the PWM connectors as those hold fairly well. I have a prototype board coming from OSHPark soon, and I’ll share that here.

I’m glad I’m not the only one that has been frustrated with the Rio’s ability, or lack thereof, to hold connections in place.

Oshpark does have a minimum of a week-long turnaround time + shipping, so be careful using them for this. I checked some quotes on Express PCB and was quoted somewhere between $80 to $120 per board if you want a week long turnaround… getting it in a few days instead doubles the price.
I think the biggest barrier to this is getting the PCBs made in time.
The Spartan board with gyro is only $140: http://www.wcproducts.net/sensors
It breaks out PWM and DIO, as well as the SPI port and adds a gyro. It also provides a more stable power supply for sensors over the regular RoboRIO power. All the pins are compatible with locking PWM housings. I don’t remember the exact P/N off the top of my head, but I’m sure emailing WCP will get the desired info. If you want fast turnaround, the Spartan Board could be a better option.

Our solution to the problem was just to route all sensors through CAN. All of the PWM/GPIO ports were taped off on our robot this year. :stuck_out_tongue:

Sorry, you are correct, they are not Wagos. These are from Pheonix Contact but essentially the same as the PCM and VRM connectors. I will say that seem to hold better than those on the VRM/PCM.

What was the proof they were finally able to provide that satisfied the enforcement of the rule? I’d like to know so I can better prepare my own team.