[Ri3D] BOOM DONE. We're not done...

[Joe Johnson]

Originally Posted by Ether

Could you please update us on what you have found out about this?

Working with the Vex folk to understand this. Reportedly, they sold a boatload of these last year with no complaints.

I've ordered AndyMark HD's as a backup in case. Backup #2 is to go with all omnis (like JVN's Blitzbuild robot but we have a CIM+MiniCIM per wheel -- shojuld be a screamer).

Dr. Joe

[marccenter]

Dr Joe,

I have tried a couple of times, within the past two hours, to download the Boom Done - MCAD.zip files without success (receive error message). Has anyone else at CD been successful?

Thanks again for a great design. For small resource teams, your design is now our prototype for potential improvement (but don't hold your breathe for too long).

Regards,

FRC3548 Coach Marc Center

[Joe Johnson]

Originally Posted by marccenter

Dr Joe,

I have tried a couple of times, within the past two hours, to download the Boom Done - MCAD.zip files without success (receive error message). Has anyone else at CD been successful?

Thanks again for a great design. For small resource teams, your design is now our prototype for potential improvement (but don't hold your breathe for too long).

Regards,

FRC3548 Coach Marc Center

Hard for me to test because we are all on the same system here. Has anyone else had the same problem?

This is what works for me:

1. Go here (there is some sort or error message here but ignore it)
2. File menu > Download.
3. BOOM DONE.

[magnets]

Originally Posted by Joe Johnson

Hard for me to test because we are all on the same system here. Has anyone else had the same problem?

This is what works for me:

1. Go here (there is some sort or error message here but ignore it)
2. File menu > Download.
3. BOOM DONE.

It works for me from this link. Thanks!

[magnets]

Originally Posted by Ether

Could you please update us on what you have found out about this?

The team I'm with used these mecanums last year, and the rollers have a TON of play (more than a new wheel), but they seem to work, so no issue here after ~40 matches.

[otherguy]

Dr. Joe,

I've pieced together what I could from the videos, and it sounds like you have a sharp IR sensor on board for ball detection, as well as some kind of acoustic sensor for measuring robot distance to the wall.

I tried digging through the java code you guys posted, but all I saw for distance measurements was based off the camera code/target detection.
If I've overlooked something please feel free to point me to the correct Java Class and I'll re-review.

Would you be able to provide any additional information on the sensor(s) used for ranging. I'm interested in knowing what make/model part was used and any usage notes (e.g. under what conditions did the sensor perform well/poorly).

Thanks, and great work.

[David Brinza]

Dr. Joe,

Congrats on winning the poll for best Ri3D robot!

The final vote was too close for comfort, given the capabilities you both demonstrated and explained.

I should have taken a class in marketing, because the second place robot certainly had a "wow" factor in their release video (cool lights, sound track, slow-motion). "Form before function?"

BTW, we're also interested in the sonar sensor you used. We've had mixed results with the Maxim devices (although we figured out some of the pitfalls).

[Ether]

Originally Posted by magnets

the rollers have a TON of play

axial or radial? or both?

[Joe Johnson]

Originally Posted by otherguy

Dr. Joe,

I've pieced together what I could from the videos, and it sounds like you have a sharp IR sensor on board for ball detection, as well as some kind of acoustic sensor for measuring robot distance to the wall.

I tried digging through the java code you guys posted, but all I saw for distance measurements was based off the camera code/target detection.
If I've overlooked something please feel free to point me to the correct Java Class and I'll re-review.

Would you be able to provide any additional information on the sensor(s) used for ranging. I'm interested in knowing what make/model part was used and any usage notes (e.g. under what conditions did the sensor perform well/poorly).

Thanks, and great work.

All the sonar stuff is on the Arduino. We currently don't send the distance data anywhere but the display. This was just an interim solution. Long term we want the CRIO to have access to the distance data. We also are thinking about some interference detection/avoidance methods. Stay tuned...

Dr. Joe

[billbo911]

Originally Posted by Joe Johnson

All the sonar stuff is on the Arduino. We currently don't send the distance data anywhere but the display. This was just an interim solution. Long term we want the CRIO to have access to the distance data. We also are thinking about some interference detection/avoidance methods. Stay tuned...

Dr. Joe

We will be doing a similar approach, but a bit backward to your approach. We already have the Arduino running the US sensor and displaying the relative distance via an addressable LED strip.
Our final design will have the cRio doing the actual distance measurements, that way it can position accurately by itself, and passing the distance info to the Arduino to display info to the driver. The display may end up on the Driver's Station, or on the robot, or both. All is still TBD.

[MikeZ]

Originally Posted by Joe Johnson

All the sonar stuff is on the Arduino. We currently don't send the distance data anywhere but the display. This was just an interim solution. Long term we want the CRIO to have access to the distance data. We also are thinking about some interference detection/avoidance methods. Stay tuned...

Dr. Joe

Dr Joe

Great job and thanks for your posts. Question on the sonar. I noted in your specs and capability document a reference to a sonar horn/cone made from sheet metal. We have tried the same ultrasonic sensor will mixed results so would appreciate if you could elaborate a bit on your cone design and how it enhanced sensor performance. Many thanks.

[Tom Line]

Originally Posted by Joe Johnson

Working with the Vex folk to understand this. Reportedly, they sold a boatload of these last year with no complaints.

I've ordered AndyMark HD's as a backup in case. Backup #2 is to go with all omnis (like JVN's Blitzbuild robot but we have a CIM+MiniCIM per wheel -- shojuld be a screamer).

Dr. Joe

Joe,

Our sister team from Richmond used Vexpro mechanums last year, and broke an extremely large number of rollers off. In the end they broke so many that they put duct-tape on the things and used them as normal wheels because they were out of spares.

Now, I didn't build their robot so I can't say for certain that they had everything set up right. So I can't necessarily state that they weren't the cause of their own problems.

Are you seeing a lot of fractures at the point where the rollers meet the main body of the wheel?

[annie1939]

We used the the 6" VEXpro mecanums. We did not have any problems through 2 regionals and Worlds. Then we did demos and the wheels started breaking. Once one goes, they all start going. Either the had been weakened or once the one roller went, the others had uneven forces and they would all start breaking. We finally changed the demo bot to 4 standard wheels because we would lose a roller halfway through. At demos, the kids were more likely to run on uneven ground and start the cascade of breakages.

I would think they would last through regionals on even ground, just make sure you have a set of replacement wheels(and design for easy wheel changes) and extra rollers.

[ekapalka]

Originally Posted by Joe Johnson

Working on a spec page but 15:1 is the short answer.

Please excuse my lack of intelligence, but aside from using the Green Cord to spare your motors, what stopped you from simply using AndyMark gearboxes with that or a similar reduction (like a pair of 8.75:1 toughboxes with reduces by 2:1 on the output)? Is there some other mechanical advantage to using the gigantic sprockets? Thanks!

[JohnFogarty]

Originally Posted by Joe Johnson

Toro Horns. You are not going to believe it: Schedule 80 PVC (thicker wall than Schedule 40). I heated up some 1/2" hex shaft and melted it in place so that a Versaplanetary shaft fits tight inside.

Dr. Joe


* which roughly stated is (6) #10 holes on a 1.875" bolt circle with a 1.125" center hole.

So. We're in our shop today trying to figure out a way to expand the 1/4" Schedule 80 PVC to fit onto the 1/2" hex shaft and it seems to be quite a challenge. How did you do it?