Electrical / Software test bed

[smclean1969]

We are looking at building a test bed that would allow us to load code and test various prototype electrical components. We've had a similar test bed that allowed electrical connections but nothing with the software. We're thinking we'll need the follow components:

- battery
- power distribution board
- Roborio
- Talons
- Laptop that would be connected to the Roborio for sending commands
- electrical and ethernet connections

Has anyone built a similar setup? Can anyone think of anything to add onto this or how to improve the basic concept? Planning on putting this together after the season.

cheers,

[CalTran]

Spare motor / actuators to actually see if the code is working?

[lethc]

You may want to add a pneumatics system depending on whether or not you feel that you need to practice with pneumatics.

[tr6scott]

http://playingwithfusion.com/productview.php?pdid=65

If you are a labview team.
We also built the simbotics Kit bot on Steroids with the 4 slot crio that is still supported in labview, with the old control system.

We also have pneumatic test board, that kept getting ripped apart whenever we needed parts. Last offseason, we rebuilt it once more, this time with non-legal FRC parts. Guess what, it is still together today.

[Coach Seb]

Are you coming to Marysville this weekend? Could you bring one of these, would like to check it out....

[GeeTwo]

Window motors are great for test beds, because you can easily see what direction they are turning without adding wheels.

We try to use the same type of controller/relay as we have on the robot to reduce/eliminate the changes needed between the testbed and the actual robot.

We have also included feedback on some of our test beds (often manually activated limit switches). We usually start with about twice as big a board as it looks like we'll need, and end up filling it by the end of the season.

[BrianAtlanta]

One issue with a programming bot is keeping it close to the competition bot, if you want to develop code while the competition bot is being built. Depending on team resources (students and money) can be a challenge.

I've been thinking this season about how start developing competition code sooner in the season. One idea I've been thinking about was prototyping a electrical harness. Here's my current thinking, using this years game. Please, suggestions are welcome.

1. After strategy, build breaks up in to 4 groups for prototyping (drive train, shooter, intake and lift)

2. Once a design is selected from each group, build starts designing and building. But at this point, electrical knows the components needed. They might change with further build enhancements, but you have a starting point.

3. As the build team works, a person or two take the required components and create a prototype harness on polycarb, rio, radio, pdp, and which ever components needed of the the subsystem.

4. Programmers can program, granted driving forward 5ft is hard to test without a bot. But you could tell that the bot motor started spinning and stopped. Once you get on the bot, then you tune the values.

5. Once build is ready for wiring, the test harness is used on the bot.

Brian

[Jaci]

I use a fairly simple Test-Bed for testing OpenRIO code during the Offseason. Here's a few suggestions:

- Battery: Don't do this. Use a 12V wallwart if you're powering just a RIO / Radio, it'll save you so much time, especially if you're testing code for long periods of time. If you're using actual motors, then the battery is a good idea. Just be prepared to charge it.

- Radio: Honestly don't even bother. Just link your computer to the RIO using USB, or your RIO to a Local Network (you could use an old D-Link router for this, just turn off the wireless). Waiting for windows to find the wireless network each time you reboot the testbed isn't worth it, and it also means if you're on the Local Network you'll have Internet Connectivity without switching networks if you need to troubleshoot something or procrastina-- help the community on Chief Delphi

- Motors: Use small, low-power motors that aren't very fast. If they aren't very fast, it'll be easy to tell which way they are turning and how fast (as a rough percentage). Low-power means you don't have to swap out your battery very often.

[GreyingJay]

Depending on what exactly you want to do, you could definitely power it all using a 12 volt wall wart (use the highest current output you can find) or a 12 volt bench power supply (5-10A or more).

If you don't need real motors and motor controllers, you can plug in servos to stand in for motors. Continuous rotation servos would give you a good visual indicator of movement. If you're not using motors or other high current draw devices, you can probably dispense with the PDP entirely.

How many RoboRios do most teams have? If we had the resources I'd want at least three: one set up in a testbed configuration like this, another one mounted in a full drive platform (either purpose built, or reusing a robot from previous years) for autonomous testing, and then the competition robot.

[virtuald]

I'm fairly lazy, so I don't bother with a testbed... I just use a low-fidelity robot simulator (pyfrc), and we can simulate most robot actions without further modifications.

[tr6scott]

Originally Posted by Coach Seb

Are you coming to Marysville this weekend? Could you bring one of these, would like to check it out....

If you are talking about the linked robot, I do not have one... Someone on CD pointed it out and I had a dad on the team talk about buying one.

If you are talking about the pneumatic test board, I can toss it in.