Who thinks the IR board's operational characteristics leave much to be desired?

in other words it vehemently inhales

I would like it to have a diode but otherwise i like the idea

i could prolly build a better one myself…

Although it is a new idea, it’s VERY unreliable. I think it gets to the point that they might as well have not given it to us and let us do fully autonomous robots. It just makes the job for us programmers a LOT more complicated.

The IR board is pretty reliable by my standards.
It survived a direct short from an energized frame(reminder, ALWAYS triple check how taped your batteries are) through the board, while only destrying 3 traces, that were easily (with a fine soldering tip and tiny wire) fixed.
It lives, and the hybrid code runs excellently. The range on the recciver is very nice when your remote uses full batteries, and it lived.

We have found it reliable, and the support from the desgner outstanding.

I don’t understand the complaints about durability when subjected to reverse-voltage. Try wiring a Victor incorrectly… (I’ve seen FETs explode!)

The biggest problem it has is not getting around IR interference - other remotes can make it unresponsive to your remote. While this is clearly something you’d want them to solve, it does bring up some pretty important real life considerations. When dealing with something like the Mars Rover, NASA fully expected to have continuous contact with it, but they couldn’t plan on it. So they have to have a certain amount of autonomy built in in case something goes wrong with the communications. If, for example, the rover landed and broke it’s modem, NASA couldn’t have communicated at all. But the rover could still go around and collect data and store it internally, which NASA could retrieve at a later time. Another great example might be a remote deep see submersible. These are things that are generally controlled from a joystick on a boat, but do have certain programming if the communications drops - like return to the surface.

Regardless, when dealing with remote systems you ALWAYS have to plan for the probability of your communications dropping out at some point. In that regard, the IR boards are perfect for use in the competition.

My beef with the reverse voltage thing is that 1) a simple diode would have solved the problem, where this is more complicated with the victors, after all, they handle alot more power. and 2) the supplied ribbon cable has a BLACK wire connected to the positive pins, and a GREY wire connected to the negative pins. Needless to say this would be confusing to many teams (including my own), especially those with an electrical team thats learning.

Ours worked for Twice. Our robot has no shorts, no current through the frame, and no shock to the board. It fried. There are no exact replacements though you can buy an equivelent for around 30 bucks. The only explanation for there spontaneous breakdowns given buy the designer was something that couldve been simply fixed by adding a few components before shipping them out to the teams. I have made equivelent boards and they’ve never broken down. unfourtunately they r built in to several projects that would be destroyed if i removed them :frowning: so i guess were going to have to go with the KOP one.

Please don’t say there is no other explanation. Unless you were actually wearing static wrist straps and practice proper ESD control there is no way you can blame this entirely on the design. The amount of static that is needed to damage a component actually falls under the value needed to perceive such a shock.

First off, I’m not sure that’s the way to put it. Remember, the company that makes the IR board can read this, too.

With that said, I would have to mention- this IR board is new to us. It is something FIRST teams haven’t done before, and will most definitely have bugs. Everything does. Think about the first week of regionals, every year, with the field control and scoring systems.

The first time anyone does anything, there will be a problem. It happens, and we have to try and fix what we can, instead of complaining about it. To quote my always-correct mother (who sometimes reads CD:slight_smile: ), whining solves nothing.

It’s not Diversified’s fault at all that this board is unreliable. It’s the technology being used that’s unreliable.

Using IR is very similar to the old “Wireless” telegraph from long ago. There are no channels… anybody starts keying their telegraph (pushing buttons on their remote) and everybody hears it.

IR wasn’t really meant to have multiple unique signals going to multiple robots at the same time. Also, remember most television remotes are made to work in a living room, and I doubt many living rooms are as big as a FIRST field (or as brightly lit).

So, not at all because of Diversified but only for the sake of physics, we aren’t using an external IR input during autonomous this year. But I think our robot does pretty well without it…

Thanks Diversified for making a great donation of Time and Money! :]

-q

Old style party lines, ehhh?

I was actually thinking about that today. Wouldn’t it be nice to be able to saturate the area with IR pulsed radiation? It would be like jamming- something several Aerospace sponsors work on for the military. Then good little programmers would have to devise methods to beat the jamming, such as perhaps specific spectral IR filters to only bandpass certain wavelengths, then authenticate with a certain code first… anyway.

We’ve had no luck with the board. I looked at it when it first came in and then handed it off to another engineer. If it works at the competition, great, but since we don’t really have a programming team I’m not too hopeful :wink:

What would I find really nice to see here? Telemetry feedback. Not every team has a huge gymnasium or build center to work on- or perhaps nascar style first person view of the action. That would impress me much more than a semi-teleoperated period.

Or even lastly… stateful machine. 4 command sequences each evaluated upon what the current state of the robot is. But since that’s specifically outlawed I think our IR board will go sit next to the EDU robot.

The board does have a lot of downfalls in its design like the brownout problem and no reverse polarity protection. Obviously these are simple problems to fix (a small code change and a diode), but this is not Diversified System’s fault. All they do is put together the boards to the customer’s expectations (FIRST). If I were to give them a circuit that would catch on fire and release tons of magic smoke, then they would, and they would not be accountable for my mistake. I believe Diversified Systems also manufactured our sensor strips, and it appears to be an issue with the schematic FIRST gave them, so again, this is an issue with FIRST. Don’t get me wrong, FIRST is a great organization that has benefited me for 6 years now, but in the past two years, I’ve seen quite a few quality control issues, some of which were not FIRST’s fault, but they could have been prevented by testing the KOP that every team gets in January.

I believe the big three (Dean, Woody, Dave) could have sat down in November with a prototype KOP and tried to make a robot with the parts given, then do stress tests like that of a competition robot for four weeks, and try to break some components by making mistakes like some students may do in the build season. By doing this, they could have found the issue with the BaneBot gear boxes that plagued many teams, they may have found the issue with the radios last year, they would have found the issue with the IR Board, and they would have found the issue with the gear tooth sensor board. By doing this, FIRST would save money and would ease the burden of the build season.

No, not really. We knew that, at best, we would be able to converse with each rover twice a day and somtimes we wouldn’t be able to converse for over a day. BTW, my wife lead the telecom team in flight operations for a few years.

Nope. The autonomous mobility software wasn’t uploaded until after each rover landed (The first cut was still being worked on after they launched). Furthermore, there is no provision for either of them to go site seeing without permission from Earth :D.

-Kevin

We have found our IR very reliable, once we got the remote we had programmed to a different setting(before we were having problems programming one button, less likely four, then once solved that problem we had no problem programming the IR and using it, it has good range and everything

The IR board, like every thing else, is part of the game challenge. If you reverse the power it fries like a few other IFI parts. Avoiding this by being careful, or modifying the board with a diode, is up to you.

The “brown out” problem is one of those unpredictable unfortunate things, leading to more game challenge that we counted on, but it is still quite surmountable. We will try Kevin’s brownout fix as soon as we get access to the robot again.

I would like to suggest that those teams who have succeeded in reloading the IR board code show up at your event with a sign that reads “IR boards reloaded here.” We will be at Portland, and San Jose with our PIC programmer. Make sure that you bring the standard hex file and the one modified according to Kevin’s directions. The other teams counting on their IR boards will appreciate it.

Eugene

Our IR board stopped working shortly after being mounted on our robot. I figured it was having the brown-out problem so we tried to reprogram it with no success. Reading the device revealled a bunch of emptyness. Then I checked the regulator and it was outputting 6 volts. I don’t know if that wrecked the PIC or not, but it wouldn’t work. The team decided that the IR isn’t worth the effort due to the high probability of IR interference during the matches. We are going with an alternative means of communication.

If you don’t like the IR board, don’t use it.

We have been using a PIC16f627a with a PNA4602M IR receiver and about six lines of Pic Basic Pro code to decipher the Sony IR protocol for the past seven years as part of a mini-sumo robotics project. There is absolutely NO magic in reading an IR remote control signal… this board just makes it easier. At least until you hook up the power supply backwards.

It’s a bit more of a challenge to program the IR routine in C than Basic, but there is no reason you couldn’t just hook the Vishay sensor (or a PNA4602M) up directly to a digital I/O port on the RC and write your own code to decipher the signal.

And if you have a specific suggestion on how the board can be improved (yes! the diode would be a great improvement!) then make the suggestion.

Jason

Excellent idea! A link to the code can be found here :D.

-Kevin