What COTS parts would you like to see?

[JesseK]

I would like to see a better solution for using LED's. Right now LED's paralyze my team with how many choices / setups there are, so usually there's a mediocre attempt at them but never a finished product.
- How do we connect power to the LED's?
- Which LED brand, model, and length do we use?
- Do we really have to care that each light is addressable, seriously?
- To control them do we use an offboard processor, use a Spike, or is there a better way?

Would love to have something simpler that doesn't depend on in-house custom circuitry or an arduino. Plug this 12" LED strip into this device, then plug this device into the PDB and into the RoboRIO using these connectors. Then use this sample code to get going with writing different colors/brightnesses or blinking lights to the LED's.

FWIW, introductory courses in embedded electronics in college were all about blinking LED's, and connection of those circuits to the corresponding code was way more straightforward than the FRC LED situation.

[Bkeeneykid]

Originally Posted by JesseK

I would like to see a better solution for using LED's. Right now LED's paralyze my team with how many choices / setups there are, so usually there's a mediocre attempt at them but never a finished product.
- How do we connect power to the LED's?
- Which LED brand, model, and length do we use?
- Do we really have to care that each light is addressable, seriously?
- To control them do we use an offboard processor, use a Spike, or is there a better way?

Would love to have something simpler that doesn't depend on in-house custom circuitry or an arduino. Plug this 12" LED strip into this device, then plug this device into the PDB and into the RoboRIO using these connectors. Then use this sample code to get going with writing different colors/brightnesses or blinking lights to the LED's.

FWIW, introductory courses in embedded electronics in college were all about blinking LED's, and connection of those circuits to the corresponding code was way more straightforward than the FRC LED situation.

Thing is, most LED strips are one of two solutions: RGB LEDs, or Addressable WS8211 or similar LEDs. I know teams that I know (I've never used LED Strips on a robot before) use the VRM's 2 Amp 5 Volt rail to power LEDs. I'd recommend one of two models: Adafruit Dotstar or Adafruit NeoPixels. These are both addressable. In reality, you don't need addressable really, but it makes controlling them easier, and sometimes it's just nicer to have control of individual sections. The main difference between the two, is that one (the Dotstar) is able to be controlled directly from the RoboRIO, albeit at a higher price. The NeoPixels are much more common, but require a special timing protocol, which must be driven by an Arduino or some other single process device. You may be able to find places that have basically identical strips to these Adafruit brands, just search for WS8211,WS8212,WS8212B, or WK6812 LED strips on Ebay or other shopping website. DotStar strips use a less common APA102 driver on the chips, so you'll find less of these types of strips on places like Ebay.

[JesseK]

Originally Posted by Bkeeneykid

Thing is, most LED strips are one of two solutions: RGB LEDs, or Addressable WS8211 or similar LEDs. I know teams that I know (I've never used LED Strips on a robot before) use the VRM's 2 Amp 5 Volt rail to power LEDs. I'd recommend one of two models: Adafruit Dotstar or Adafruit NeoPixels. These are both addressable. In reality, you don't need addressable really, but it makes controlling them easier, and sometimes it's just nicer to have control of individual sections. The main difference between the two, is that one (the Dotstar) is able to be controlled directly from the RoboRIO, albeit at a higher price. The NeoPixels are much more common, but require a special timing protocol, which must be driven by an Arduino or some other single process device. You may be able to find places that have basically identical strips to these Adafruit brands, just search for WS8211,WS8212,WS8212B, or WK6812 LED strips on Ebay or other shopping website. DotStar strips use a less common APA102 driver on the chips, so you'll find less of these types of strips on places like Ebay.

I get it - there are options. Yet they're wide, varied, sometimes incompatible and sometimes take the power ports we'd rather put to other uses.

Or perhaps CTRE could come out with a couple more 5V ports on the VRM?

[Greg Needel]

Originally Posted by JesseK

I would like to see a better solution for using LED's. Right now LED's paralyze my team with how many choices / setups there are, so usually there's a mediocre attempt at them but never a finished product.
- How do we connect power to the LED's?
- Which LED brand, model, and length do we use?
- Do we really have to care that each light is addressable, seriously?
- To control them do we use an offboard processor, use a Spike, or is there a better way?

Would love to have something simpler that doesn't depend on in-house custom circuitry or an arduino. Plug this 12" LED strip into this device, then plug this device into the PDB and into the RoboRIO using these connectors. Then use this sample code to get going with writing different colors/brightnesses or blinking lights to the LED's.

So we built a prototype of a product about a year ago to do simple LED control called the "blinkin Board"

The basic idea was to allow on a single board the ability to drive the WS281* LEDs or 12v RGB LEDs. The voltage buck 12 to 5 with 3 amps of supply is on board, so just give it 12 volts and it would work.

The real beauty (in our minds) was the fact that you can control either kind of strip with a PWM input, we were able to sub-divide the standard PWM signal into tons of discrete signals and using a provided chart teams could make their LEDs do things with no hard programming required. There are a bunch of jumpers on the board so you can select things like your team colors & strip length.



We got the prototype almost ready for production (needed a bit more DFM work and software) but we ran out of time as we needed to focus on other things (like the SPARK). We also were not sure how many teams would be interested in something like this to justify the effort to bring it to market. It is still something we can definitely make, we just need to figure out where on the priority list it falls compared to some of the other large initiatives REV is working on.

[nuclearnerd]

Originally Posted by Greg Needel

So we built a prototype of a product about a year ago to do simple LED control called the "blinkin Board"

We got the prototype almost ready for production (needed a bit more DFM work and software) but we ran out of time as we needed to focus on other things (like the SPARK). We also were not sure how many teams would be interested in something like this to justify the effort to bring it to market. It is still something we can definitely make, we just need to figure out where on the priority list it falls compared to some of the other large initiatives REV is working on.

Oh I would definitely buy this. The need for a dedicated power supply and controller has kept me from prioritizing blinkinlights in the past. Your board looks like a pretty simple drop-in solution.

[mrnoble]

Originally Posted by nuclearnerd

Oh I would definitely buy this. The need for a dedicated power supply and controller has kept me from prioritizing blinkinlights in the past. Your board looks like a pretty simple drop-in solution.

I'd be all over this too. I sure hope your market research shows demand, I'd love to see it get built.

[mrnoble]

I think it was a little late last year when I voiced a hope for inverse mounted CIM gearboxes (like http://www.chiefdelphi.com/forums/sh...d.php?t=140713) as a COTS item; maybe this year the Wizards at WCP will bring one to market. I'd also love to see the Raw Box made more like Joey Milia's original version; my team would probably keep several sets of that box in stock, it seems like one of the better designs I've ever seen.

[jwfoss]

Originally Posted by Greg Needel

So we built a prototype of a product about a year ago to do simple LED control called the "blinkin Board"

A system like this definitely makes LEDs less cumbersome. 558 would be very interested in this set up.

[Chak]

A bigger version of the versaplanetary.

[nuclearnerd]

Some of these are (unmet) repeats from my 2015 wishlist:

1. seconding a larger versa planetary. A stronger output stage (with more than two extremely thin #10 mounting holes), longer output shaft (on two bearings maybe, steel ring gears, aluminum encoder stages, and would actually installing the anti-torque pins at the factory would help. We went through hundreds of dollars of busted stages this year For larger arms, teams shouldn't have to build their own chain stage - a strong final stage could work as a direct drive!
2. A system of hubs and bearings for 1" OD tube axles. 1/2" 7075 hex is not up to the challenge of transmitting large torque, over long distances with little twist. 1" tube would fit in AndyMark and Vex bolt circles, and be a lot stiffer
3. Repeated from up-thread: A pneumatic brake and/or ratchet stage for a versa planetary. This could be small and lightweight to go on the motor side for mechanism holding, or beefy to go on the output side for something like a catapult release
4. A clutch stage for a planetary output (or maybe just something that slips on a hex shaft. This would let teams use air-cooled motors like the 775 pro without worrying about stalling and burning them out
5. Something like Vex's 217-4183 bearing mount, but longer, with set distances between holes to fit standard belt-pulley / chain-sprocket spacings. Teams that don't have machining resources have a heck of a time mounting pulleys at the proper center spacing
6. Longer ribbon cable for talon SRX's, and/or cheaper breakouts

I'll probably come up with more...

[JesseK]

Originally Posted by nuclearnerd

A clutch stage for a planetary output (or maybe just something that slips on a hex shaft. This would let teams use air-cooled motors like the 775 pro without worrying about stalling and burning them out

Hmm, I generally really like this one...

[Munchskull]

I have mentioned this before but I would love a VP output stage with a female 1/2" hex coupler.

Also more small usable motors would be great. Stuff in the RS-550 size.

Said this earlier but an electric solenoid and coupler for shifting with no pneumatics would be epic.

[Ari423]

Originally Posted by Munchskull

I have mentioned this before but I would love a VP output stage with a female 1/2" hex coupler.

Why not just use the male 1/2" hex output shaft and one of these 217-4008?

[Sperkowsky]

Originally Posted by Ari423

Why not just use the male 1/2" hex output shaft and one of these 217-4008?

Its a matter of form factor.

[Munchskull]

Originally Posted by Ari423

Why not just use the male 1/2" hex output shaft and one of these 217-4008?

Because it would be nicer just to take a piece of hex shafts of any length and have that be your output.