MXP board design help

[Munchskull]

Recently I have started working on a custom MXP breakboard, its main design purpose is a place to mount a I2C driven IMU unite sourced from either Adafruit or Sparkfun. I feel fairly confident there, where I am running into problem is with LED based continuity signal lights I want to include to show if PWM cables are plugged in.

Long story:
The problem is the rules with the MXP boards, specifically the part about being active vs passive. To have the LEDs light up, my plan had been to run the 5V power across all the power lines for the PWM terminals. Then on the PWM ground each cabled connected to the board would have an individual path to an LED that would then connect to the shared digital ground. My thinking was that the 5V of power would go into the power line on the PWM cable, then in the component (talon, victor, spike, ect) that the PWM cable plugs into the power would then connect to the PWM ground. Back on the MXP board before reaching the digital ground the the PWM cable would power an LED. That way your LED would only light up if the Cable was plugged in at both locations.

Short story (or long story continued):
The rules (according to FRC 2015 game Manuel) do not allow active component in the PWM path way if it controls motors (it will). To my understanding LED are an active component because they are diodes.

So is there any other way that I could hook these LEDs up that would show continuity and still be considered a passive board?

[GeeTwo]

You would have to get it approved by FIRST, but my recollection of active vs passive was whether the signal was modified between the 'RIO and the motor controller. If the LED was used purely as an indicator (not as a rectifier), you should be able to get it approved.

[Munchskull]

I am sure FIRST would allow it but my understanding is that I would then have to have a way to market the board to other teams, making it fiscally impossible to make a custom board.

[marshall]

Originally Posted by Munchskull

I am sure FIRST would allow it but my understanding is that I would then have to have a way to market the board to other teams, making it fiscally impossible to make a custom board.

I think you'll be ok and either way you should make it. It's a good learning opportunity.

Which EDA software package are you using to design the board?

[Munchskull]

Originally Posted by marshall

Which EDA software package are you using to design the board?

I have been using PCB express. It was free and I could figure it out easily.

[marshall]

Originally Posted by Munchskull

I have been using PCB express. It was free and I could figure it out easily.

A license for Altium comes with the kit and I'm a big fan. They also just released a community edition of it called CircuitMaker that seems good too: http://www.circuitmaker.com/

EEVBlog just did a review of it: http://www.eevblog.com/2015/06/16/ee...t-impressions/

[Michael Hill]

Originally Posted by marshall

A license for Altium comes with the kit and I'm a big fan. They also just released a community edition of it called CircuitMaker that seems good too: http://www.circuitmaker.com/

EEVBlog just did a review of it: http://www.eevblog.com/2015/06/16/ee...t-impressions/

Getting the license for Altium is a PAIN. IN. THE. ***! I've been able to get it every time I ask, but from the time you request your license, it can take weeks for it to actually get to a working state. That said <3 Altium Designer.

[marshall]

Originally Posted by Michael Hill

Getting the license for Altium is a PAIN. IN. THE. ***! I've been able to get it every time I ask, but from the time you request your license, it can take weeks for it to actually get to a working state. That said <3 Altium Designer.

Really? I haven't had a problem for the last few years. I'm also a big Altium fan. It's good software.

[SteveGarward]

Originally Posted by Munchskull

I am sure FIRST would allow it but my understanding is that I would then have to have a way to market the board to other teams, making it fiscally impossible to make a custom board.

This is correct. However that does not prevent you from making it to try things out, which I highly recommend.

This year we made a number of custom boards. After our initial boards made through the beta test period, we decided with the current rulings, we were best to avoid anything that would potentially present itself as a problem during inspection (due to interpretation of rules), and made 3 boards which would all communicate over I2C.

Due to space issues, we ended up cramming all three boards into one Mega Ultra Maxi board, which would plug into the MXP directly. Essentially, this board houses 3 Arduino clones, repackaged to suit our needs. These all still only communicated over I2C. The presence of rows of pins for our hall effect sensors always raised some questions during inspection, since they look like PWM headers. Ultimately, the flow chart given in the team update was our friend - we could point to it and say 'it doesn't control anything - it's just a custom circuit'. When inspectors see ANY components on an MXP board, the initial reaction has been to call it an ACTIVE board.

Where was I going with this?

I'd stay away from altering or using any power or signal pathways on any IO/PWM pin from the MXP. As much as it would be nice to have it, if it rules your board illegal, where you could have the board without LEDs and not have to set yourself up as a vendor and have distributions channels, I personally think that would be the better route to take. You can still make one that lights up for practice/testing/off-season, and who knows what the rules will state next year!

[GeeTwo]

Originally Posted by SteveGarward

I'd stay away from altering or using any power or signal pathways on any IO/PWM pin from the MXP. As much as it would be nice to have it, if it rules your board illegal, where you could have the board without LEDs and not have to set yourself up as a vendor and have distributions channels, I personally think that would be the better route to take. You can still make one that lights up for practice/testing/off-season, and who knows what the rules will state next year!

Or you could put 2-pin headers where the diodes are on the circuit board and plug diodes in for in-shop testing, and pull them for competition. Even easier to use would be a DIP LED strip; just one component to pull/place.

[protoserge]

Originally Posted by Munchskull

I am sure FIRST would allow it but my understanding is that I would then have to have a way to market the board to other teams, making it fiscally impossible to make a custom board.

I believe you could achieve this by posting the board to OSHPark where teams could order it as a set of 3 bare boards.

[Alan Anderson]

Originally Posted by Munchskull

...To have the LEDs light up, my plan had been to run the 5V power across all the power lines for the PWM terminals. Then on the PWM ground each cabled connected to the board would have an individual path to an LED that would then connect to the shared digital ground. My thinking was that the 5V of power would go into the power line on the PWM cable, then in the component (talon, victor, spike, ect) that the PWM cable plugs into the power would then connect to the PWM ground. Back on the MXP board before reaching the digital ground the the PWM cable would power an LED. That way your LED would only light up if the Cable was plugged in at both locations.

I don't understand how this is supposed to work. Can you post a schematic that shows the connections, especially the part that provides the LED current?

[teslalab2]

We made our own mxp board with a perf-board then we printed the shell, we didn't to get it approved by first because its only for sensors, it would be considered active if it was driving motors, or something, at inspection they just simply asked if it was active or passive and they also wanted to see what it looked like on the inside.

[Munchskull]

Originally Posted by Alan Anderson

I don't understand how this is supposed to work. Can you post a schematic that shows the connections, especially the part that provides the LED current?

Sorry I am currently in transit and do not have acess to me schematics. I will post them later today. For now I will try to explain a little better. +5v goes to the power pin on the PWM terminals. The ground pin on the PWM terminal leads to a single LEDs positive side, then the negative side of and LED would lead to digital ground. Give me three minutes and I can up load a hand drawing.

[Alan Anderson]

Originally Posted by Munchskull

+5v goes to the power pin on the PWM terminals. The ground pin on the PWM terminal leads to a single LEDs positive side, then the negative side of and LED would lead to digital ground.

Are you expecting to power the LED using the ground current from the speed controller's PWM connection? That might work if the LED were connected in the other direction, and if the controller's optically-isolated input will still function with the extra LED's additional voltage drop in the circuit. But it can't work if the robot is disabled, and I can't see it telling you any more than what the LED on the speed controller already does.