734 replies to this topic

[cyril92]

Hi mjr,

Sorry for my newbie question, I apologize in advance : but for the plunger we need to wire 2 blue leds . Could you please tell us a little bit more on how to wire them .... because I read that considering the different voltage we use, we need to put different resistances if we don’t want to burn the leds ...


Cyril.

[parabolic]

Ok - NEXT STEP : TABLE CONFIGURATION!!!

So to not derail this topic on the plunger build and setup I have a question regarding adjusting tables to work with the plunger over here:

http://www.vpforums....showtopic=28926

 

Need some help on it....

Thanks!

[jkimbrell]

Try increasing the tilt value.  It's backwards of what you would assume.  A setting of 0 is no tilt.  Try setting it to 750 or 1000.

Edited by jkimbrell, 01 September 2014 - 12:50 AM.

[mjr]

Sorry for my newbie question, I apologize in advance : but for the plunger we need to wire 2 blue leds . Could you please tell us a little bit more on how to wire them .... because I read that considering the different voltage we use, we need to put different resistances if we don’t want to burn the leds ...

 

I guess I did leave that pretty vague, now that you point it out! 

 

You're right that you need resistors with the LEDs.   The exact resistor value you need depends on the type of LED and the voltage of the power supply you use.  There are on-line LED resistor calculators that do the math for you, like this one: http://led.linear1.org/1led.wiz.  You need three parameters: the forward voltage and forward current of the LED, and the voltage of the power supply.  The forward voltage and current should be on the LED packaging or should be available from wherever you bought them.  Typical numbers for the little glass bubble LEDs are 20mA for the current and around 2.4V for red LEDs and 4V for green and blue LEDs.

 

Once you know the resistor size, the actual wiring is straightforward.  Just wire the long leg of the LED to the positive (+) voltage on the power supply, wire the short leg to a resistor, and wire the other end of the resistor to the negative on the power supply (also known as 0V or ground or common).

 

Hope this helps!

[cyril92]

Yes it's crystal clear.

 

Thank you mjr

[Zablon]

This looks awesome! Can someone direct me to the docs for this? The links back in page 3 or 4 are dead.

[mjr]

This looks awesome! Can someone direct me to the docs for this? The links back in page 3 or 4 are dead.

 

Everything is on the mbed project page now:

 

http://mbed.org/user...ape_Controller/

 

(Sorry about the dead links; I initially posted the files on dropbox as a convenient default, but once I had the mbed project page set up, I consolidated everything there.  I just went back and edited my posts with the old dropbox links to point to the mbed page.  I think I got them all - if you run into another dead link in one of my posting, let me know and I'll fix it.)

[did014]

Hello,

 

this is great, the nudge is amazing

 

Thanks for your work !!!!

Edited by did014, 01 September 2014 - 06:40 PM.

[mrarcade]

I ordered some items today, including the freescale board from mouser.

[mrarcade]

The free scale board is here and I have it programmed and win recognizes it as USB game controller. So far so good.

I didn't read enough before ordering from mouser. I guess I should have ordered some pin headers, crimp terminals, and wire housings. The shopping list has an 8+8 and a 10+10. Is there a reason not to get another 8+8 and a 6+6 for j10 & j9?

[mjr]

I didn't read enough before ordering from mouser. I guess I should have ordered some pin headers, crimp terminals, and wire housings. The shopping list has an 8+8 and a 10+10. Is there a reason not to get another 8+8 and a 6+6 for j10 & j9?

 

It certainly wouldn't hurt - makes for a cleaner install if you do.  On mine I soldered the wires for the handful of connections to those jumpers, and put in a couple of free-hanging cable connectors (the Molex .062 type), one for the plunger sensor and one for the calibration button.  My reasons for going that route are explained around p10 ("Wire the plunger sensor"), but briefly, there isn't a way with the pin layout to wire the plunger sensor to a single header - it has to be connected to pins on both J9 and J10 - so I found it more convenient to set up one Molex connector to the plunger rather than wiring it to two pin headers.

 

What you could do, and what I might do if I were to build another one, would be to put pin headers on all four jumper pads (J1, J2, J9, J10) and then wire the Molex connectors to those.  It's extra connectors but it would be a little cleaner for servicing.

[mameman23]

Yup ordered up the headers for mine, should make for a really clean and easy install into a cab.

[mrarcade]

Which 6+6 header and housing did you order? Can the housing be handled like a PC case header....using multiple 2 pin, 4 pin housings on the same 6-10 pin header?

[mameman23]

Just ordered some 10 pin units off eBay and will cut the length down on the shorter ones if needed. I'll dig the link out.

[sliderpoint]

Just wanted to post a follow-up so say how well this works. The plunger is awesome!

 

Thanks and nice job MJR!

 

-Mike

[mrarcade]

I made a quick file to help me visualize where things go on the freescale headers so I can decide which housings to order.

I want to say up front that I am not complaining or not happy with the pinscape controller. I am just curious if it would be easy to make some adjustments, I have never programmed, so I have no idea. There might be a simple reason or reasons why the following won't work. I am just wondering...

 

I am curious about the unused pins.For instance, pin J10-12 is unused. Could button 4 be moved from J10-4 to j10-12? Then three 1x2 housings would take care of the button inputs from J10, a 1x2 would take care of the calibrate and a 2x2 would take care of the plunger ccd. Similarly, button 24 could be moved from J2-20 to J2-15. Either five 1x2 or one 1x10 would take care of the button inputs from J2. Ledwiz port 22 could be moved from J2-18 to J2-12 so three 1x2 or a 1x6 would cover the ledwiz ports on J2.

 

Again, I am in no way complaining, I am guessing there are reasons for the unused pins and for the gaps etc, I hope I am not going to upset anyone for asking.

 

[mjr]

Which 6+6 header and housing did you order? Can the housing be handled like a PC case header....using multiple 2 pin, 4 pin housings on the same 6-10 pin header?

 

I don't see why that wouldn't work.

 

Be sure to order the plain vertical pin headers, like these:

http://www.mouser.co...855-M20-9760846

 

Specifically, don't get the socket or housing type with a wall around the outside.  The pads on the KL25Z are too close together for two of the walled ones to fit side by side on J1-J2 or J9-J10.  You need the plain pin-only type.

Just wanted to post a follow-up so say how well this works. The plunger is awesome!

 

Really glad to hear it's working!

[mjr]

I am curious about the unused pins.For instance, pin J10-12 is unused. 

It's entirely likely that there's a better way of assigning the pins.  It's an incredibly complex web of dependencies and constraints because of (a) the way the KL25Z is physically wired, and (b) the way the mbed system library uses the pins.  And I had to piece together the constraints from multiple sets of documentation - no one seems to have compiled the full set of constraints anywhere.

 

J10-12 is PTC1, which is assigned by the mbed system library to the real-time clock.  The library sets things up with the MCU mux to assign that pin internally.  It happens to also be attached to this jumper pin, but we can't use it for any other purpose because it's physically connected to the on-board clock signal.

 

 

Similarly, button 24 could be moved from J2-20 to J2-15. Either five 1x2 or one 1x10 would take care of the button inputs from J2. Ledwiz port 22 could be moved from J2-18 to J2-12 so three 1x2 or a 1x6 would cover the ledwiz ports on J2.

 

J2-15 is physically not connected to anything on the board.  I'm not a good enough coder to do anything about that!    It's very odd that they left this disconnected, since there are numerous GPIO ports on the MCU that aren't connected to pins.  You'd think they would have found an otherwise unused port and connected it here, but alas...

 

And J2-12 is wired to the blue segment of the on-board RGB LED.  I think it's possible to repurpose that pin by cutting a trace on the bottom of the board, but I didn't want to do that or require anyone else to do that; plus it would lose the diagnostic LED function (the blue segment, at least).

 

Sorry to put the kibosh on these ideas!  But note that you certainly can move things around - I tried to make it really simple in the software to do that by putting all of the pin assignments into the configuration block at the top of the main source file.  It should be fairly self-explanatory if you take a look at the code.  mbed is set up to let you create your own fork of the repository if you want to make custom changes like that, so feel free to experiment - there's no danger of breaking the public version since your changes will go in a separate branch.  If you have any other ideas for reassignments that you want to run by me, free free to post them here - I can try to advise on whether or not they'll run into any of the hardware obstacles such as the ones above.

Nice pin-out diagram, by the way!

Edited by mjr, 12 September 2014 - 05:15 AM.

[mrarcade]

Here is what I came up with:

Pin headers: two 8x2, one 10x2, one 6x2.

Pin Housings:one 8x2, one 2x2, two 9x1, one 8x1, one 3x1, six 2x2. I will not be using button inputs 4 (in 2x2 ccd header) or 6 (nothing attached)

 

Opinions? I plan to order within the next few days

 

[mjr]

That looks like it'll work - complex, but I can't see how to simplify it given the pin layout constraints.  

 

The only other approach I could see would be to use housings that match the headers, and run wires from the housings to a set of Molex .062 connectors (or something similar - some kind of free-hanging wire plugs).  That way you could use four Molex connectors that map to the functional groups - e.g., a 24-pin plug for the buttons, another 24 for the LedWiz outs, a 5-pin for the CCD, and a 4-pin for the calibration button.  Each Molex connector would connect to multiple housings, and vice versa, so it'd be a little bit of a web, but everything would still be completely modular.  The total number of connectors would be the same in both setups (4 headers + 12 housings in your setup vs 4 headers + 4 housings + 4 plugs + 4 receptacles in this setup), but it would be a little easier to figure out what plugs into what, since almost everything would be uniquely mated pairs.  It'd be a little more work to wire.

Edited by mjr, 13 September 2014 - 07:48 PM.