TCCoA Forums banner

1 - 8 of 8 Posts

·
Registered
Joined
·
2,634 Posts
Discussion Starter #1 (Edited)
I thought some of you guys might appreciate this new comm system I built for my 24 Hours of Lemons track car (endurance racing).
UPDATED March 2017 - Ver 2 now

PROBLEM
* Since our first race back in 2010, we've relied on a DIY system involving some proprietary digital spread spectrum FRS radios (it's digital so you don't step on anyone else's chatter even with 170 other teams on track), a DIY antenna, and some DIY headsets.
* Even using a "motorsports" connector (one plug with microphone+speaker), the wiring inside is very, very thin and I'm sick of repairing headsets when people forget to disconnect when jumping in/out of the car.
* The headsets never really worked anyway after we started renting a pit garage space at Sears Point -- they are essentially giant faraday cages. Even with a 30' antenna sticking out of the garage window track-side, the grandstands block your line of sight to most of the track.

REALIZATION
* I don't want anyone chattering in my ear while trying to keep my tbird away from the 169 other idiots on the track. We are fast enough now with a V8 that we are trying to weave through the traffic (vs just trying to thread our way fron the front of a pack if faster cars to the back of the pck).
* The driver really only needs to tell the pits one thing "hey, I'm coming in."
* The pits SHOULD have the ability to signal the driver to come back in though.

MY EPIPHANY
* Cell phones work inside the Sears Point garages.
* Arduino compatible cell phone modules are really cheap (esp 2G ones). AT&T will be turning off 2G service by EOY 2016 but TMOBILE doesn't show any signs of turning away all the cheapie IOT biz.

INGREDIENTS
(1) Sainsmart ARDUINO Nano ($11.99)
(1) NRG Steering wheel with buttons on it (already had it. Because 2 fast 2 Furious).
(1) Sainsmart Small GSM SIM900 comm module ($27.99)
(1) LION Battery: needs to output 2A for the comm module
(1) SIM with a few bucks loaded in. NOTE: the really sketchy cheap guys, FreedomPOP, use an android app for SMS texts so it won't work with my module. (TING is a Tmobile MVNO who offers SIMs for $1 on sale. 1 month of service with 1K SMS cost a mere $11)
(MISC) Small proto board, connectors, scrap box, USB cables, etc - <$5.
(1) 12V to USB 3A DC-DC adapters hardwired into car. You need two because the Cell modem takes too much power from the converter for the arduino to boot correctly
http://www.ebay.com/itm/DC-DC-Converter-Module-12V-To-5V-USB-Output-Power-Adapter-3A-15W-/231458070635


WHAT THE SYSTEM DOES - VER 3
* SMS texts the pit crew upon boot up "Ready to Rock". Current code sends messages to ALL drivers.
* Steering wheel buttton replaced with a two switch panel mounted where the sun visor used to be. Even with a helmet + hans, a driver can quickly operate the system without looking up and he won't accidentally trigger it.
- Right switch = New Driver/More fuel
- Left switch = Repairs needed
- BOTH Switches = OMG! Some **** went wrong. Get ready for big repairs.
* Pitcrew only texts car when THEY want the car to return to the pits (no acknowledgement). If the correct callerID and phrase is matched (screw you, rival team haxxors), a giant red LED light is illuminated in front of the driver.

PROBLEMS ENCOUNTERED
* Driver/reference code availability. I eventually found a reference to a usable GSM library and sample code in someone's product review on amazon. Thanks, more-determined-nerd-than-me.
* It took me a bit to realize that the comm module will work with only 100mA of current to stay on network but needs up to 2A to transmit. If you give it too little current, it starts acting squirrelly.
* When sending multiple SMS messages, you need to wait a decent amount of time (10 sec delay) between sending messages or the modem (which works asynchronously from the Arduino) will mess up. The alternative is to redo the GSM library to actually read back the AT commands the cell modem returns vs the "fire and forget" mentality it uses today.


LESSONS LEARNED AFTER FIRST RACE DEPLOYMENT
After my 7th Lemons race, I learned two key lessons:
1) Do NOT put the button on the F&F style race wheel because someone will tap it accidentally. Move it. Even better, don't make it a button -- make it a switch.
2) The communication protocol should be that the Pit crew only turns the "RETURN TO PIT" light on for the driver when THEY want the driver to pit and not as an acknowledgement of msg received. if the driver wants to pit, hes going to do it anyway. We had two inadvertant pit stops when the driver accidentally pushed the button, the pitcrew acknowledged the message, and the driver saw the light and thought that the pitcrew wanted him to come in. Whoops.
3) You need more info from the driver to the pit to decide what to do. For example, for fuel or driver change stops, you need someone on the pitcrew suited up in a driving suit. For repair stops, you need the "mechanics" of the team (we have two mechanics and one chef) suited in their repair coveralls and ready to work.

STATUS
1) Works Great




 

·
Super Moderator
Joined
·
10,408 Posts
Being an electronics guy:

Use Power tool batteries for power for this kind of stuff; a cheap C3 flashlight is handy to make a C3 Li-ion socket. (craftsman)

A newer Raspberry pi can run arduino code; but if you have Linux guys on your team, Pi's rock. :)

You really wouldn't believe what we've run on a Pi with an FPGA front end... :grin2:
 

·
Registered
Joined
·
2,634 Posts
Discussion Starter #3
Being an electronics guy:

Use Power tool batteries for power for this kind of stuff; a cheap C3 flashlight is handy to make a C3 Li-ion socket. (craftsman)

A newer Raspberry pi can run arduino code; but if you have Linux guys on your team, Pi's rock. :)

You really wouldn't believe what we've run on a Pi with an FPGA front end... :grin2:
Interesting comment about using a Sears LION battery pack.
For my needs, I've found that the USB battery packs (power banks) that have become all the rage in the last few years work like a charm. All of them output 5V (in the change of what I need for these embedded modules so no voltage conversion is required) and with 18650 cells being so cheap nowadays, you can get them with 5-15K AH without too much fuss. Challenges:
* I've found that some are "too smart" and shut themselves off if the devices plugged into them don't draw enough current.
* As with all these no-name sketchy chinese products, the current supplying capabilities and/or actual AH capacities tend to be all over the place (just like with rechargable NIMH AA/AAA cells). It helps to read the Amazon reviews closely as with enough buyers -- one of them will turn out to be an obsessive nerd who will test things for you.
* For the DIY set, I've also found some power banks that are shipped empty. This may be great for people who reclaim old 18650 cells from laptop packs or have a bunch of good ones but honestly, I'd rather stick to the units that are available.

---

Q: What do you use rasberry PI boards for?
I've always found that there's another device that does it better even if it's not as small as a Pi. Here are the popular projects I've read about:

1) Home Theater PC - I'd rather just have a full blown PCI with a better GPU for the occasional game

2) NAS Server - again, a Win 10 box does this just fine even if it's bigger

3) For embedded projects, my use cases have always been pretty simple control code projects which mostly involve: read data, make a decision about what to do about it, and send it out. There might be multiple functions but probably only a few interrupts beyond the main loop. Arduino has been perfect for my needs. Once they are built/deployed, they are set with little/no need to alter them.

Prior to using an Arduino for my gauge monitoring project, I tried using an STM32 Discovery Board a friend gave me. I ultimately shelved that thing b/c the lack of infrastructure was a PITA (eclipse based tools were meant for much more substantial developments than what I was doing). Even with the EE degree (albiet 16-20 years out of college now), just creating a basic project by trimming one of their demo packages took days of fiddling. In comparison, I got the base code of my Arduino project working on a breadboard up and running in 30 minutes.

I suppose if I was building more complicated devices (say a fireworks controller with banks of dependencies that might change from show to show) or something to automate production in a machine shop that may change over time, I might consider a larger processor than Arduino. However, I'm not sold that Linux makes the most sense for this embedded OS (even though linux means connectivity with an IP network is a snap). I would suspect that a RTOS would make more sense.

Q: So what do you do with a Pi and an FPGA front end? Audio processing/filtering?
 

·
Registered
Joined
·
2,472 Posts
I run a Pi 2 for KODI/XBMC, still use my desktop as the server though, as for graphics it does 1080i just fine, add a powered USB Hub and it is a DVD/BR player too.

I'm currently thinking of using a PI for a print server and NAS, would like to simplify what my PC does.

I'd say be the biggest benefit of/for a PI is LINUX.
 

·
Registered
Joined
·
3,726 Posts
Cool project, way to solve an issue in a creative way.

I have never really created anything grand with an Arduino, but I love to play around with them, moving servos, stepper motors, making LEDs blink, etc. I did create a module for my Mark VIII that 2 seconds after it starts it "presses" a button twice, another button once, and a final button once to disable traction control and set it to show average economy. That was so small and with only 3 I/O I programmed it into an ATTiny. Something similar for Leaf owners to disable the fake car noises so pedestrians wouldn't walk behind them when backing up. Played around with GPRS when that was a thing, and running a web server and some networking.

I am currently using a PI to create a small arcade machine that runs emulators for old school arcade games and consoles. Working much better than a low end atom and USB interface board, just using the GPIO.

I haven't had much use for the STM32, other than a little background into how to unlock and flash quad copters and fine tune them.
 

·
Registered
Joined
·
2,634 Posts
Discussion Starter #6
Cool project, way to solve an issue in a creative way.

I have never really created anything grand with an Arduino, but I love to play around with them, moving servos, stepper motors, making LEDs blink, etc. I did create a module for my Mark VIII that 2 seconds after it starts it "presses" a button twice, another button once, and a final button once to disable traction control and set it to show average economy. That was so small and with only 3 I/O I programmed it into an ATTiny. Something similar for Leaf owners to disable the fake car noises so pedestrians wouldn't walk behind them when backing up. Played around with GPRS when that was a thing, and running a web server and some networking.

I am currently using a PI to create a small arcade machine that runs emulators for old school arcade games and consoles. Working much better than a low end atom and USB interface board, just using the GPIO.

I haven't had much use for the STM32, other than a little background into how to unlock and flash quad copters and fine tune them.
Hrm. I never knew about the ATTiny; I'll have to check it out.
My current program is about 8KB worth of code compiled -- from what I've read the ATTiny has about 6KB after the bootloader is installed. At <$3/ea, they might be really useful for projects.
-g
 

·
Registered
Joined
·
3,726 Posts
While the price and size of the attiny's is compelling, with the number of 2 and 3 dollar Arduino Micro/Mini/Nano and even cheap ones with the ATmega32U4 micro controller, it is tough to not just buy a number of those and use them.

Even the $1.20 Digispark Kickstarter boards are handy.





This is my I am bored, lets use a rotary encoder, a joystick from a PS4, servos, laser module, proximity sensor, and see what we can blow up box o stuff:
 

·
Registered
Joined
·
2,634 Posts
Discussion Starter #8
After my 7th Lemons race, I learned two key lessons:
1) Do NOT put the button on the F&F style race wheel because someone will tap it accidentally. Move it. Even better, don't make it a button -- make it a switch.
2) The communication protocol should be that the Pit crew only turns the "RETURN TO PIT" light on for the driver when THEY want the driver to pit and not as an acknowledgement of msg received. if the driver wants to pit, hes going to do it anyway. We had two inadvertant pit stops when the driver accidentally pushed the button, the pitcrew acknowledged the message, and the driver saw the light and thought that the pitcrew wanted him to come in. Whoops.
3) You need more info from the driver to the pit to decide what to do. For example, for fuel or driver change stops, you need someone on the pitcrew suited up in a driving suit. For repair stops, you need the "mechanics" of the team (we have two mechanics and one chef) suited in their repair coveralls and ready to work.

Total cost for a race: $11 (TING.COM service with 1K messages for 1 month. Not bad for a tmobile MVNO).

Ver 2 is now complete:
- Buttons replaced with a two switch panel mounted where the sun visor used to be. Even with a helmet + hans, a driver can quickly operate the system without looking up and he won't accidentally trigger it.
- Right switch = New Driver/More fuel
- Left switch = Repairs needed
- BOTH Switches = OMG! Some **** went wrong. Get ready for big repairs.
 

Attachments

1 - 8 of 8 Posts
Top