[Riley Porter] posted a picture of his EvalBot USB power hack. In the photo above we’ve put a box around D6 and D7. The development board ships with a 0 Ohm resistor in the D7 location, patching in power from the USB-B connector labeled USB DEVICE. He found that by moving that resistor to D6 he can power the board from the USB-B connector labeled ICDI.
That connector is the In-Circuit Debug Interface. TI sent us an EvalBot bundle so we pulled it out and tried it ourselves. If you plug in the ICDI it doesn’t power the board, and no USB devices register. Shorting the D6 pads changes this and the following USB device registers:
Bus 002 Device 062: ID 0403:bcd9 Future Technology Devices International, Ltd Stellaris Evaluation Board
So it looks like you need to have two USB connections or be using batteries in order to program the board via USB. The uC/OS-III hardcover book that ships with the EvalBot bundle includes board schematics. We took a look and were surprised to see that they show diodes installed on both pads. Rev A of the online schematics have been corrected, showing an omitted diode on D6 and the 0 Ohm resistor on D7. Images of both schematics are included after the break.
It would have been nice to see a selector switch installed here to give you a little more flexibility when prototyping.
Schematic from uC/OS-III page 700:
Stellaris Robotic Evaluation Board Schematics Rev A (available here):
24 thoughts on “Moving A Resistor For EvalBot Power When Programming”
We started discussing this at http://ti-evalbot.proboards.com/index.cgi?board=general&action=display&thread=25
To me, this was a HUGE oversight on TI’s part but it’s easily fixed.
I would highly recommend using a Schottky diode in D6 instead of a 0 ohm resistor. Using a diode protects your computer from over voltages that may be seen due to batteries, motors back feeding or another supply that might get plugged in. No need to accidentally blow up a USB port, a simple diode will cost less than a quarter and can save your motherboard.
mine gets shipped out on the 12th of next month and i might try this
yes i agree if you have one on hand or want to wait for one to ship
The power switcher onboard is pretty good at preventing overvoltages, the reason for the lack of diode is that the draw on the batteries during motor use would drop the voltage to the programmer enough to disconnect it. Also, the motors on startup can draw too much current from the USB supply, causing the entire board to power off.
is a 0 ohm resistor even a resistor because it does no resisting O_o
@biozz: It’s one of those Resistance is Futile things…
Someone have a link to the recommended software stack for developing on the platform? I managed to get one, but haven’t had any research time.
What makes it a zero ohm resistor rather than a zero farrad capacitor or a zero legged, bodied, headed armed and tailed kangaroo?
Is a zero ohm resistor stupid talk for a non resistive link between components?
A zero Farad capacitor is pretty much what you’ve got when there’s nothing installed in location D6.
So, you’re replacing a zero Farad capacitor with a zero Ohm resistor :-)
It’s interesting that DigiKey carries six different size 1206 1/4 Watt zero Ohm resistors IN STOCK!
I’ve got a bag of infinite ohm resistors around here somewhere…
TI’s got a couple of those 0 ohm resistors around the crystal pad on the Launchpad as well. It threw me for a loop the other day when I had to solder some load caps onto the board for a HF crystal I was installing. I guess they must use them as placeholders in case they want to put something else there in a future revision.
Also, I wish TI would have shipped the EvalBot in the order that they were purchased. I have a lower order number than the people who are getting theirs. I also have a ship date of 11/4 for a couple more Launchpads, but they’re still stuck in “Processing.” It’s great that the EvalBot and Launchpad are so cheap, but frustrations with the estore will probably leave me a bit unsatisfied in the end.
Zero Ohm resistors are extremely common and very handy.
They are commonly used to activate/deactivate portions of pcb at population time. It serves the exact same purpose as a jumper, but is much smaller.
There are other reasons, occasionally it can also be used to adjust R values in case mods need to be made.
Hilarious, yes. Practical yes.
They’re also used to exclude unpopulated parts of a board (if not soldered in) and as small fuses in signal lines to save hardware from computer faults (they hopefully “pop” before your computer does).
Sorry, I meant connection faults.
I think these have been discussed here before…but the zero ohm resistor is much easier to use in automated assembly systems than a jumper wire.
Yep I see 0-Ohm resistors all the time where I work.
-way easier than a jumper at SMD scales.
“The uC/OS-III hardcover book that ships with the EvalBot bundle includes board schematics”…
AFAIK the book is not in the 25$ bundle that you could order, guess you guys got an extra :).
Is the schematic somewhere online (could not locate it that easy)?
ok, stupid. did not say the link.. need coffee
I finally got some code running and debugging on the Evalbot last night. There are really 2 ways to go for a toolchain.. (high level i mean)
1. Buy one. This way is Super easy. In fact. I used Code Sorcery Stellaris EABI. What you need to do is download the trial register and get a 30 day setup. However, its not done there. You also need to get the StellarisWare from TI. (its in a the form of an exe.. greeeat. So run on windows copy to linux, osx)
2. Build one. This way is not easy. I am working on cross compiler now. Its been pretty hard so far. I have NOT got this option working yet.
Long story short I plan on posting a blog with links for setting up the Evalbot for development. All of us who got the $25 special didnt get a book. Oh and to make things worse. TI officially does not support the evalbot as a “stand alone” dev platform. Not until q1 2011. Meaning, you have to get the book and run their $$$ RTOS if you want to dev for it now.
I have been googling for a few days and I have not seen anyone else get code working on this evalboard. I will post a write up to my blog once its all setup and working however. If you want to start writing code now I would suggest just get the trial CS and the stellaris header files (stellarisware) and blinky blink!
BTW, I ran CS on Linux (ubuntu 10.10) which is pretty cool since most Arm dev has to be done on windows… Oh and the personal CS dev environment is $99. So time * effort > $99?
blog.synthetos.com <— Check back here. I will post it when its done.
Everyone who is having trouble programming the board:
I’m working on putting together test code for most of the peripherals (got the motors, oled, UARTs, and USB mouse working so far), I’ll post them on that forum when its ready.
Programming in Windows is relatively easy, once I figure out the bootloader, Linux development should be trivial.
LOL, I just looked at your source code.. Its almost EXACTLY how I got blinky working :) I used PF4 and PF5 too. Cool stuff.
Have you got GNU working ? Or just using IAR?
they may have mistaken, i was one of the first hundred orders, i got the book and all for 25$ and it shipped within 2 weeks. dunno, maybe they ran out of the standard models and shipped what ever they had. :/
Mine arrived a few days ago with D6 populated with the 0 ohm resistor rather than D7. It appears that TI is taking note of user feedback.
Please be kind and respectful to help make the comments section excellent. (Comment Policy)