This is a decent board, but I think I will stick with the one I found if I do a project with a mini-ITX board. It’s a bit more expensive, but I like the feature set with 8 USB 2.0 ports and its fast enough to build software natively. I have been wondering about access to the SMB bus also, and none of the descriptions of the mini-ITX boards seem to indicate access one way or the other. Maybe the manufacturer website has more info on this.
How the heck did I miss this post! Excellent work on the tracked vehicle Eddie! 8) If you are using 4 of the GHM-04’s you should be ok. You can also PWM them at 12vdc, run them at 60% max most of the time and take them up from there for short bursts of power when needed. Just take care not to run over 60% for too long.
hey thanks for the compliments and advice. I actually only have two of the motors at the moment and will see what happens. I have a couple of options thought out if they don’t have enough oomph though so I should be covered.
I have actually been picking up stuff slowly over the last few weeks, including more parts for its arm, sensors, and the mini-itx board and stuff to support it. waiting on a couple packages this week and next but I did get some of the sensor stuff done last weekend.
the plan as it is… I’m working on using a PIC and a USB-to-serial eval board (left over from a project) to interface the encoders, sharp ir sensors, and the ping to the mini-itx. I’ll use the mini-itx serial port to talk to the ssc-32 for controlling the servos and motor controller. I’m just going to put an inexpensive usb webcam on the arm for a camera for the moment. I don’t really have any other kind of end effector planned so I’ll just use it as an inspection ROV for now. Trying to balance powering it all is an interesting problem that I have been trying to work out. I will probably just run with several smaller batteries to start and then do something more sophisticated when I rev-up the chassis later. Right now I’ve got 9.6V packs for the motors, the 6V pack from the Brat for the servos, and a 10.4V DR202 Li-ion pack and charger for the mini-itx. Lotta weight.
Anway, I will likely post a few updated pictures next week when I’ve got more things assembled. I am doing too much firmware/software work on the PC at work lately to feel like working on the control application at home, and mother nature doesn’t look like it is going to co-operate with weather this weekend for flying so I’ll just have to find something else mechanical to do.
Is possible to design a power supply for it? I’m thinking of a circuit board custom designed to use one power source, and like a PC power supply, have several supply cables coming off of it going to the various peripherals.
Yes and I might at some point. the most efficient approach is to put it all on one high voltage battery and switch down to the different voltages and currents I need. to be honest it is too much like stuff I do at work for me to want to take it home and do it there too. you saw how I burned out on the wiport / fpga board for the lantronix competition. so for now I’m just going to do it the simple way. I did order a different picoPSU for the miniITX that can take from 6-26VDC so I’ll try running that from the same pack as the motors and see how it goes, maybe be able to eliminate the Li-ion pack and just upgrade the main motor pack to a 10 or 12 cell sub-C pack or something. right now I am researching how to make windows xp have the smallest footprint possible on disk so I can run the whole thing off of a 2gb flash card and have no hard disk to worry about.
No hard drive? Sounds very cool. I have heard there may some day be internal flash drives to replace internal PC hard drives.
Have you seen the Nano ITX boards? they are smaller than the normal ITX boards but the processors that you can use are limited.
In this image, it uses a VIA 1ghz processor and the board is 4"x4" in size. You can tell how small it is in comparrison to the size of the video connector. I’m not sure if windows xp will work with the VIA processors though.
unfortunately nano-itx boards are out of the $$$ range I wanted to be in. I went with a Via EPIA ME6000 600MHz fanless which was about $140 and I already had ram and an oem xp pro package on hand. I bought one of the CF200 compact flash adapters that I am going to modify to just plug onto an IDE header (so no cable) with a 2GB cf card as a disk module. As an IDE device the cf cards are considered non-removable and bootable. Really from the motherboard standpoint they look, smell, and run just like a hard disk, although a slow one. The Via processors run windows XP btw, as well as most other x86 based OS.
I would like to use the nano ITX like the one I have in the previous post. It’s expensive but it’s nice and small. I would like to use it for real-time onboard video processing, and have it comunicate to the Atom for motor control.
Eddie, you going to use Robo Realm for the web cam?
This is really interesting. Linux would really rock on something like this with 512 Meg or 1 Gig of RAM.
The size of this board is great! However, there is a premium for the small size - higher cost for the board. I think Windows XP would not be a good idea even if it could be run on it. Windows XP is just too much of a resource hog.
This project really isn’t planned to be autonomous yet so I have not gone very far into looking at how I would do that. Seems like there is a fair amount of renewed chatter on the board about Robo Realm lately though so I have got to figure I will give it a look through when the time comes. I’ve certainly eaten the speech and voice recognition curves at work so those parts will be a no brainer when the time comes to make the jump from ROV to autonomous.
Initially though I’m planning on using a biological image processing and control system.
I meant to say SSC-32 and not the atom. You said that windows xp will be to much of a hog for the VIA 1ghz processor? Perhaps I should stick with the mini itx?
I’d recommend going with the first configuration illustrated, and attach the liquid-filled heat-dissipation unit to the heat sink, in order to increase the thermal mass. The second one illustrated is shinier, but that’s about it.