Older blog entries for earlwb (starting at number 8)

Tomorrow, Thursday 20th of March, I am driving up to Illinois from the Dallas Area to pick up a steel Sumo ring from www.Lynxmotion.com on behalf of the DPRG club. This ought to be very interesting as I get to see the top secret Lynxmotion facilities, where they make all those great robots. The new Japanese Sumo rules have us using a 1/8" thick ring of steel as the sumo ring for the matches. This allows the use of magnetics and vaccum assist to improve the traction of the robots. Of course you can't just stick a robot in the ring, it still has to move around and attempt to push the other robot out of the ring. This is for the 1 or 3kg 22cmx22cm class of robots. Also www.DPRG.org supports autonomous robotics thus radio control isn't used.

My nano-sumo bot project is still going. I've had a number of attempts at getting under the 1 cubic inch size barrier. And it looks a lot closer now. I just discovered some 402 sized SMD LED's from LUMEX. So I did a Atmega8 (MLF formfactor) PCB using 402 sized parts on it as well. It eeks out a little bit more space on the PCB. I still have a 603 sized component PCB just in case.

Now if I can only find some small rechargable Lithium-ion battery cells small enough to work. using non-rechargable cells sort of sucks as they are very expensive. I see tantalizing pictures, but no one selling them yet. Or at least selling in small quantities.

I can't beleive it is so hard to get two identical speed controls.

I wanted to get two ESC's for a firefighting robot I am trying to build. So I ordered two ESC's from a nice supplier. They both arrived, except one was DOA. So I sent that one back to the manufacturer for warranty repair/replacement (that's what the instructions say to do). So they send me a even better one that the one I sent them (heck what do you do?, they did really good on the warranty DOA). Big difference too, a 140 amp (.012 ohm) ESC replaced with a 162 amp ESC (.004 ohm). Doesn't exactly allow for a robot to track straight. Anyway, I order another 140 amp ESC from the supplier and they accidentally send me the wrong one a 65amp one. I ordered two of a different model ESC from a different dealer, and I wind up with them accidentally only shipping one instead of two.

I've called them all and they are all correcting the orders and sending the stuff out. So everyone is bending over backwards to correct the problems. But it's just the bad karma of it all. Jeeshhh. Only four weeks trying to get two matching ESC's. Besides all the time lost waiting. I have to sort of squish everything into the robot and build up on top of that in the robot, so there isn't a lot of room to go trying to disassemble everything.

I should have known better and just built the stupid things myself. But I haven't been able to get the 140-160 amp motor controllers that small yet. Mine are too big.

But before anyone says anything. A set of motors needs maybe 20 amps max to work. But a 140 amp motor controller doesn't need neat sinks as the transistors have such low on resistance (PWMing them) that they don't even get warm. Thus you can cram them into a small, tight, low to little air circulation place, and they don't get hot.

Definitely going to be hard to build a firefighting robot at this rate. Maybe for next year's firefighting contest, I'll be ready. Don't know about this year.

I got some Atmega8 MLF's and some Atmega128 MLF's in last Friday. The Atmega8's are really small in the MLF package format. Basically, it sort of looks like the old PLCC packages, except the little pads are about .020" wide and about .015" apart. Plus they have a center square pad that serves as a solderdown and holddown point. The pads are also arranged around the bottom edge or lip of the chip case, and don't extend up the side. The pads are also plated with a gold like substance and not tinned with solder like you would typically expect.

Now all I have to do is come up with a PCB that has the Atmega8 and two 3v motor hbridges on one PCB that is less than 1 inch square, and still have a place to put in a IR object detector or two and a line detector as well. What would be really neat is to get a small LM2621 or MAX1621 DC- DC 1.2v tp 3.3 ot 5v step up converter to work on that same size as well. The MAX DC-DC converter can handle up to 1 amp max, so I could run the two tiny swiss gear motors off the converter via the motor controllers too maybe.

I am thinking about whether backing up is really important or not. The nano-sumo bot can go forward and turn easily enough, one or the other motor off. Or maybe only use one hbridge for reversing one motor and run the other forward only. In this case, if the robot detects a line, it can do a reverse turn to the left and turnaround to the right and continue on it's way. Forward only saves a lot of space, only one small SMD transistor per motor, the other nethod uses one transistor and one hbridge.

Atmel Atmega8 and Atmega128 MLF package MCU's from digikey or Pioneer are finally available in some small quantities. A Atmega8 MLF chip is about 5mm square, really small. I ordered some, I hope to shrink my nano-sumo robot PCB's down to one PCB with the MCU, dual hbridge motor controllers, and really hopefully maybe even the 1.2v to 5v DC-DC converter all onto one under 1 inch PCB. Currently, my nano-sumo PCB has a MCU board, two motor controller boards, and a DC-DC converter board. I actually use a LM2621 DC-DC converter that puts out about 1 amp max, so I use it to drive all the electronics including the two tiny motors as well. Then I can use one nicad or nmh battery cell just like the little Microsizer RC cars.

Bad luck building up my Can Can DPRG contest robot. One of the ESC's appears to program but not drive the motors. One ESC works and one doesn't. So I have to send it off to get fixed and wait forever for them to send it back. I hate it when that happens. New right out of the box. I guess I should just stick to building my own motor controllers, at least I can fix them myself.

Last week I was successful at building a second Gyro/Accelerometer combo board. And it works as well. It was a little scary at first, as I accidentally caused a short on the gyro chip and had to remove the chip, clean up the chip and the PCB and resolder it back down. In this case I used a ADXRS300 chip with 300 gegress per second performance versus the earlier PCB using a 150 degree per second chip. Those hot air SMD rework stations are worth their weight in gold.

It was a busy week, I got most of my new balancing robot built up. I made up a PCB co-processor board using a Atmel ATiny26 MCU to use with the gyro combo board, and I also got the Atmel ATMega128 MCP main PCB built up and tested as well. Now comes the fun part, wiring in all the stuff and getting it all programmed. I am using two of the new Hbridge motor controllers from NewMicros and I'll be running the motors using 16 NMH AA cells for about 19.2v. The plan is to have the ATiny26 read the Gyro and accelerometer chips and keep the robot in balance. The Atiny26 will also monitor the serial port for commands from the main MCU. The main MCU will tell it to go forward, reverse, slow, faster or turn left or right as needed. I've mounted a couple of encoder discs to the wheels with encoders so I can get distance traveled information or RPM indications as well. Lots of little things to do now.

Sorry I was so excited that I forgot to mention that this is the new Analog Devices ADXR150 Gyroscope IC and ADXL202 Accelerometer IC.

It's alive, It's Alive!!

I was successful at mounting my Gyro chip on a PCB. Plus the circuit works too. Really, really neat. I had a PCB layout done for me by ExpressPCB.Com, I then applied some flux onto the PCB and carefully placed the chip. Then I used a hot air rework station to carefully heat up the stuff and solder it down. Of course you have to carefully hold the chip in place to prevent it moving as the hot air flows under and around it. It took about three seconds to heat it up real nice. You could tell as the solder started flowing as the chip would drop down ever so slightly.

I am planning a project submission for the DPRG website detailing my PCB layout for a combo low cost Gyro/Accelerometer system on one PCB.

Of course low cost is relative. Right now the Gyro chip costs $39.90 in single units, and the Accelerometer costs about $29.90 in single units as well. So we have about $20.00 for the PCB (piggy back with other layouts), $70.00 for the two chips, plus maybe about $10.00 for other assorted parts like capcitors, inductor, regulators, connectors, etc. So it looks to be about $90.00 or so, to build one up. But then a while back another company wanted something like $4500 for a gyro chip setup. So in that case it is low cost.

But it is really cool, works nice, the gyro is real fast and very sensitive to small tilt changes. The Accelermeter allows you to compensate for the creeping gyro errors that occur in things like a two wheeled balancing robot.

I love it when things actually work. It is so hard to debug and fix things in the unknown never been there before area.

Share this page