Spaming the comments on this blog has gotten out of control. I had 83 in one day. In self defense I implemented a simple Captcha system. So far i have had zero spam comments and it would appear zero failures for legitimate people from posting comments. Sorry for the inconvenience.
If anyone needs it, I have 83 internet companies that would love to sell you your favorite pharmaceutical.
Finally getting close to finishing the gyro calibration table. This was made from a re-purposed record player. I gutted the electronics, the tone arm, and even the motor. Replaced the motor with an electric starter motor for a nitro RC car.
I had problem with the speed control because the controller I first got was good for up to 1.5 amps. I had measured the motor resistance and 1.5 seemed enough. But when running the motor draws closer to 2.5 amps and the controller went into thermal shutdown. So I replaced the H-Bridge with a simple FET and a mechanical reversing switch. The gate of the FET went to the PWM output of the Arduino. Now I can control the speed for rather slow to ridiculously fast.
The next step was feedback on the speed. For this I painted on 96 black strips on the outside edge of the polished aluminum platter. I then used a reflective IR sensor to watch the stripes go by. And this is where I hit my second snag.
I could not get a single interrupt per stripe. I would get as many as twenty or thirty. After significant time investigating, I learned that the Arduino digital input pins have a Schmidt trigger with only 60mV of hysteresis. As the black stripe would swing into view, the voltage on the output would ramp up and not snap on. This allowed the noisy signal to be in the hysteresis band for enough time to trigger a gazillion interrupts. I was forced to condition the signal with a comparator wired for 1 volt of hysteresis before I could get good clean interrupts.
Next I needed to insure that the rotation speed calculations being done on the Arduino were correct. For this I turned to a trick my son taught me (one good reason to send your munchkins to college!). Place a contracting dot on the item that is spinning then take a picture of it with your SLR. Using ImageJ, measure the amount the dot was smeared. Now you can multiply the smear angle by the shutter speed and get the angular rate. My two values were very similar and I think I can now trust the Arduino calculations.
Progress is now on hold while I replace the master hard disk drive in my file server. SMART warned me it was going bad so out I went to by a replacement. Glad this happened now and not during the Taiwan flood when drives were horribly expensive.
Last week sometime, Sparkfun announced next year’s AVC competition. I am sooooooo far behind schedule it is not funny. By now I had hoped to have sensor data I could use for simulations. I had wanted an ARM processor up and running. I had wanted all kinds of things that just did not get done this summer.
Do last night I got the chance to go over to my testing ground (the local elementary school) to perform a GPS survey on some control dots I had painted on the pavement the day before. But Murphy was working overtime and when I arrived a gazillion parents and students where there for an open house. So I turned around and came home. I did use the time to survey my road, but that data looks poor since the road is heavily wooded.
I started to experiment with GPS positional error. I placed a Trimble Lassen IQ receiver in the shadow of my house. Most the the satellites vehicles (SV) in the south are masked. I logged position data all night and loaded the data into Excel. I then converted the lat/lon data to relative Northing and Easting based on a position dictated by the average lat and the average lon. The results were rather shocking. Continue reading
I was on a sailing vacation in Maine for the last two weeks. My type of sailing is small boat sailing that leaves you bruised and sore, no the Margarita sipping type. Need to get ready to teach a class in LEGO next weekend, but now that my kids are back in college, it will mean more time on projects.
Having to take another short break. Time to take a small vacations and to get the kids off to college. I had been working on a gyroscope calibration table but that will all have to be put aside for a few weeks.
After I got Vectronix to agree to a sponsorship of my AVC bid, I was compelled to get approval from my employer as we do business with Vectronix. After numerous exchanges, my companies policy prevents me from accepting the compass. Even though Vectronix would be lending me a unit and not gifting it to me and even if I was proving a service in return, it was considered to be over the $25 gift allowance. So I had to back out of the arrangement and will need to find another heading solution.
I ordered two HB100 Doppler radar modules from China. They arrived and I took them down to the lab for a quick test. First off, the boards look nice. Good solder masks and silkscreens. On the surface they look of good construction and materials. Continue reading
So the electronics deck on my RC car is hard mounted to the chassis. When the robot vibrates so do all the electronics including the IMU. This points tons of noise into all the channels of the IMU. I have talked to a few of my mechanical friends and they all have come back with various suggestions including floating the IMU in a bottle of glycerin. Like that would work on an RC car!
I think I have a few ideas on my own. First I think I need to stiffen the electronics deck. I could see it flexing like crazy under the vibration. Then I need to modify the hard mount system with perhaps a rubber band mount. And lastly I am going to need some kind of viscous damper. Here is where I am stuck. Perhaps I can find a VERY light duty air filled shock absorber.
So here is another job for the todo list.
Got all the channels on the IMU to work last night. I then started to perform some crude testing. First thing I did was to get a few minutes of data without any motion. This seems ok, although there is more noise than I would have liked.
Next I simply turned on the car motor. I was fully expecting to see the magnetometer channels go ballistic due to the magnetic field, but I was not expecting the accel and gyro channels to go nuts as well. But looking at the car, the sensor deck was visibly vibrating to the point it was blurry.
So now I think I need to balance the tires (will my mechanic do this for me?) and vibration isolate the sensor deck from the car. Need to think on this one.