The date is: Fri Jan 29 17:03:37 MST 1999 This week has been eventful. We have decided on a chassies design and a basic plan of how to approch the project. We have decided to go with an ackerman sterring vechical, as well as use many components I allready have. We have agreed to meet on the weekends to work on the project. Secondly, We have decided how to decide the work load. We will split in to two teams and each work on a specific piece of hardware. Then all work together on the programing necessary for that piece of hardware. Then move on to the next piece. That is about it. The date is: Fri Feb 5 14:14:01 MST 1999 February 5, 1999 Dr. Wedeward, We are currently working on a design for the chassis and are planning to meet again tomorrow to complete the construction and attach the necessary components. We have decided to go with the acerman chassis design and to build it such that the rear end pivots in conjunction with the front wheels to aid in turning. Also, we have tested the speed controller we plan to use. It turns out that it requires a 5V pulse width modulated signal that goes from 7% (stop) to 9.5% (full throttle) at a frequency of 50Hz to operate properly. This range is adjustable on the speed controller and I plan to take it apart this weekend to see if I can widen the range a little to make the PWM a little easier. The output is a 3kHz PWM signal whose voltage is determined by the battery it is connected to. This controller can handle more than 30V DC at 300A, much more than we will ever use. Secondly we have also tested the steering servo to determine how to control it. It requires a similar signal to the speed controller, a 5V, 50Hz PWM signal that varies from 7% to 5% (counter clockwise) and 7% to 9% (clockwise). Being that these two signals are similar we can write one program and modify it slightly so it will work for the other. Since the duty cycle is so short on both signals we have decided to implement all off the PWM in Altera with 16bit resolution. This will not only allow better control but will free many of the Hc11's resources so that they may be used elsewhere if necessary. We are hoping to have a fully operational chassis by the end of the week. The date is: Fri Feb 12 15:02:06 MST 1999 Memo To: Dr. Wedeward From: Clint Ayler Sub: Weekly Report Date: February 12, 1999 This week we didn't finish the construction of the preliminary chassis because we were not able to get the correct gear for the motor. I am going to Albq. this weekend to remedy this. This week we tested the new 2012 sensor we received and are happy with the performance. We plan to mount four of them on the robot for wall following. Secondly, we plan to use one of the infrared sensors we initially received in the kit as a forward wall detector. Depending on how close it will detect a wall, we may not modify it, so that it remains a simple digital detector. As for the second one, we would like to use it as a white line detector. Rather than connecting the four 2012's to the A/D converter on the HC11 we plan to connect them to external 8bit serial A/D's since we would loose the use of four pins on the A/D port of the HC11. We are trying to keep from using the peripheral ports B&C so that he HC11 will not have to share the memory expansion ports, allowing the HC11 to run faster. Thus far we have accomplished this by: 1. using the external A/D converters 2. going with Acerman chassis so we only have to use one motor so there are less control lines 3. implementing the motor and steering control in Altera requiring a minimum of three control lines if we run the timing of the Altera chip off of the E clock on the HC11 This saves a minimum of five lines to use for other things. Also, we are going to connect the motor speed output to a decade divider so that we may use the HC11's pulse accumulator to measure motor speed so that we don't have to convert back and forth between digital and analog signals. This weekend we are meeting to prepare for the design review and hopefully to get a mechanically functioning chassis working. The date is: Fri Feb 19 17:02:10 MST 1999 Memo To: Dr. Wedeward From: Clint Ayler Date: Feburary, 19 1999 This week there is not much to report because preperation for the design presentation took much of the time. On Monday I am going to contact Analog Devices to see if we can get some sample A/D converters to interface the sensors. This weekend we will meet and start working on both the Altera and Hc11 code for motor control. This should not be too difficult sice our design is simple. Also, we expect to have a fully functional chassis, Jon is working on the servo placement while the rest of us start coding. The date is: Fri Feb 26 15:10:19 MST 1999 Memo To: Dr. Wedeward From: Clint Ayler Subject: Weekly Report Date: February 26, 1999 Dr. Wedeward, This week we began focusing on the programming of the Hc11 and the Altera chip. Before now we thought that the Hc11 only had 4 A/D converter inputs, but after talking to George we found out that the inputs are multiplexed so we can access the upper and lower 4pins of the port. Thus eliminating the need for the external A/D's. So knowing this we began to interface the 2D12 sensors to the A/D converter on the Hc11. We are also working on the necessary code to properly read this information. Mainly Nate and Greg have been working on this part of the project, I have been working on the Altera PWM code to run the servos, motor controller, and the motor direction relay. The 50Hz PWM signal works well in this application. Using the E-clock off the Hc11 divided by 10, the math works out that 50Hz is 4000 pulses giving us nearly an 8-bit resolution (200 pulses) for our motorcontrollers and servos duty cycle range of 4.5% to 9.5%. So far we have not had to use the PIA ports or the data\address lines, and in the event we need more port space this program can be easily converted from using a parallel port to memory mapping making efficient use of the Hc11 data lines. I had Dr. Rison look at it and he seems to think it will work, but I still have to compile and simulate it. Luckily this program fits on one 7664 Altera chip with room to spare. This weekend I plan to write the Hc11 portion of this code. Next week I will begin building the relay interface circuitry. We ordered the Altera chip today so by the time we receive the chip and the 2D12's, we may have much of the interfacing completed. The date is: Fri Mar 5 17:00:17 MST 1999 Memo To: Dr. Wedeward From: Clint Ayler Subject: Weekly report Date: March 5, 1999 This week things are moving out pretty well. Yesterday we received our altera chip, so we are going to be able to test our PWM program, which not only controls the motor but the steering and fire sensing servos. I have to write a temporary HC11 program to test the different functions but other than that most everything is ready to go. Also Greg and I worked out a nice flow diagram so that we can start the main programming in an organized fashion. We are attempting to try and keep the program as modular as possible. By this I mean have some parts of the main program (motor speed, steering, and sensor updates) triggered by interrupts and call to more generic subprograms to keep the program as efficient as possible. Greg and I plan to work on the PWM and possibly other programming this weekend. The main chassis is almost complete. By Tuesday we believe we will have a fully functional chassis. So, as soon as our sensors arrive, we will be able to start to interface the sensors and possibly have the robot wall following over spring break. The date is: Fri Mar 12 12:28:20 MST 1999 Memo To: Dr. Wedeward From: Clint Ayler Subject: Weekly Report Date: March 12, 1999 This week was especially productive. I got the PWM program working for the motor control, including the interface to the H-bridge, as well as the steering servo and the fire servo. This was quite the task since the program just fit in the Altera chip. In order to get it to fit I had to reduce the motor speed resolution to 6 bit (64 positions is more than adequate) and the steering servo to 7 bit which is also adequate. This should give you some clue as to just how full the chip really is. Also, I worked with Greg a little on the programming and we figured out a neat way to only use one interrupt in our program, this interrupt reads the motor speed, distance, sensors, and updates the Altera PWM program. We think it will work well. The basic chassis is finished all that is necessary is to connect the rear end to the steering servo and this is a simple task. Yesterday, Nate and I got the white line sensor tested and it works well. I am working on a circuit to amplify and buffer the output of the sensor to increase resolution and dependability. Jon has been doing a nice job on the sensors and is going to work on increasing the turning angle of the front wheels, adding necessary levels for sensors, electronics and batteries, and is going to start working on the fan. For the fan we are going to probably us one of my high-speed DC motors (45,000 Rpm unloaded) with a fan blade approximately 3 inches in diameter. This should be able to blow out the candle when it enters a room, if that were allowed. The control circuit is simple, a power MOSFET that is controlled by the HC11 will act as a switch. The FET will go into saturation turning the motor all the way on. This weekend Greg and I will meet to start physically interfacing the Altera and the wall sensors and hopefully have a wall follower before break is out. So, as you can see things are coming together well. The date is: Fri Apr 2 17:15:21 MST 1999 Dr. Wedeward There is not much to report this week, I have been fighting to get the PWm to run in this room. It works in the Analog lab well but not here. I have been trying to get this going all week with little luck, I think it has to do with, 60Hz Emf and since we are using a 50Hz pwm it is not easy to figgure out. Greg has been working on the programing end of things, but without dependable PWM we have not yet been able to test anything. Other than that, there is not much to say. The date is: Fri Apr 9 16:29:29 MDT 1999 Memo To: Dr. Wedeward From: Clint Ayler Subject: Weekly Report Date: April 9, 1999 This week has been OK. We managed to determine the source of the glitching in the PWM circuitry. It turns out that there were two problems. The first was that the H-bridge triggers at about 4.8 volts and the power supply was outputting at this level making the controller very susceptible to noise. The second factor was the noise on the circuit. We had originally thought the noise was a direct result of the E clock signal coming from the Hc11 so we buffered it with an OR gate but this turned out to work against us. The OR gate was cleaning up the signal at the expense of the rest of the circuit. When it was removed we found that the noise level dropped by about 50%. One thing that made this so hard to find was that we did most of our testing in the analog lab using the coaxial test leads which as it turns out act as a large capacitor. So the circuit worked well in there due to the added capacitance of the test leads. Once it was discovered that the leads where allowing the circuit to work it was realized (later) that a larger capacitance between the PWM out to the H-bridge and Ground was necessary. So, to fix the glitch Dr. Bond recommended a larger capacitor on the output and to increase the voltage level to the H-bridge to be sure that noise didn't trip up the circuit. Now that our PWM circuitry is functioning properly I have been working with Greg on programming or more like debugging his code. It seems that I am the only on who has any circuit board design experience so I also have been helping Jon & Nate design and build permenent circuit boards in P-Spice and/or Protel which I am now down loading off of the net. Greg and I have open loop control via HC11 now and are working on the closed loop version. Maybe we will see some wall following action this weekend. Who knows? Well that is about it for this week. The date is: Fri Apr 23 15:56:52 MDT 1999 MEMO To: Dr. Wedeward From: Clint Ayler Subject: Weekly Report Date: April 23, 1999 This week over all has been a little slow but today it ended well. For the last week I have been placing the components on the PC boards and starting the process of connecting and testing them. I have been a little nervous since I am the one who did all of the hardware design, including the PC board layout. Earlier today my worries were put to rest after testing because ALL OF THE HARDWARE WORKS!!!!!!! One of the questions I had about the boards was the noise level and, as expected, the noise level on the digital voltage source decreased and the noise on the analog lines is low which is a relief since our sensors are susceptible to noise. We have program ready to wall follow, so as soon as Greg and I can get together we are going to start testing all of the programs he has been writing. Also this week we are going to build a more permanent structure to mount the boards and neater wiring harnesses which will make things easier to work with. We still have to mount the fire and white line sensors on the robot, but they have been built, tested, and we know where we want to put them so it is just a matter of doing it. I think we will be ready for the competition on Saturday if our luck holds out. The date is: Fri Apr 30 15:55:22 MDT 1999 MEMO To: Dr. Wedeward From: Clint Ayler Subject: Weekly Report Date: April 23, 1999 resubmitted This week over all has been a little slow but today it ended well. For the last week I have been placing the components on the PC boards and starting the process of connecting and testing them. I have been a little nervous since I am the one who did all of the hardware design, including the PC board layout. Earlier today my worries were put to rest after testing because ALL OF THE HARDWARE WORKS!!!!!!! One of the questions I had about the boards was the noise level and, as expected, the noise level on the digital voltage source decreased and the noise on the analog lines is low which is a relief since our sensors are susceptible to noise. We have program ready to wall follow, so as soon as Greg and I can get together we are going to start testing all of the programs he has been writing. Also this week we are going to build a more permanent structure to mount the boards and neater wiring harnesses which will make things easier to work with. We still have to mount the fire and white line sensors on the robot, but they have been built, tested, and we know where we want to put them so it is just a matter of doing it. I think we will be ready for the competition on Saturday if our luck holds out. April 30, 1999 As it turns out things didn't fall together as well as we would have liked it to. We had some other trouble with either an IC or something having to do with a PC board. We have etched it but that is about all. Much of our focus week has been on the presentation tomorrow, but there is still hope that the robot will by the end of next week. The date is: Fri May 7 16:32:39 MDT 1999 Memo To: Dr. Wedeward From: Clint Ayler Subject: Last Weekly Report Date: May 7, 1999 This week has been a good week for us; we have the robot wall following dependably. All of the PC boards are built and have found permanent, neat mounting locations on the robot. I have spent a majority of this week debugging the hardware, especially the sensors. We have had a lot of trouble interfacing the sensors to the HC11. I think it has been the HC11 causing the problems. When we connect the sensor buffer to the HC11 it draws a lot of current and when we don't buffer the 2D12's there is so much nose on the line that the signal is nearly useless. We have been trying to use a 2.5V reference for our sensors but once again when we do so the power consumption jumps and if we try to specify a low reference the Hc11 pulls it nearly to ground. So in future designs I think it is a good thing that the HC11 probably won't be used. The other main thing I have been working on is a permanent solution for the PC boards. This ended up working so well that we went from 4 layers to two! Also, this got rid of a lot of the wire mess. Greg and I are going to work on Glitchy for a little while this weekend so as to try and get some of the oscillation out of the wall following.