Friday, December 25, 2015

More PID ramblings

First of all, Merry Christmas!

Since I cannot seem to turn my brain off on this topic I have been semi-immersed in for a number of weeks, I wanted to jot down a few thoughts on the topic of PID controllers.

My goals in using a PID controller with a mechanical fader that is driven by DC motor are simple, to wit:

  1. Instantaneous snapping to a new position once loaded into the controller.
  2. Positional accuracy and repeatability when the same position is loaded.
  3. Smooth movement of the fader control under PID control when set point changes are small and frequent.
  4. Tolerant of "stair-stepped" set point changes arriving at non-regular or varying intervals while still maintaining smooth movement.
So far, I have achieved the first two goals.  To date, I have been using a positional PID in an attempt to meet all these requirements.  A positional PID is one that takes a positional set point and a current position as input and produces the necessary output to drive the fader to its new position.  It takes care of acceleration and deceleration as necessary to arrive quickly at the new position.

However, when trying to achieve smooth movement that is quite slow, I find that a positional PID is insufficient for the task for the following reasons:
  1. A PID tuned to aggressively snap to a new position is a poor experience when trying to achieve slow but smooth changes to the position.  The fader tends to be very jerky and/or noisy.  The effect looks like a long-term crack user trying to be cool when the authorities arrive.  I have attempted to use multiple tuning settings where I reduce the gains of the P, I and D terms, but while it improves things, it is still unacceptable.
  2. I find that it takes approximately 30% of the motor speed range just to get the fader to move if it is already stopped.  Once it overcomes inertia and friction it will suddenly lunge forward and likely overshoot the desired set point if it is close at hand.
  3. Once the fader is in motion, it takes a lot less energy from the motor to keep it in motion, but there is still a fairly significant dead band in the motor speed range where the fader will stop again if the PID output drops too low.  Anytime the PID calculation must reverse direction in order to slow motion, it must necessarily pass through this dead band on both sides of zero.
What is needed for smooth motion out of a PID is to have the PID control the velocity of the motor, not just to move it to a new position.  I have attempted to simulate this using a positional PID by limiting the range of the output of the PID to something lower than the maximum speed of the motor when it is desired to be non-aggressive about positioning.  However, what I find to be true is that while it helps the overall effect, the results are not sufficiently predictable or repeatable from one fader to the next.  They all have their own friction profiles and while they are carefully manufactured, the motor movement transfer function from one fader to the next varies considerably.

So, to provide smooth movement for non-aggressive position changes, what is needed is a velocity PID that will control the speed of the motor according to a motion profile.  I will be implementing a trapezoidal motion profile where the motor will accelerate to a maximum speed, hold that speed and then decelerate to a stop at the desired position.

So, I think I need the following:
  1. ​Desired velocity - I will calculate desired velocity based off the current positional error.  When the desired position changes, we start moving in that direction according to a motion profile.
  2. Current velocity - I currently calculate this as the derivative term of positional PID.  The first derivative of position is velocity.
  3. A second PID to control motor velocity when being non-aggressive about positioning that will control motor speed according to a motion profile in a tight control loop.

We can visualize position, acceleration and motion profiles as follows:


To determine the desired velocity, I will calculate the positional PID error.  This is simply the difference between the current position and the desired position.  The further I have to go, the faster I will go up to a maximum limit.  As the fader moves, the error changes (approaches zero) as the fader approaches its target.  The fader loop will calculate the desired velocity as the current position changes towards the target.  As friction changes, voltage changes and other external disturbances affect the actual velocity, the PID will calculate the necessary correction in velocity to stay on the motion profile.

More to come...

Sunday, December 20, 2015

Christmas Math

Just for fun...



I wish you all the best in this holiday season.

Saturday, November 28, 2015

PID Controllers

Wow...  Where to begin.  I have had the pleasure of trying to get my head around implementing a fairly simple controller for a linear fader control that is also motor controlled.

This is an Alps fader which is basically a linear 10K potentiometer that has a small DC motor mounted on the back.  It has a 100mm travel length and the fader control is touch sensitive.  Mouser and other suppliers have them, though I cannot recommend hobbyists buying them through these channels as they are quite expensive.  I have seen them on eBay and other auction sites at significant savings.


The motor is mounted along the axis of the linear fader making a very low profile solution.  The motor drives a toothed belt to move the fader to any position desired remotely.  The motor is a simple DC motor that can be driven by up to 10 VDC.


So, one could envision simply putting a voltage across the pot and connecting the wiper to an Arduino analogue input.  Moving the pot would report a digital number from 0-1023 at any given instant in time representing the voltage across the pot at the wiper and therefore, the relative position of the fader control.

 However, if we wanted to simply tell the fader to position itself to any of its 1024 positions it is capable of, we would basically need to turn the motor on and drive it in the correct direction while reading the pot position until it attains the correct value.  This would be an appropriate application for a PID controller.

To control a DC motor, it is most convenient to use an H-Bridge circuit.  As can be seen by the circuit diagram below, turning on Q1 and Q4 will drive the motor in one direction whilst turning on Q3 and Q2 will drive the motor in the opposite direction.  As an addition feature turning on Q1 and Q2 or Q3 and Q4 will short out the power supply and allow for stress testing your circuit protection implementation...


In reality, there are nice single chip H-Bridge solutions that not only allow for controlling the motor direction, but also allow for PWM control of motor speed.  Most suppliers of robotics components will have typical examples of these controllers as well as places like Adafruit and Sparkfun.

In my next posting, I will get into the details of the hookup and programming of the PID controller for this fader.  More to come...

Saturday, October 17, 2015

Fun with PIDs

PID?  What the heck is a PID?

According to Wikipedia, PID is a Proportional-Integral-Derivative controller (PID controller) which is any kind of control loop feedback mechanism.  These controllers find common usage in industrial control systems.

What a PID controller does is continuously calculate an error value as the difference between the measured value of something that is changing and the desired value.

Perhaps a simple example is in order.  Let's say you want to fill a pan of water at a specific temperature from the tap.  You have two sources of water, cold and hot.  You will need to control the mix of hot and cold water to obtain the desired temperature which you are going to measure with your finger.  So you start with a guess of the hot and cold water settings, measure the result and adjust the hot and cold flow until satisfied.  Making a change that is too big may result in overshooting or undershooting the desired temperature.

Lets call the water temperature sensed by your finger the "process variable" or PV.  The temperature that is desired, let's call the "set point" or SP.  The amount of change of the water flow mix of hot and cold, let's call the "control variable" or CV.  And lastly, the difference between the measured temperature and the desired temperature is the "error" or e.

The most obvious way to change the water flow is in proportion to the current error.  The bigger the error, the bigger the change.  A slightly more complex methodology might want consider the rate of change of the error adding more water flow control depending on how fast the error is approaching zero (a derivative action).  Another refinement might be to consider some historical accumulated error information to detect whether the temperature is settling out too low or too high and make the appropriate corrections (an integral action).  Another way to perform this integral action would be change the current water tap position in steps proportional the current error.

When making changes that result in overshoot or undershoot of the desired set point, continuing to do so will result in oscillations around the desired set point that either grow or decay with time.  If the oscillations decay, the system is stable.  If they grow, the system is unstable.  If they remain at a constant amplitude, the system is marginally stable.  The desired goal is a gradual convergence to the desired set point, so the controller may try to dampen future oscillations by tempering its adjustments by a process called reducing the loop gain.

When the controller starts from a stable state with no error (PV = SP), any changes made by the controller will be in response to changes in other inputs to the process that affect it.  These changes are known as disturbances.  An example of a disturbance in our simple system described above would be a change in the tap water temperature caused by an external event such as the water heater deciding to turn on its heating element.

In theory, a PID controller can be used to control any process which has a measurable output (PV), a known ideal for that output (SP) and an input to the process (CV) that will affect the PV.  So, as you can see PID controllers find uses in any problem domain that needs to regulate temperature, force, speed, pressure, flow rate, weight, position and just about any other variable for which a measurement exists.

So, now just a little bit about the theory of PID controllers.  The name as we have seen above comes from its three terms (Proportional, Integral and Derivative) which are summed to calculate the output of the controller.

The clasasic formula for the algorithm (from Wikipedia) is:


where:
   Kp - Proportional gain.  These first three are algorithm tuning parameters
   Ki  - Integral gain
   Kd - Derivative gain

   e   - Error = SP - PV

   t   - Time or instantaneous time (the present)
   T  - Variable of integration (takes values from time 0 to present time t)

What is desired is a controller that will smoothly make adjustments to the desired set point with minimal over-shoot and under-shoot.  Directly implementing this formula in code would lead to some simple solutions that would most likely suffer from a number of short-comings such as the following:

  • Sample Time - The formula requires recalculation at a regular interval since both integration and derivatives are a function of time.
  • Derivative Kick - Any time the set point (SP) is changed, the error is changed (SP-PV) and the derivative of this change is infinity.  In practice however, since the change in time is never zero, it ends up being a very big number which when fed into the calculation results is an undesirable spike in the output.
  • On the fly Tuning Changes - Changes in the three tuning parameters affect the integration of the error value over time.  Any change in Ki will be multiplied by the entire error sum that has been accumulating when we really only want it to affect the result going forward.
  • Reset Wind-up - Most controllers have some limit on operational ranges.  For example a water valve can only be set in the range of completely closed to completely open.  If the PID controller does not know about these ranges, it may calculate an output value that is out of range.  Over time when it tries to continue to add more water flow beyond maximum, only to have it clamped by the physical size of the pipe, the algorithm will continue to ask for more and more beyond the limit.  The result is that the output gets "wound up" way beyond the maximum limit.  Where the problem reveals itself is when the set point (SP) is dropped, the algorithm needs to wind back down again to the maximum before it will even affect the output.  This results in what looks like a lag in response of the controller to the new set point.
  • Switching the PID Controller on/off - This occurs when you validly decide that regardless of what the PID controller is doing, you want to override its decision for a period of time.  Now, when you stop overriding it's decision, you get a sudden, huge change in the output.  The controller keeps trying to adjust the output to get the desired result, but it doesn't see any change, so it adjusts the output a little more and so on.  It would be like externally overriding a volume control of a stereo system while trying to adjust the volume internally.  No change internally has any effect.  Then you switch off the external override and suddenly you have a huge volume change.
  • Initialization - While it is useful to be able to turn off the PID and set your own override, when you turn it back on, the PID jumps back to the last output value it had set resulting in another spike in the output.  These transitions need to be seamless.
  • Changing Direction - The PID algorithm may be used to drive a system that is either "direct acting" (an increase in the output causes an increase in the input) or "reverse acting" (an increase in the output causes a decrease in the input).  A refrigeration system for example is a reverse acting system as an increase in the cooling results in a decrease in temperature.  Changing the sign of Kp, Ki and Kd allows proper control of a reverse acting systems.
Rather than detail the solutions for all of these potential issues, let me refer you to a most excellent blog by Brett Beauregard detailing the improvements one-by-one starting with the most basic code example.  It is a great read and not one I could not improve upon.

There is a great PID implementation for the Arduino also available at the Arduino Playground which will be the subject of my next posting.

Sunday, October 11, 2015

Postage Stamp Micro-controller

I have updated my ATMega328 controller board with a new layout to fix the previously posted deficits and have sent the gerbers off to OshPark for production.  Hopefully I will have the set back in a week or so.  After validation, I will make the design public at OshPark so anyone interested can purchase the boards.  This is a one inch square PCB with an ATMega328 processor, full set of IO pins and programmable via an ISP connector.


See my previous postings for the schematic and further details.  This was originally a project done for purposes of learning how to use KiCAD.

Sunday, September 20, 2015

Arduino Satellite/Sun Tracking

I have put together the recent bits and pieces of test code that I have been playing with to control my AZ/EL servo camera mount and an Arduino port of the old Plan13 code, originally written in Basic back in 1983 by James Miller G3RUH.  You can read about his efforts over on the AMSAT web site here.  Jim gives a great treatment of the math involved and example code in Basic.


A port of the code described on the page above has been made to the Arduino and I leveraged this code from the QRPTracker project web site.  With some simple changes, I have pulled together a simple satellite tracker to drive my AZ/EL camera mount.  It should be simple to scale this up to drive an actual satellite antenna array.


Here you can see the project pointing to the point in the sky where the ISS would be located at a point in time today.  I took the AMSAT Keplerian elements and processed the six passes today where the satellite was above the horizon at my location.  The code currently processes the entire 24 hour period in a day, and tracks in real time each of the passes.

The implementation of Plan13 is not particularly accurate due to the limited precision available in single precision floating point on the Arduino, but with the broad bandwidth of most amateur 144 and 440 MHz antennas, it is of little consequence even if the accuracy was only to the degree level.  In fact most low earth orbital satellites can be worked with a fixed elevation on a standard yagi, but the project has been interesting.  If there is interest, I will post my code, though most of it was gleaned from the efforts of others.

Beyond the code I have previously published, I have changed the mapping of the azimuth servo to be clockwise from north to match a magnetic compass.  I limit the calculated elevation values to those angles physically possible with my AZ/El mount.  I prevent movement of the servos if the satellite is currently below the horizon.

I would like to put a real-time clock in the circuit and have it calculate and pre-position the antenna to the point where the satellite rises, add a simple display, add the ability to load Keplerian elements from a wifi connection to the internet and provide for calibration of the servos.

On an Arduino Uno, I am able to load the entire two line elements (TLE) set from here into flash.  I have written a routine that will print out the start of every visible pass for my location for every satellite in the list (89 of them).  I suspect that it will be unusual to have the entire set in flash at any given time, but it can be done and still have room for the code.  Alternatively a storage card could be used to store the data.

On the mighty Arduino Uno, this table fills much of the flash and calculating the position of all 89 satellites for every minute of a 24 hour period takes quite a while.  A better approach might be for the process that grabs the KEPS also pre-calculate the start of pass and just generate a table along with the KEPs table.  This table could then drive a higher level process  to allow the selection a which satellites you want to track today.

The code calculates the effects of doppler and provides a tuning frequency for both the uplink and downlink if the frequencies are provided prior to the calculation.  For now I am passing in zeros, but a table of satellite frequencies and a bit of glue code to drive tuning a transceiver could fully automate tracking and tuning tasks.

This has been lots of fun and a precursor to building a LEO satellite station, but for this particular hobby AZ/EL mount, my next task will be to calculate the position of the sun and use this mount to keep a small solar panel array pointed at the sun for battery charging.  For that particular task, I am thinking of trying out a couple of approaches.  The first would be to calculate the position of sun relative to me as an observer.  The second would be to use an analogue approach where the output of the solar panel is used as feedback to drive a correction to the AZ/EL servos to keep the panel at maximum output regardless of the position of the sun.  I suppose in theory, I could just set the device out in the sun and let it figure out where to point to maximize output.  Who knows, it might decide to point at a white wall rather than at the sun when currently behind tree cover for example.  Should be fun!

Sunday, September 13, 2015

Arduino and Servos

This little project has been lying around waiting to be done for far to long, so I have resurrected it on a rainy Sunday and hope to make something useful out of it.

I have this little AZ/EL camera mount pair of servos that are quite stout.  They have external feedback potentiometers for the servo position, so even though they are geared down, they can be accurately positioned.


Coupling these up to an Arduino UNO is trivial.  I need one signal pin, power and ground.  Do not try to power servos from the USB power on the Arduino, use an external power supply.

The first task at hand was to decide on the range of pulse widths that would be allowed for the servo.  The defaults are 544 us to 2400 us.  The APIs decide that if you set a pulse width less than the minimum, then it must be an angle in degrees 0-359.  I didn't take the time to sort this out.  Instead, I decided arbitrarily that angle zero is 500 us. and everything is counterclockwise from there.

I placed a mark on the Azimuth gear right at the servo gear and through experimentation determined that a change of 768 microseconds would rotate the gear 360 degrees.



From there, it was simple to create a function that would convert an angle in degrees to the required microseconds pulse width to position the servo as desired.

// For the azimuth servo, angle is mapped 0-360 degrees to 500-128 us
int deg2usAZ(int angle)
{
  return map(angle, 0, 360, 500, 1268);

}

Now in the case of the elevation servo, we have some limitations to the range of motion due to the physical characteristics of the assembly.  I decided first to figure out where level was and call this zero degrees of elevation.



Next I determined the setting for negative 45 degrees of elevation (pointing down).



And finally, I determined the setting for pointing straight up.  Now it was trivial to create a function to map degrees to these settings.

// For elevation servo, angle is mapped from -45-90 to 1112 to 2220 us
int deg2usEL(int angle)
{
  return map(angle, -45, 90, 1112, 2220);

}

Now with the addition of a button, I have a simple test application that will postion to zero degrees azimuth and elevation when started and then move to 90 degrees elevation and 45 degrees of azimuth when the button is pressed..

// Servo test code

#include <Servo.h>

int button1 = 4; //button pin, connect to ground to move servo
int press1 = 0;

Servo servoEL;
Servo servoAZ;

void setup()
{
  pinMode(button1, INPUT);
  servoEL.attach(7, 500, 2400);
  servoAZ.attach(9, 500, 2400); 

  servoEL.writeMicroseconds(deg2usEL(0));
  servoAZ.writeMicroseconds(deg2usAZ(0));

  digitalWrite(4, HIGH);
}

// For the azimuth servo, angle is mapped 0-360 degrees to 500-128 us
int deg2usAZ(int angle)
{
  return map(angle, 0, 360, 500, 1268);
}

// For elevation servo, angle is mapped from -45-90 to 1112 to 2220 us
int deg2usEL(int angle)
{
  return map(angle, -45, 90, 1112, 2220);
}

void loop()
{
  press1 = digitalRead(button1);
  if (press1 == LOW)
  {
    servoEL.writeMicroseconds(deg2usEL(90));
    servoAZ.writeMicroseconds(deg2usAZ(45));
  }
  else 
  {
    servoEL.writeMicroseconds(deg2usEL(0));
    servoAZ.writeMicroseconds(deg2usAZ(0));
  }

}

So, next I obtained the Arduino port of Plan13 which is a port of some very old basic code for determining a satellite location given the latitude/longitude/elevation of an observer, the Keplerian Elements for the satellite of interest and the date/time of interest. Keplerian Elements (or just Keps), named after Johann Kepler [1571-1630] are a set of seven numbers called satellite orbital elements that define an ellipse oriented about the earth and place the satellite on the ellipse at a particular time.

I have the Arduino port of Plan13 up and running and calculating azimuth and elevation values.  The precision is really insufficient on the Arduino to have very accurate calculations due to the fact that double precision floating point is required, but the Arduino defines double precision as a single precision value.  So the calculations do not have the precision they might otherwise have, but I find it is quite sufficient for simple experiments like this.

The remaining work is to provide a little more glue logic to have what amounts to a desktop pointer to the satellite as it passes overhead.

More to come.
.

Saturday, September 5, 2015

Salmoncon X QSL cards

July 10-12 this year our QRP ham radio group had our annual camp-out at Valley Camp in North Bend, WA.  We have a lot of fun hanging out together and have great technical talks as well as ham radio operating at QRP levels (low power).  We also have fun activities like hidden transmitter hunts.  Here are a couple of the participants on the hunt.




This year, the group had about 100 contacts split 75%/25% between US and DX contacts.  Since this was the 10th annual event, I decided to make up some custom QSL cards to send to all the contacts we made over the weekend.



I got all of these cards finished up today, addressed, stamped and in the post.  So if you worked K7S during the weekend of 10-12 July, look for your QSL card in the post soon.

Arduino Experiments with Serial Speed

I have been having a play around with seeing how much serial speed I can get out of a standard Arduino Uno when it isn't really doing anything else.

I decided to try 1M baud initially on the standard Serial port.  The code I used is as follows:

void setup() 
{
  Serial.begin(1000000);
}

void loop()
{
  unsigned int a = 0;
  do
  {
    Serial.println(a);
  } while (++a > 0);  

}

I didn't even consider using the standard Serial Monitor application built into the IDE and instead grabbed a copy of PuTTY and set the serial port up at 1000000 baud.




At this baud rate, the Arduino had no issues and PuTTY correctly decoded the serial stream without issue.  So, if 1M is good, then 2M should be twice as good, right?

Well, not exactly.  at 2,000,000 baud the per character rate is twice the rate, but the overall throughput actually dropped visibily.  By my crude measurements, the throughput dropped by about 15%.  But, it is good to know that Arduino can push serial bits out at a respectable speed.

One side effect to note is that the IDE must override the Serial baud rate in order to program the device.  When running at these very fast serial rates, it can be a bit hit and miss.  If you have trouble, press the reset button and release it just before the compile finishes and you will help it recover.  Another alternative would be to use an external ISP programmer.

Another thing to keep in mind is that the IDE and PuTTY are both going to try and use the same COM port.  The IDE will use it for programming the chip.  PuTTY will use it to monitor the serial output.  In order for the IDE to be able to program the chip, it must be able to open the serial port.  So, you will have to terminate PuTTY before attempting to reprogram the Arduino.

I have read a lot of complaints lately about trying to debug Arduino code with the serial port output.  While upping the serial speed will not help in situations where interrupts are disabled or where time required to output debug data adversely affects the running code, being able to output the data orders of magnitude faster is certainly going to help in general.  If you are doing a lot of edit, compile, flash, run cycles, it might be better to use an ISP programmer so that PuTTY or whatever serial terminal application you use can stay connected to the debug serial port.

In my work with faster embedded processors, 2-3M baud is about the limit even on these devices with much faster core and peripheral clocks.  The Freescale K60 for example is running at 96 MHz in my configuration and at 2M baud on the debug port, I still get data loss, but it is solid at 1M baud.  I think that for high speed output or where you need to output debug information during interrupt processing, it may be a better choice to use SPI for debug output to a dedicated device that can capture and save the output such as a PC with a USB/SPI interface or another Arduino that can write to a storage device such as an SD card.

Tuesday, August 18, 2015

Android Phone (Samsung Galaxy Note 4) Power Usage

I was playing around recently with my Android phone and decided to take a look at it's power consumption when charging from my laptop USB port.



While the device comes with a quick charge power adapter, my little USB power meter is not expecting to draw a couple amps from a USB circuit, so in the interest of not damaging it, I gave it a standard USB 2.x port as a power source.  As we can see it is quite hungry and draws a consistent 2.1 watts in this mode, with only about 75 mA headroom on a standard USB charging rate of 500 mA.  The device does not seem to change its power consumption when using the device while charging or allowing it to sleep.  The device uses the typical method of various sense resistor values between the USB pins to determine if the special charger is in use or a standard one to limit current draw to USB specifications.

When charging on the quick charger, I am seeing about 1.27 amps when measured with my multi-meter, but it seems to walk around a bit from about 1.0 to 1.45 amps.

Once the charger has brought the battery up to 100% charge, I put it back on my little USB power meter and find a much more respectable 43 mA current draw.  Again, it doesn't seem to matter if the display is lit or not.



Sunday, July 26, 2015

Controller check-out

I have FINALLY found my ISP programmer and been able to check out the postage stamp controller and made it to the typical embedded "Hello World" application of a blinking LED.



The small board in the upper left is an ESP8622 wifi module that is hanging out from a previous project.  I have flashed the ATMega328 part with the following typical example code:

void setup()
{
  pinMode(13, OUTPUT);
}

void loop() 
{
  digitalWrite(13, HIGH);
  delay(1000);
  digitalWrite(13, LOW);
  delay(1000);

}

The USBTinyISP programmer had no problem flashing the code to this part, but at initial blush, it appeared that nothing was working.  Then after a very long delay, I saw the LED turn on.  Reducing the delay values to 100 resulted in approximatly 1 second timing of the LED flashes.  So, it appears that the clock was not running at 16 MHz.

On a whim, I decided to flash the optiboot boot loader to the device and let the device reset and then erase the chip and reflash the test application above with no boot loader present.  This in fact fixed the problem.  

In looking at the datasheet, I see the following:



So, it appears that out of the box, a new ATMega328 is going to default to a 1 MHz system clock.  I will have to look at the optiboot code, but I am sure it is fiddling with the clock select bits which would explain the behavior I have seen.

In any event, I now have a tiny little microcontroller that needs no boot loader (saving that space in flash) and can be used for my projects that can use such a controller.  I have a couple updates that I need do to the PCB and will publish my updates on OSHPark where anyone can order the boards as desired.  $5 for 3 boards is a bargain.  More to the point is that I got something useful out of an attempt to learn a little more about KiCAD.

Saturday, July 18, 2015

Postage Stamp Micro-controller

In a previous post I described a little one inch square micro-controller based on the ATMega328 which I decided to build in order to learn KiCAD.  I have the boards back from OshPark.



To my way of thinking, it is a pretty good bargain to be able to obtain this lot of boards for the grand total of $10 plus shipping.  In any event, I have built up one of the units as seen below.


I have a couple of errors I will need to fix.  The first is that the row of header pins on the left and right side are not quite on 0.1 inch centers.  I plugged the headers into a prototype board while soldering and they both bend towards the centre of the board, so I need to space them a little closer together.  Oops...  The top and bottom headers are mounted on the top side of the board to allow plugging the board into a prototype board and still having all pins available.  I would expect to typically hard-wire the board into whatever circuit it will be employed in and then mount it down with double sided tape.  This one however is for bread-boarding.

The second error is that I used the wrong footprint for the thru-hole crystals that I have on hand and so it cannot sit completely down on the board.  Oh sigh...  Easy enough to fix however and I will make those updates before I publish the board at OshPark.com.

The ATMega328 chips are raw ones with no boot loader, so I will have to program them with the ISP header.  This can be done with another Arduino or any of the ISP programmers such as the USBTinyISP.  No boot loader is required on this device as there is no USB or other programming interface other than the ISP connector so therefore the entire flash is available for your code.




Thursday, July 2, 2015

KiCAD

Since being laid off at Dynon Avionics, I have been looking around at schematic capture and PCB CAD packages.  Chris at Contextual Electronics has encouraged me to try out KiCAD.  I decided to give it a go and design a small 1 square inch microcontroller board with a full-on ATMega 328 processor and the minimum circuitry necessary to build an (almost) UNO clone.  The "almost" part is the fact that there will be no USB port on the board and it will need be programmed using the ISP connector.

The circuit diagramme I decided on looks like this from KiCAD:



Nothing here that doesn't need to be here for a minimal controller.  I have the entire I/O pin set that can be found on an UNO, plus since I use the TQFP package, I have two additional analogue input pins.

The board layout is pretty dense at one inch square.  Here is the board layout showing all of the layers from KiCAD.  I am not terribly impressed with KiCAD's copper pour capabilities, at least as far as I have explored them, but what do you want for free, eh?



Producing the Gerber files was trivial and I zipped up the lot of them and shipped them off to OSHPark to be manufactured.  Here is the screen shot of their rendering of the gerber files for the front and back of the board.



So, now we wait for the little purple envelope to arrive.  Meanwhile I will collect up the set of parts to build a few of these up.  This was an experiment to enable me to learn KiCAD and at the same time to produce a useful gadget for other projects.  Once I have verified the functionality is correct, I will publish the OSHPark boards so others can order them directly if desired.  At $5 for 3 of them, they are pretty cheap.

Overall, I found KiCAD relatively easy to learn, once I got a couple of key concepts down.  The main one was that hot keys apply to whatever your mouse is hovering over.  I found myself wanting to click to select something before hitting a hot key and this lead to all sorts of grief.  Moving objects once they are placed also needs a lot of work.  Rubber-banding of traces pretty much sucks.  Once I figured out how to drag an object, the best that KiCAD could do is straight line connect all the existing traces.  This ends up being pretty much worthless for anything other than adjustments of a few centimetres and you end up deleting all the traces and redrawing them anyway.

Selection of wires in schematics needs an easy way to select only a line segment or the entire line without having to draw box around it.  When you move the line, it needs to rubber-band the end points.  Straightening out a series of segments needs to collapse them into one segment and it needs to be able to easily break a line segment in two without disconnecting them.  All of these comments also apply to traces in the PCB layout editor.

I did like that KiCAD included the ability to view gerber files, a comprehensive footprint editor as well as building the notion of hierarchical schematics into the product as a key feature.  I cannot however get the bill of materials functionality to work, so I must be doing something wrong.  Always more to learn. 

For a free, open-source product, it is quite compehensive and there are a lot of part footprints available out-of-box and from 3rd parties.  Overall, I would continue to use it for hobby projects.  I would need to try and build at least a four layer board to have an opinion about suitability for commercial projects.

Thursday, June 25, 2015

Change of status

Today I accepted a Firmware Engineer position with LOUD Technologies in Woodinville, WA working with the Mackie Digital Audio Mixing team.  I will be starting mid-July, so my job hunt status appears to be over.  Yay!  Check out their web site when you get a chance.  http://loudtechinc.com/

Wednesday, June 3, 2015

Diode Ring Mixer - UPDATED

In an attempt to firstly melt more solder than I have of late, and secondly to try and actually understand the operation of a mixer, I have decided to build one up from discrete components, including (gasp) winding my own transformers.

The rather rough diagramme of the circuit I propose to build looks something like this which is based on the fine design by Pete Juliano N6QW found here:



I found a few 1N6263 Schottky diodes in my junk box and out of the lot was able to find four that matched on the forward voltage drop to the millivolt level at 0V300.  A quick look around found a bit of PCB and a 1/4 inch sheet metal hole punch allowed me to create simple Manhattan pads.  Here is the layout at the start of the build with just the diodes installed.



I wound coils on FT37-43 cores.  I seem to be out of enameled wire, so I am going to try this with 30 gauge insulated wire-wrap wire as the number of turns is small, though the coils are trifilar.  I wound the trfilar wires together using a drill motor and then wound the three wires onto the core.  Ten turns, both coils are identical.



Separating the coil winding ends, I check continuity on them and mark them with a Sharpie (why is a dull felt pen called a Sharpie?) on each end of the wire with no mark, a single mark or a double mark.  The no mark wire will be the single coil while the one and two mark wires will be hooked in series (observing phasing).  The left side as drawn in the circuit above, will have a 100 ohm variable resistor at the centre tap of the transformer secondary to allow small balance adjustments.  As suggested by Pete Juliano N6QW, but not drawn above, I plan to also include a 39 ohm resistor that is switchable to ground from the centre tap in order to intentionally unbalance the mixer.  This is useful for injecting carrier for tune-up purposes for example.

The final build, minus the output filter can be seen below.  I hope to get out of interview preparation mode soon so that I can have some time to check this out.  The trim pot is a 100 ohm multi-turn pot to allow fine tuning of the balance.  Connecting the top of the 39 ohm resistor to ground will deliberately unbalance the mixer.



Finally got a chance to see if this little guy is mixing or not.  I set the 100 ohm multi-turn pot to the middle and attached the output of a 20 MHz crystal oscillator to it along with a signal generator output and had a look at the output on the spectrum analyzer.  Firstly, we needed some connectors.


I gave it 20 MHz on one port, 1 MHz on the other, both at 7 dBm and obtained the following output.  Both the sum and difference were within 0.08 dBm of each other.


So, we seem to be mixing and with proper IF filtering following, should be good to go.  Many thanks to Wayne NB6M for helping with the testing.



Tuesday, May 26, 2015

Version 1.0 of Si5351 Signal Generator

I have completed version 1.0 of my signal generator.  There will definitely be a revision as there are a number of things still to straighten out, but overall I am pleased with the result.

The 3D printed box with 20% fill on the box walls, was easy to drill, though I recommend a drill press.  I made the top cover thick enough that my controls (rotary encoder and switch) could be threaded into the plastic and no surface nuts required.  This seems to work well, though it remains to be seen how it holds up over time.


The point-to-point wiring inside is a pain in the tush.  Version B will definitely have a single PCB with all the components (display, power switch, rotary encoder, Trinket PRO, Si5351 and battery) plugged into it and mounted on stand-off spacers to the front panel.  I also definitely need to screw down the display.  The friction fit to the panel is good, but not good enough.  The picture below was taken before the Si5351 was installed.


So, overall I am pleased and look forward to using this new addition to my tools.

Monday, May 25, 2015

So, what do you think?

The second round of 3D printing I think is very close to what I need for this simple Si5351 signal generator.  The new mounting method for the display works well.  A little bit of plastic needed to be trimmed in order for the display to fit into place, but not bad.


The display fits very snugly so I will let it be for now.  I will likely drill out the corner holes and install screws to keep things in place.


The rotary encoder will  be mounted to the right of the display.  In reality, with a single line of code I can flip the rotation of the display by 180 degrees so it is easily adaptable to left or right handed operation.

I have been considering how to implement a single control interface using the rotary encoder and its built in push button.  I think that I will use a single press of the encoder knob to cycle between the three clocks.  The active clock is the one displayed in yellow.

Turning the rotary encoder will change the currently selected digit of the active frequency.  To change the tuning rate, I am considering rotating the encoder while holding the knob depressed to select which digit is changed by the encoder.

Otherwise, there will be a power switch and the three clock SMA output connectors.  I am unsure if I will mount them on the face along side of the display or if I will mount them on the side of the case.  I plan to use a small LiPo battery to power the entire device.

Overall, I am pleased with 3D printing as a medium for prototyping things like this case.  The aesthetics are not the greatest, but the strength of the case and the amount of accuracy of the print to the object design is quite good.  It will be interesting to see if my son's calibration efforts on the printer will have any effect on the aesthetics.

Si5351 Signal Generator (continued)

My first attempt at 3D printing a case for my signal generator project was, shall we say a good learning process.  The printer printed what I created, but a number of rather fatal flaws in the design indicated the need to scrap it and try again.

Meanwhile, I have decided on the controller that will be dedicated to this project.  Given that the Si5351 and my TFT display are both 3V3 devices, I decided to utilize the Adafruit PRO Trinket 3V3 device.  This is essentially a 3V3, 12 Mhz Arduino Uno without the USB controller and with a couple of I/O pins being dedicated to other purposes and therefore unavailable (pins 2 and 7).



Since I developed the code on the ATMega2560, it seemed that my next task would be to convert the code to run on the ATMega328 processor found in this device and to make some final I/O pin assignments with this device in mind.  The conversion was a little more complicated than I first anticipated due to my sloppy coding on the first go-around.

I didn't want to put header pins on this board for this project, so I used my Arduino Uno to verify the code changes (and to clean up my sloppiness).  I have wired everything using the 3V3 source, but since the I/O pins are 5V on the Uno, I used an NTE4050B buffer to do the level conversion to 3V3 on the pins going to the display.  This will not be necessary once I wire up the intended controller as it is a 3V3 device.  The Si5351 is not shown in this image.



The setup of the 1.6.4 IDE to support the Trinket PRO was fairly straight forward, but here still be dragons.  The Adafruit web site recommends just downloading the IDE with the Adafruit board support definitions already installed.  This is convenient, but creates a set of additional problems which I will likely make the subject of a separate post.

The other thing I learned is that if a class name defined in your code happens to collide with a class name in an existing library in your installation, even if that library is not used in your project, that library will still be compiled and can cause duplicate symbol errors or can surface dependent library missing errors.  I ran into this because I have in my project a file that defines a class for the Si5351, but I also have installed the Adafruit_Si5351 library into the IDE which also defines a class with that name, though in a different name space.  Even though I am not using it, the Adafruit_Si5351 library gets compiled by my project, but since that library depends on the Adafruit_Unified_Sensor library which I had not installed, I got a compile error.  Had the compiler error not occurred, I would have likely gotten link errors.  I need to dig into this in more detail as I believe this to be a bug in the current 1.6.4 IDE from Arduino.cc but need to simplify the reproduction of the issue before filing a bug report.  A lot of work has gone into this version's handling of libraries, but here still be dragons.  My workaround was to delete the Adafruit_Si5351 library from my installation and able to recompile without errors.

So, now I need to make some software changes to allow me to eliminate the extra push button in favor of just the rotary encoder, its push button and the display being the entire user interface.  I also want to provide a mechanism to enable/disable each clock as desired.

My son meanwhile has been busy printing my latest attempt at a case for this project.  I am certain this will not be the final print, but the changes made were to eliminate the plastic posts to hold the display and to make the front panel thicker to give it more rigidity.  I increased the lip size and provided a recess for the display to sit in.  Hopefully it will be a nice friction fit, but if not, I will drill the case and install plastic screws to hold it in place at the corners.





3x4x1.5 inch project box
My son is in the process of dialing in his 3D printer.  This one looks a little rough as he has sped up the head speed considerably in order to cut down on print time.  His extrusion feed rate is a little high and his X vs. Y axis accuracy in this print is about 0.06 mm (about 2.5 mil).  After calibration his accuracy is now about 0.02 mm (about 0.78 mil) so the next print should be a lot cleaner.  His extrusion feed rate causes his interior dimensions to suffer a bit as he is putting down too much material.  He should have that dialed in before he reprints this for me.  We also went to a 20% fill rather than solid, so it will be interesting to see how well that can be drilled.  The top is a tight fit so I will likely need to cut a thumbnail slot along the edge so I can get it off the case once the display fills the hole.  Once he has the new calibrations in place and I have verified my geometry will work with the parts I have in hand, we will print one more and it should look a lot better.

More to come...