Action Potential Generator Circuit

Few biological cells are as interesting to the electrical engineer as the neuron. Neurons are essentially capacitors (with a dielectric cell membrane separating conductive fluid on each side) with parallel charge pumps, leak currents, and nonlinear voltage-dependent currents. When massively parallelized, these individual functional electrical units yield complex behavior and underlie consciousness. The study of the electrical properties of neurons (neurophysiologically, a subset of electrophysiology) often involves the development and use of sensitive electrical equipment aimed at studying these small potentials produced by neurons and currents which travel through channels embedded in their membranes. It seems neurophysiology has gained an emerging interest from the hacker community, as evidenced by the success of Back Yard Brains, projects like the OpenEEG, and Hack-A-Day’s recent feature The Neuron – a Hacker’s Perspective.

In pondering designs for complex action potential detection and analysis circuitry, I realized that it would be beneficial to be able to generate action-potential-like waveforms on my workbench. The circuit I came up with to do this is a fully analog (technically mixed signal) action potential generator which produces lifelike action potentials.

 

Cellular Neurophysiology for Electrical Engineers (in 2 sentences): Neuron action potentials (self-propagating voltage-triggered depolarizations) in individual neurons are measured in scientific environments using single cell recording tools such as sharp microelectrodes and patch-clamp pipettes. Neurons typically rest around -70mV and when depolarized (typically by external excitatory input) above a threshold they engage in a self-propagating depolarization until they reach approximately +40mV, at which time a self-propagating repolarization occurs (often over-shooting the initial rest potential by several mV), then the cell slowly returns to the rest voltage so after about 50ms the neuron is prepared to fire another action potential. Impassioned budding electrophysiologists may enjoy further reading Active Behavior of the Cell Membrane and Introduction to Computational Neuroscience.

The circuit I describe here produces waveforms which visually mimic action potentials rather than serve to replicate the exact conductances real neurons employ to exhibit their complex behavior. It is worth noting that numerous scientists and engineers have designed more physiological electrical representations of neuronal circuitry using discrete components. In fact, the Hodgkin-Huxley model of the initiation and propagation of action potentials earned Alan Hodgkin and Andrew Huxley the Nobel Prize in Physiology and Medicine in 1936. Some resources on the internet describe how to design lifelike action potential generating circuits by mimicking the endogenous ionic conductances which underlie them, notably Analog and Digital Hardware Neural Models, Active Cell Model, and Neuromorphic Silicon Neuron Circuits. My goal for this project is to create waveforms which resemble action potentials, rather than waveforms which truly model them. I suspect it is highly unlikely I will earn a Nobel Prize for the work presented here.

The analog action potential simulator circuit I came up with creates a continuous series action potentials. This is achieved using a 555 timer (specifically the NE555) in an astable configuration to provide continuous square waves (about 6 Hz at about 50% duty). The rising edge of each square wave is isolated with a diode and used to charge a capacitor*. While the charge on the capacitor is above a certain voltage, an NPN transistor (the 2N3904) allows current to flow, amplifying this transient input current. The capacitor* discharges predictably (as an RC circuit) through a leak resistor. A large value leak resistor slows the discharge and allows that signal’s transistor to flow current for a longer duration. By having two signals (fast and slow) using RC circuits with different resistances (smaller and larger), the transistors are on for different durations (shorter and longer). By making the short pulse positive (using the NPN in common collector configuration) and the longer pulse negative (using the NPN in common emitter configuration), a resistor voltage divider can be designed to scale and combine these signals into an output waveform a few hundred mV in size with a 5V power supply. Pictured below is the output of this circuit realized on a breadboard. The blue trace is the output of the 555 timer.

*Between the capacitance of the rectification diode, input capacitance of the transistor, and stray parasitic capacitance from the physical construction of my wires and the rails on my breadboard, there is sufficient capacitance to accumulate charge which can be modified by changing the value of the leak resistor.

This circuit produces similar output when simulated. I’m using LTspice (free) to simulate this circuit. The circuit shown is identical to the one hand-drawn and built on the breadboard, with the exception that an additional 0.1 µF capacitor to ground is used on the output to smooth the signal. On the breadboard this capacitance-based low-pass filtering already exists due to the capacitive nature of the components, wires, and rails.

A few improvements naturally come to mind when considering this completed, functional circuit:

  • Action potential frequency: The resistor/capacitor network on the 555 timer determines the rate of square pulses which trigger action potentials. Changing these values will cause a different rate of action potential firing, but I haven’t attempted to push it too fast and suspect the result would not be stable is the capacitors are not given time to fully discharge before re-initiating subsequent action potentials.
  • Microcontroller-triggered action potentials: Since action potentials are triggered by any 5V rising edge signal, it is trivially easy to create action potentials from microcontrollers! You could create some very complex firing patterns, or even “reactive” firing patterns which respond to inputs. For example, add a TSL2561 I2C digital light sensor and you can have a light-to-frequency action potential generator!
  • Adjusting size and shape of action potentials: Since the waveform is the combination of two waveforms, you can really only adjust the duration (width) or amplitude (height) of each individual waveform, as well as the relative proportion of each used in creating the summation. Widths are adjusted by changing the leak resistor on the base of each transistor, or by adding additional capacitance. Amplitude and the ratio of each signal may be adjusted by changing the ratio of resistors on the output resistor divider.
  • Producing -70 mV (physiological) output: The current output is electirically decoupled (through a series capacitor) so it can float at whatever voltage you bias it to. Therefore, it is easy to “pull” in either direction. Adding a 10k potentiometer to bias the output is an easy way to let you set the voltage. A second potentiometer gating the magnitude of the output signal will let you adjust the height of the output waveform as desired.
  • The 555 could be replaced by an inverted ramp (sawtooth): An inverted ramp / sawtooth pattern which produces rapid 5V rising edges would drive this circuit equally well. A fully analog ramp generator circuit can be realized with 3 transistors: essentially a constant current capacitor charger with a threshold-detecting PNP/NPN discharge component.
  • This action potential is not all-or-nothing: In real life, small excitatory inputs which fail to reach the action potential threshold do not produce an action potential voltage waveform. This circuit uses 5V rising edges to produce action potential waveforms. However, feeding a 1V rising edge would produce an action potential 1/5 the size. This is not a physiological effect. However, it is unlikely (if not impossible) for many digital signal sources (i.e., common microcontrollers) to output anything other than sharp rising edge square waves of fixed voltages, so this is not a concern for my application.
  • Random action potentials: When pondering how to create randomly timed action potentials, the issue of how to generate random numbers arises. This is surprisingly difficult, especially in embedded devices. If a microcontroller is already being used, consider Make’s write-up on the subject, and I think personally I would go with a transistor-based avalanche nosie generator to create the randomness.
  • A major limitation is that irregularly spaced action potentials have slightly different amplitudes. I found this out the next day when I created a hardware random number generator (yes, that happened) to cause it to fire regularly, missing approximately half of the action potentials. When this happens, breaks in time result in a larger subsequent action potential. There are several ways to get around this, but it’s worth noting that the circuit shown here is best operated around 6 Hz with only continuous regularly-spaced action potentials.

In the video I also demonstrate how to record the output of this circuit using a high-speed (44.1 kHz) 16-bit analog-to-digital converter you already have (the microphone input of your sound card). I won’t go into all the details here, but below is the code to read data from a WAV file and plot it as if it were a real neuron. The graph below is an actual recording of the circuit described here using the microphone jack of my sound card.

import numpy as np
import matplotlib.pyplot as plt
Ys = np.memmap("recording.wav", dtype='h', mode='r')[1000:40000]
Ys = np.array(Ys)/max(Ys)*150-70
Xs = np.arange(len(Ys))/44100*1000
plt.figure(figsize=(6,3))
plt.grid(alpha=.5,ls=':')
plt.plot(Xs,Ys)
plt.margins(0,.1)
plt.title("Action Potential Circuit Output")
plt.ylabel("potential (mV)")
plt.xlabel("time (ms)")
plt.tight_layout()
plt.savefig("graph.png")
#plt.show()

Let’s make some noise! Just to see what it would look like, I created a circuit to generate slowly drifting random noise. I found this was a non-trivial task to achieve in hardware. Most noise generation circuits create random signals on the RF scale (white noise) which when low-pass filtered rapidly approach zero. I wanted something which would slowly drift up and down on a time scale of seconds. I achieved this by creating 4-bit pseudo-random numbers with a shift register (74HC595) clocked at a relatively slow speed (about 200 Hz) having essentially random values on its input. I used a 74HC14 inverting buffer (with Schmidt trigger inputs) to create the low frequency clock signal (about 200 Hz) and an extremely fast and intentionally unstable square wave (about 30 MHz) which was sampled by the shift register to generate the “random” data. The schematic illustrates these points, but note that I accidentally labeled the 74HC14 as a 74HC240. While also an inverting buffer the 74HC240 will not serve as a good RC oscillator buffer because it does not have Schmidt trigger inputs.

The addition of noise was a success, from an electrical and technical sense. It isn’t particularly physiological. Neurons would fire differently based on their resting membrane potential, and the peaks of action potential should all be about the same height regardless of the resting potential. However if one were performing an electrical recording through a patch-clamp pipette in perforated patch configuration (with high resistance between the electrode and the internal of the cell), a sharp microelectrode (with high resistance due to the small size of the tip opening), or were using electrical equipment or physical equipment with amplifier limitations, one could imagine that capacitance in the recording system would overcome the rapid swings in cellular potential and result in “noisy” recordings similar to those pictured above. They’re not physiological, but perhaps they’re a good electrical model of what it’s like trying to measure a physiological voltage in a messy and difficult to control experimental environment.

This project was an interesting exercise in analog land, and is completed sufficiently to allow me to move toward my initial goal: creating advanced action potential detection and measurement circuitry. There are many tweaks which may improve this circuit, but as it is good enough for my needs I am happy to leave it right where it is. If you decide to build a similar circuit (or a vastly different circuit to serve a similar purpose), send me an email! I’d love to see what you came up with.


     

TTL Triggered Stimulus Generator

I was presented with a need to rapidly develop a pulse generator to take a TTL input and output a programmable output (for now 0.1 ms pulses at 20 Hz for as long as the input is high). I achieved this with a one-afternoon turnaround and the result looks great! This post documents the design and fabrication of this prototype device, with emphasis placed on design considerations and construction technique. It is also worth noting that by stocking large quantities of frequently-used items, inventors can build beautiful and functional prototypes for new ideas at the drop of a hat. While it’s easy to inexpensively accumulate tens of thousands of passive components (resistors, capacitors, etc.), it’s the slightly more expensive components that people tend to order only when they need it for a project. However, paying high shipping rates or waiting months for items to arrive from overseas dramatically increases the barrier for initiating new projects. In my own workshop I have noticed that stocking large volumes of slightly more costly items (inductors, microcontrollers, connectors, enclosures, LED bezels, etc.) lowers the barrier for me to start new projects, and has proved to be a good investment! Now I can build a product on the same day that I have the idea! Today’s idea takes the form of a TTL-controlled pulse generator for physiology applications.

I designed the enclosure before I designed the circuit. Metal enclosures are always expensive compared to their plastic counterparts. Steel enclosures are difficult to drill, and aluminum enclosures are expensive. My most cringe-worthy stocking expenditure is ordering metal enclosures in quantities of 10+. The last I checked this specific one is listed as, “Aluminum Instrument Box Enclosure Case+Screw For Project Electronic 26X71X110MM” and is a little under $4 each. Brass hex stand-off nuts and black steel screws don’t exactly match the aluminum, but they’re what I had on hand. I knew I would need power and a BNC input and output, so I put those 3 on the back. I wasn’t sure about the exact functionality of this device (and it may change after it is initially implemented) but I thought a single button and two LEDs would be a good starting point.

The circuit demonstrates the general flow of this device: a microcontroller-controlled project with a buffered output. I drew this schematic after I finished the build (I kept adding passives here and there as I tested it out) but before I started I knew the gist of how I would organize the project. Mentally I knew that as long as my microcontroller (ATTiny2313) could sense the TTL input and had control over all outputs (LEDs and BNC alike), I had a lot of flexibility to control the operation of this device in software. I used a generic LM7805 linear voltage regulator with a few decoupling capacitors to take a who-knows-what input voltage (up to 40V) and turn it into a stable 5V output. Note that both inputs (the BNC TTL input and the push-button) have decoupling capacitors near the microcontroller input pin to aid in debouncing.

I’m leaning on a 74HC541 inverting line driver to clamp the output voltage firmly at TTL levels. The microcontroller (an ATTiny2313) isn’t really designed to source of sink much current (I think it’s rated to 20 mA max) and I don’t know about the input circuitry of the stimulus isolator I intend to control (and don’t forget about the impedance of 50-ohm cable). The line driver helps me take some of the pressure off the microcontroller and help me feel better about reliably driving the output BNC.

Should I have optically isolated the input? Well, probably not… the application at hand is low importance. If I wanted to rely on optical isolation I would probably lean on the H11B1 as previously used in my opto-isolated laser build. In retrospect I kind of wish I had just because it would have been cooler!

I added a header to allow me to program the microcontroller with a programmer configured with test clip grabbers. I have an AVR ISP MKII (clone), and building a programming adapter that uses test clips was one of the best decisions I ever made! It makes programming (and the inevitable re-programming) a breeze.

The program isn’t too complex. It uses a polling method to continuously check for the state of the input TTL. When it’s high, it starts a new cycle (0.1 ms pulse, 49.9ms delay, yielding 20 Hz). The code is ready to add a “mode select” feature (which uses the front-panel push-button to select different stimulation protocols), but that functionality is not included in the example below. Note that a lot of the millisecond and microsecond delays are empirically determined by picking a value and checking its output on the oscilloscope. I should note that absolute timing isn’t critical for my application, as long as it’s consistent. For this reason I’m not relying on the internal RC clock (which is temperature sensitive), but instead am using an external 20MHz crystal as a time source. It’s still temperature sensitive (and so are the loading capacitors on each side of it), but dramatically less so than the RC option. Note that the crystal wasn’t in the original photos, but it was added for later photos.

Configure the ATTiny2313 to use an external crystal clock source

@echo off
avrdude -c usbtiny -p t2313 -U lfuse:w:0xff:m -U hfuse:w:0xdf:m -U efuse:w:0xff:m
pause

The core program (main.c)

#define F_CPU 20000000UL
#include <avr/io.h>
#include <util/delay.h>

volatile char state=0;

void output_HIGH(){PORTB|=(1<<PB4);}
void output_LOW(){PORTB&=~(1<<PB4);}
void LED1_ON(){PORTB|=(1<<PB3);}
void LED1_OFF(){PORTB&=~(1<<PB3);}
void LED2_ON(){PORTB|=(1<<PB2);}
void LED2_OFF(){PORTB&=~(1<<PB2);}

void singlePulse_20Hz_100us(){
	output_HIGH();
	_delay_us(100);
	output_LOW();
	_delay_us(900);
	LED2_ON();
	_delay_ms(20);
	LED2_OFF();
	_delay_ms(28);
	_delay_us(920);
}

void poll(){
	if ((PIND&(1<<PD4))){singlePulse_20Hz_100us();}
}

int main(void){
	DDRB=255; // all outputs
	DDRD=0; // all inputs
	PORTD=(1<<PD5); // pull front button high
	LED1_ON();
	for(;;){
		poll();
	}
}

Compile “main” and load it onto the ATTiny2313

@echo off
del *.elf
del *.hex
avr-gcc -mmcu=attiny2313 -Wall -Os -o main.elf main.c -w
avr-objcopy -j .text -j .data -O ihex main.elf main.hex
pause
avrdude -c usbtiny -p t2313 -U flash:w:"main.hex":a 

I didn’t think to check my height profile! I got lucky, and things fit fine. Socketed ICs can be close calls, and so can vertically-installed electrolytic capacitors. Now that it was programmed and everything fit, it was time to seal it up and make labels.

The finished product looks great! Never underestimate the power of clear labels and square outlines. Following deployment, a couple screws will let me open it up and access the programming header in case I need to change the stimulation protocols stored in the microchip. I am pleased with how professional of a result I was able to achieve in one sitting! I look forward to seeing how this device works for my application.

PS: Microcontroller code for this project (and many others) is stored in my ever-growing AVR-Projects GitHub page: https://github.com/swharden/AVR-projects

PPS: I smiled when a Google search revealed the PulsePal, “A low-cost programmable pulse generator for physiology and behavior”.


     

Controlling Speakers with RADAR

I just finished building a device that uses RADAR to toggle power to my speakers when it detects my hand waiving near them! I have some crummy old monitor speakers screwed to a shelf, and although their sound is decent the volume control knob (which also controls power) is small and far back on my work bench and inconvenient to keep reaching for. I decided to make a device which would easily let me turn the speakers on and off without having to touch anything. You could built a device to detect a hand waive in several different ways, but RADAR (RAdio Detection And Ranging) has got to be the coolest!

 

This project centers around a 5.8 GHz microwave radar sensor module (HFS-DC06, $5.22 from icstation.com, + 15% discount code haics) which senses distance (sensitivity is adjustable with a potentiometer) and in response to crossing a threshold it outputs a TTL pulse (the duration of with is adjustable with another potentiometer). I ran the output of the module through divide-by-two circuit (essentially a flip-flop) so that an object-detect event would toggle a line rather than pull the line high for each detection. I didn’t have a cheap flip-flop IC on hand (the 74HC374 comes to mind, $0.54 on Mouser) but I did have a 74HC590 8-bit binary counter on hand ($0.61 on Mouser) which has a divide-by-two output. I used the radar sensor and this IC to produce a proximity-toggled TTL signal which enabled/disabled current flowing through a power n-channel MOSFET. All together this let met create a device with two DC barrel jacks (an input and an output), and current delivery on the output could be toggled with proximity sensing.

Design

I wanted this device to be extremely simple, with a single input DC jack and single output jack and no buttons or knobs. It’s a funny feeling making a user input device with no drill holes in the enclosure! The design is so simple it’s not worth reviewing in detail. The 15V line (which in reality could be almost anything) is brought to 5V with a LM7805 linear voltage regulator. Decoupling capacitors are commonly placed on the input and output of the regulator, but since the function is to toggle a switch I didn’t find it necessary (there is no downstream signal I wish to preserve the integrity of). The radar module has only 3 connectors: +5v, GND, and OUT. Out produces a high pulse when it detects something close. The output of the OUT signal is fed into a divide-by-two stage which is really just a 74HC590 8-bit binary counter taking output of the div/2 pin. That output is fed into an IRF510 N-channel MOSFET to switch current flow on the “DC output” on and off. A Darlington transistor (i.e., TIP122) would probably work fine too, but there would be a slightly greater voltage drop across it. Any power MOSFET would have worked, but I had a box of IRF510s on hand so I used one although they are more expensive ($0.82 on Mouser). Not shown is a status LED which is also on the output of the divide-by-2 chip (with a current limiting series resistor).

Construction

glued the radar module to the wall of a plastic enclosure. Isn’t radar messed-up by glue and plastic? Yes! But I’m not sensing things 50 feet away. I’m sensing a hand moving a few inches away. To this end, I experimented with how much glue and how thick plastic I needed to distort the signal enough so that it would only activate when I put my hand near it (as opposed to sitting down at my desk, which could also trigger the sensor without these attenuating structures). I found that aluminum tape further dampened the response, but luckily for the aesthetics of the build I didn’t have to use it. Also, rather than take time making a PCB (or even using perfboard), I found point-to-point construction quite sufficient. I hot glued the counter IC to the radar module, wired it all together, and it was done! A little black plastic LED bezel was a nice touch with the diffuse blue LED.

Alternative Designs

RADAR was a cool way to accomplish this task, but there certainly are additional methods which could achieve a similar result:

  • IR (infrared) – By pulsing IR and sensing the reflected signal intensity with an IR-filtered photo-transistor, you could invisibly detect presence of an object in front of the sensor. Adjusting the amplification of the photo-transistor (and/or diffusion of its lens or enclosure) could adjust for distance. In fact, many companies make paired IR-LED / IR-phototransistor modules specifically for this task. However, if you have a TV remote control or other device which uses IR to communicate, it could screw with this signal.
  • Sonar – Instead of light, pressure waves (sound) could be used to sense distance due to the time delay between an audio emission and detection of its reflection. Presumably an ultrasonic transducer would be used to prevent perpetual annoyance of those living in the area. Ultrasonic distance sensor modules can also be found online for this purpose. These technologies are what is commonly used by vehicles to detect objects in their path while backing-up, alerting the driver with a beep. The downside of this method is that it would not work inside an enclosure. Aesthetically, I didn’t want to have two silver screen-covered cans staring at me.

Radar Module Teardown

I had two of these modules on hand, so after I got this project working with one I used the other to conduct a destructive teardown. What I found inside was interesting! If someone were really interested, there may be some potential for hackability here. Aside from the microwave PCB goodness I found two primary ICs: the LM2904 dual op-amp and an ATTiny13 8-bit microcontroller. I was really surprised to find a microcontroller in here! With so much analog on these boards, it seemed that a timed pulse could be accomplished by a 555 or similar. A single-quantity ATTiny13 is $0.58 on Mouser (as compared to $0.36 for a 555) but maybe when you add the extra discrete components (plus cost of board space) it makes sense. Also, I’m not entirely sure how this circuit is sensing distance and translating it into pulses so perhaps there is some more serious computation than I’m giving it credit for.

The presence of an ATMEL AVR in this RADAR module is a potential site for future hacks. I’d be interested to solder some wires to it and see if I could extract the firmware. In any large scale commercial products the read/write access would be disabled, but with small run modules like this one seems to be there’s a chance I could reprogram it as-is. If I really wanted to use this layout but write a custom program for the micro I could desolder it and lay my own chip on the board. For now though, I’m really happy with how this project came out!


     

Precision Pressure Meter Project

I just completed building a device capable of measuring temperature to one hundredth of a degree Celsius and pressure to one ten-thousandth of a PSI! This project is centered around an ICstation MS5611 temperature sensor breakout board which was small enough to fit inside of a plastic syringe. The result is a small and inexpensive pressure sensor in a convenient form factor with a twist connector (a Luer-Lok fitting) that can be rapidly attached to existing tubing setups. Although the screw attachment would work well for industrial or scientific applications, I found that the inner connector (the non-threaded plastic nub with 6% taper) made a snug and air-tight connection with my CO2-impermanent aquarium tubing.

this MS5611 breakout board is small enough to fit inside a 10 mL syringe

I documented this project thoroughly so others can learn about the design process that goes into making one-off prototypes like this. The video is quite long considering how simple the task seems (read a number from a sensor and display it on a screen), but it gives a lot of tips and insights into rapidly making professional looking one-off projects like this. Reading datasheets can be intimidating for newcomers too, and this video walks through how to figure out how to bang out I2C commands to a new sensor using a Bus Pirate – a really convenient skill to have for hobby electrical engineers like me! After it’s working well with the sensor/computer interface you can move to the microcontroller level with confidence. Since no one has posted code for how to interface this sensor directly with the microcontroller platform I intended to use (AVR-GCC, notably not Arduino), my build process started by poking around with a Bus Pirate to learn how to interact with the device using  I2C commands. Once I was able to initiate temperature and pressure readings and pull its values by hand using the Bus Pirate, I wrote a Python script to automate the process (using PySerial to interact with the Bus Pirate) and allow recording and graphing of real-time pressure and temperature information. I then used a logic analyzer to glance at the data exchanged between the Bus Pirate and the pressure sensor (mostly for my own satisfaction, and to help with debugging in the future). Finally, I ditched the computer and had an ATMega328 microcontroller pull temperature/pressure readings and display them on a 16×2 HD44780 character LCD display beautifully framed with a laser-cut LCD bezel (from Tindie user widgeneering). I used a USB connector to give the device power (though there’s no reason it couldn’t run off of 3xAA batteries) and CAT5 cable as a convenient connector between the display and the sensor. After assembling everything and making some labels, the final product looks quite professional!

Project Summary Video

This video is quite extensive. It explores the design process for one-off projects like this, with extra time spent on the difficult parts that often pose the greatest challenges to newcomers (exploring datasheets, banging out I2C commands with a new sensor). I don’t see this part of the design process discussed too often in engineering videos, so I hope it will be an insightful and inspiring resource to people just starting to work with custom electronics and prototype design. Another group of people who benefit from watching the video are those who don’t know much about the design process of embedded devices, but will quickly realize that building a prototype device to do something as simple as reading a number from a sensor and displaying it on a screen can take an immense amount of insight, work, troubleshooting, and effort to create.

 

About the MS5611 Temperature and Pressure Sensor

The breakout board I’m using provides 5V access to the I2C interface of the MS5611. This is convenient because the MS5611 requires 3.3V and many microcontroller applications run at 5V. The MS5611 itself is the small (5mm by 3mm) silver rectangle on the side of the board. The MS5611 datasheet has all the information we need to know to get started poking around its I2C bus! The general idea is that it has an imperfect pressure sensor on board. During production the pressure sensors are produced with slightly different offsets and gains. Further, the pressure sensor varies its output as a function of temperature. They included a temperature sensor on there too, but that also varies by offset and gain due to production! To yield highly precise absolute pressure readings, the factory calibrates every device individually by storing six 16-bit calibration values in a program memory. They represent the sensitivities and offsets of these sensors.

When run through an algorithm (given a whole page in the datasheet), the 6 factory-programmed calibration values (16-bit integers) can be combined with the raw temperature and pressure readings (24-bit integers) to yield incredibly accurate and precise temperature and pressure readings down to 0.01 degree Celsius and 0.012 millibar (0.00017 PSI). This accuracy is enough to be able to measure changes in altitude of 10 centimeters!

These are some photos of the break-out board from the company’s product page and a few more taken from my USB microscope zoomed in on the sensor itself. If I feel inspired, I may use my hot air tool to lift the sensor off the board and incorporate into a future, smaller design. I’ll save that project for another day!

Using a Bus Pirate to Communicate with the Sensor

After reading the datasheet I learned the general flow of how to read data from this sensor. It was a three step command process for both temperature and pressure:

  • Tell the device what to measure and with what precision [by sending 1 byte]
    This is in the commands section (page 9/20) of the datasheet. Command 0x48 will tell it to use maximum oversampling ratio (OSR) to convert D1 (the digital pressure value). Highest OSR (4096) means the most precise reading but a slightly slower reading (9.04 ms) with higher current draw (12.5 µA at 1 Hz) as compared to the lowest OSR (256, 0.6 ms, 0.9 µA). 
  • Tell the device you are ready to perform an ADC read [by sending 1 byte]
    The byte you send to read the ADC is always 0x00. Don’t proceed to this step until the conversion has been given time to complete or your reading will be zero.
  • Read the ADC result [by reading 3 bytes]
    The ADC result will always be an 18-bit integer.

This was a great use for my Bus Pirate! Without the Bus Pirate in order to debug this device I would have needed to make a circuit board, wire-up a microcontroller, figure out how to program that microcontroller to interact with the sensor (with very limited hardware debug tools), and send readings (and debug messages) to a computer via a USB serial port. Also, I’d have to add bidirectional serial communication code if I wanted it to be interactive. What a nightmare! Recently I started really valuing my Bus Pirate as a way to immediately hook up to a new sensor out of the box and interactively pull data from it within a few seconds. To hack this to my Bus Pirate I soldered-on female headers (instead of soldering on the pins that came with the breakout board). The Bus Pirate pin descriptions page shows how to hook up an I2C device. It’s important to note that the sensor board will not receive power (and its LED won’t light up) until you send the “W” command to the Bus Pirate.

Here are the commands I use with the Bus Pirate to connect with the sensor. If you can’t get this part to work, I don’t recommend challenging using a microcontroller to pull I2C data from this part! This is kind of fool proof, so this stage not working means you’ve read the datasheet incorrectly and don’t know how to interact with the sensor as well as you thought you did, or that there is a hardware or connectivity issue with the circuit. All of this is in the video posted above, so watching that part of the video may help you get an idea of what it looks like interacting with circuits like this with a Bus Pirate. Also, be sure to review the Bus Pirate I2C guide.

  • Open RealTerm and connect to the Bus Pirate
    • change display mode to Ansi
    • set baud to 115200 baud (no parity, 8 bits, 1 stop bit)
  • # to reset the Bus Pirate (optional)
  • m to set mode
  • 4 to select I2C
  • 3 to select 100 KHz
  • W to enable power (the red LED on the sensor should light up)
  • P to enable pull-up resistors (no errors should be displayed)
  • (1) scan for I2C devices (the sensor should be displayed, likely as oxEE)
  • Let’s make a read! This is how to read raw pressure:
    • [0xEE 0x48] to do the 4096 OCR D1 read
    • [0xEE 0x00] to prepare to read the ADC
    • [0xEF r:3] to read 3 bytes

For the most accurate readings, use the algorithms on page 7/20 of the datasheet to use the calibration variables (C1-C6) in combination with pressure (D1) and temperature (D2) to produce an accurate temperature and pressure measurement.

Enclosing the Pressure Sensor

My application requires me to sense pressure in air-tight tubing. My solution was to insert this sensor inside a 10 mL syringe and seal it up with epoxy such that the only opening would be the twist connector I could attach to the air line. I accomplished this by cutting the syringe with a rotary tool, removing the rubber stopper from the plunger and puncturing it so I could pass the wires through, then sealing it up as tightly as I could. I crossed my fingers and hoped it wouldn’t leak as I mixed-up some epoxy and poured it in. After an hour of setting time, I was delighted to learn that it sealed air tight! I could now attach needles and tubes with the screw connector, or leave it disconnected to measure atmospheric pressure.

Sniffing I2C with a Logic Analyzer

Right off the bat my Bus Pirate could pull sensor data but the C code I wrote running on a microcontroller could not. What gives? Was the sensor hooked up wrong? Was the microcontroller sending the wrong commands? Were the commands not being read by the microcontroller properly? Were the messages not being transmitted to the LCD display properly? There are so many points for failure and such limited hardware debugging (I’m not using JTAG) that my first go-to was my logic analyzer. As you can probably tell by the video I don’t use this thing too often, but good gosh when I do it usually saves me hours of head scratching.

logic analyzer sniffing functional I2C between Bus Pirate and sensor board

In this case, I immediately saw that the I2C lines were always low (!) and realized that the problem was my reliance on microcontroller pull-up resistors to keep those lines continuously high. That was a rookie mistake. I guess I could have seen this with an oscilloscope, but at the time I hooked it up I thought it was a protocol issue and not a dumb hardware issue. I slapped on a few 10K resistors to the VCC line and it worked immediately. Regardless, it was nice to have the capability. See the video for details.

Building the Enclosure

I still can’t get over how good the silver aluminium looks against the black laser-cut display bezel in combination with the dark backbit LCD display. I couldn’t have done this without the LCD bezels I just found being sold on Tindie! Mounting character LCD displays on metal or plastic enclosures is a chore and usually looks awful. I cringe at some of my old projects which have displays loosely meshed with square cut-outs. My square holes look nicer now that I use a hand nibbler tool, but there’s just no way that I know of to make an LCD display look good in a square cut-out without a good bezel. Another advantage of a large bezel is you don’t have to make a perfectly square cut-out, since it will all get covered-up anyway!

I then proceeded to epoxy the connectors I wanted (USB and Ethernet) and drill holes for the PCB mount. I added the microcontroller (ATMega328) and the circuit is so simple I’m not even going to display it here. If you’re really interested, check out the video. My logic is that a 5V noisy power supply is fine since all we are doing is simple, slow, digital signaling, and that the sensitive stuff (analog pressure/temperature sensing) is done on a board which already has a linear regulator on it presumably filtering-out almost all of the supply line noise. Plus, my application is such that 0.1 PSI is good enough and measuring it to a ten-thousandth of a PSI is quite overkill and I didn’t even end-up displaying the last digit of precision.

I used CAT5 to carry I2C, which I understand is a bit iffy. I2C is designed to be sent over small distances (like across a circuit board), and not really for long distance transmission. That’s not to say long distance I2C isn’t possible; it just requires a few extra design considerations. The basic idea is that a long line has a lot of capacitance, so it would take a lot of current (sinking and sourcing) to rapidly pull that line fully high and fully low at the high speeds that I2C could use. The longer the cable, the greater the capacitance, and the lower speed I2C you have to use and/or higher current you need to drive it. I2C drivers exist to help with this purpose, and although I have some I found I didn’t actually need to use them. For more information, google on the topic of sending I2C over twisted pair. This Hackaday article on sending I2C over long distances is also quite nice. For the purposes of this prototype, it’s working with straight-through wiring (sensor-to-microcontroller) so let’s call it good and move on.

I had to use a slightly larger aluminum enclosure than I initially wanted because there was limited vertical space with the LCD risers as well as the risers I used for my own board. It was a tight squeeze when all was said and done, but it worked great!

Programming the Microcontroller

Let’s just say I programmed the microchip to do exactly what we did with the Bus Pirate. The code is messy as heck (and even is using two different I2C libraries to signal on the same I2C line!) but it works and the prototype is done and sealed so I don’t really have a lot of reason to fine-tune the software. The full project can be found on the GitHub page, and a few relevant lines of code are below.

Here are a few key points about the microcontroller software:

  • I added a “baseline reset” which resets the current pressure to 0.000 PSI.
  • I’m intentionally not showing full precision because I don’t need it for my application.
  • I hard-coded the calibration values in C rather than look them up each time. This is fine since this code will only run on this one microchip with this one sensor. If this were a production device, obviously they would be read on startup.
  • I am not using the formula provided in the datasheet to integrate the calibration values with temperature to calculate pressure. Instead, I came up with my own formula (essentially just Y=mX+b) which was fit to an ADC/PSI curve I plotted myself using the calibration values for this one sensor and the temperature (72F) where I know the device will be held.
  • Since I’m controlling for temperature and hard-coded my calibration values, I can get good enough precision without the need for floating point math. Adding floating point libraries to an 8-bit AVR consumes a lot of memory and can be slow. However, in a production unit this would probably be a must.
  • Adding logging / PC connectivity would be really easy since there’s already a USB connection there! In this circuit I’m just using it for the +5V power, but there’s no reason we couldn’t attach to the data lines and send our temperature and pressure readings via USB. The easiest way to do this would be by adding an FTDI TTL serial USB adapter such as the FT232 or its breakout board. The microcontroller already has TTL USART capability so it would only be a few extra lines of code.

Code to pull 16-bit calibration values from program memory:

volatile unsigned int prog[8]; // holds calibration values
uint8_t data[2];
char i,j;
for (i=0;i<8;i++){
	data[0]=160+i*2; // addresses from datasheet
	i2c2_transmit(0xEE,data,1);
	i2c2_receive(0xEF,data,2);
	prog[i]=data[0];
	prog[i]*=256;
	prog[i]+=data[1]; // prog[1] will be C1
}

Code to pull a 24-bit pressure sensor reading:

uint8_t data[3];
data[0]=72; // command 72 is "set register to pressure"
i2c2_transmit(0xEE,data,1); 
_delay_ms(10); // time for conversion to complete
data[0]=0; // command 0 is "ADC read"
i2c2_transmit(0xEE,data,1); 
i2c2_receive(0xEF,data,3);
pressure=0;
pressure+=data[0]; // pull in a byte
pressure=pressure<<8; // and shift its significance
pressure+=data[1]; // pull in another byte
pressure=pressure<<8; // shit it all again
pressure+=data[2]; // add the last byte

Example Application

It’s great to have an inexpensive precision temperature and pressure sensor design ready to go for any application I want to use it for in the future. This is a project I’ve been wanting to build for a long time for an aquarium purpose involving monitoring the rate of CO2 injection through the intake valve of an aquarium filter (which I am aware is discouraged because bubbles can be rough on the impeller) as part of a DIY yeast reactor, all to encourage aquatic plant growth. Botany in a sentence: plants use light to grow and metabolize carbon dioxide (CO2) while producing oxygen (O2). By supplementing (but not saturating) the water with CO2, I get better plants. There’s also an application to monitor the positive pressure (rather than negative pressure / suction) of a microcontroller-pressure-controlled reaction chamber this way. If I code it wrong, and the pressure isn’t released, 1 gallon of sugary yeasty water will end up bursting over my living room floor. (I guess this means the pressure is on to get the design right?) Alternatively this prototype may serve a role as a pressure sensor for scientific applications such as electrophysiology to monitor fluid pressure, pipette pressure, or incubator pressure and temperature. Most importantly, this project encouraged me to check out some new hardware I am really glad I found (laser-cut character LCD bezels), read-up on I2C transmission lines and power drivers, and get experience with a new type of sensor that a lot of the Internet has not seen before.

my aquarium
Test configuration measuring filter intake suction (negative pressure)

Components Used

Tools Used

Software

Additional Resources


     

Logging I2C Data with Bus Pirate and Python

I’m working on a project which requires I measure temperature via a computer, and I accomplished this with minimal complexity using a BusPirate and LM75A I2C temperature sensor. I already had some LM75A breakout boards I got on eBay (from China) a while back. A current eBay search reveals these boards are a couple dollars with free shipping. The IC itself is available on Mouser for $0.61 each. The LM75A datasheet reveals it can be powered from 2.8V-5.5V and has a resolution of 1/8 ºC (about 1/4 ºF). I attached the device to the Bus Pirate according to the Bus Pirate I/O Pin Descriptions page (SCL->CLOCK and SDA->MOSI) and started interacting with it according to the Bus Pirate I2C page. Since Phillips developed the I2C protocol, a lot of manufacturers avoid legal trouble and call it TWI (two-wire interface).

Here I show how to pull data from this I2C device directly via a serial terminal, then show my way of automating the process with Python. Note that there are multiple python packages out there that claim to make this easy, but in my experience they are either OS-specific or no longer supported or too confusing to figure out rapidly. For these reasons, I ended up just writing a script that uses common Python libraries so nothing special has to be installed.

Reading data directly from a serial terminal

Before automating anything, I figured out what I2C address this chip was using and got some sample temperature readings directly from the serial terminal. I used RealTerm to connect to the Bus Pirate. The sequence of keystrokes I used are:

  • # – to reset the device
  • m – to enter the mode selection screen
    • 4 – to select I2C mode
    • 3 – to select 100KHz
  • W – to turn the power on
  • P – to enable pull-up resistors
  • (1) – to scan I2C devices
    • this showed the device listening on 0x91
  • [0x91 r:2] – to read 2 bytes from I2C address 0x91
    • this showed bytes like 0x1D and 0x20
    • 0x1D20 in decimal is 7456
    • according to datasheet, must divide by 2^8 (256)
    • 7456/256 = 29.125 C = 84.425 F

Automating Temperature Reads with Python

There should be an easy way to capture this data from Python. The Bus Pirate website even has a page showing how to read data from LM75, but it uses a pyBusPirateLite python package which has to be manually installed (it doesn’t seem to be listed in pypi). Furthermore, they only have a screenshot of a partial code example (nothing I can copy or paste) and their link to the original article is broken. I found a cool pypy-indexed python module pyElectronics which should allow easy reading/writing from I2C devices via BusPirate and Raspberry Pi. However, it crashed immediately on my windows system due to attempting to load Linux-only python modules. I improved the code and issued a pull request, but I can’t encourage use of this package at this time if you intend to log data in Windows. Therefore, I’m keeping it simple and using a self-contained script to interact with the Bus Pirate, make temperature reads, and graph the data over time. You can code-in more advanced features as needed. The graphical output of my script shows what happens when I breathe on the sensor (raising the temperature), then what happens when I cool it (by placing a TV dinner on top of it for a minute). Below is the code used to set up the Bus Pirate to log and graph temperature data. It’s not fast, but for temperature readings it doesn’t have to be! It captures about 10 reads a second, and the rate-limiting step is the timeout value which is currently set to 0.1 sec.

NOTE: The Bus Pirate has a convenient binary scripting mode which can speed all this up. I’m not using that mode in this script, simply because I’m trying to closely mirror the functionality of directly typing things into the serial console.

import serial
import matplotlib.pyplot as plt

BUSPIRATE_PORT = 'com3' #customize this! Find it in device manager.

def send(ser,cmd,silent=False):
    """
    send the command and listen to the response.
    returns a list of the returned lines. 
    The first item is always the command sent.
    """
    ser.write(str(cmd+'\n').encode('ascii')) # send our command
    lines=[]
    for line in ser.readlines(): # while there's a response
        lines.append(line.decode('utf-8').strip())
    if not silent:
        print("\n".join(lines))
        print('-'*60)
    return lines

def getTemp(ser,address='0x91',silent=True,fahrenheit=False):
    """return the temperature read from an LM75"""
    unit=" F" if fahrenheit else " C"
    lines=send(ser,'[%s r:2]'%address,silent=silent) # read two bytes
    for line in lines:
        if line.startswith("READ:"):
            line=line.split(" ",1)[1].replace("ACK",'')
            while "  " in line:
                line=" "+line.strip().replace("  "," ")
            line=line.split(" 0x")
            val=int("".join(line),16)
            # conversion to C according to the datasheet
            if val < 2**15:
                val = val/2**8
            else:
                val =  (val-2**16)/2**8
            if fahrenheit:
                val=val*9/5+32
            print("%.03f"%val+unit)
            return val
    
    
# the speed of sequential commands is determined by this timeout
ser=serial.Serial(BUSPIRATE_PORT, 115200, timeout=.1)

# have a clean starting point
send(ser,'#',silent=True) # reset bus pirate (slow, maybe not needed)
#send(ser,'v') # show current voltages

# set mode to I2C
send(ser,'m',silent=True) # change mode (goal is to get away from HiZ)
send(ser,'4',silent=True) # mode 4 is I2C
send(ser,'3',silent=True) # 100KHz
send(ser,'W',silent=True) # turn power supply to ON. Lowercase w for OFF.
send(ser,'P',silent=True) # enable pull-up resistors
send(ser,'(1)') # scan I2C devices. Returns "0x90(0x48 W) 0x91(0x48 R)"

data=[]
try:
    print("reading data until CTRL+C is pressed...")
    while True:
        data.append(getTemp(ser,fahrenheit=True))
except:
    print("exception broke continuous reading.")
    print("read %d data points"%len(data))

ser.close() # disconnect so we can access it from another app

plt.figure(figsize=(6,4))
plt.grid()
plt.plot(data,'.-',alpha=.5)
plt.title("LM75 data from Bus Pirate")
plt.ylabel("temperature")
plt.xlabel("number of reads")
plt.show()

print("disconnected!") # let the user know we're done.

Experiment: Measuring Heater Efficacy

This project now now ready for an actual application test. I made a simple heater circuit which could be driven by an analog input, PWM, or digital ON/OFF. Powered from 12V it can pass 80 mA to produce up to 1W of heat. This may dissipate up to 250 mW of heat in the transistor if partially driven, so keep this in mind if an analog signal drive is used (i.e., thermistor / op-amp circuit). Anyhow, I soldered this up with SMT components on a copper-clad PCB with slots drilled on it and decided to give it a go. It’s screwed tightly to the temperature sensor board, but nothing special was done to ensure additional thermal conductivity. This is a pretty crude test.

I ran an experiment to compare open-air heating/cooling vs. igloo conditions, as well as low vs. high heater drive conditions. The graph below shows these results. The “heating” ranges are indicated by shaded areas. The exposed condition is when the device is sitting on the desk without any insulation. A 47k resistor is used to drive the base of the transistor (producing less than maximal heating). I then repeated the same thing after the device was moved inside the igloo. I killed the heater power when it reached the same peak temperature as the first time, noticing that it took less time to reach this temperature. Finally, I used a 1k resistor on the base of the transistor and got near-peak heating power (about 1W). This resulted in faster heating and a higher maximum temperature. If I clean this enclosure up a bit, this will be a nice way to test software-based PID temperature control with slow PWM driving the base of the transistor.

Code to create file logging (csv data with timestamps and temperatures) and produce plots lives in the ‘file logging’ folder of the Bus Pirate LM75A project on the GitHub page.

Experiment: Challenging LM7805 Thermal Shutdown

The ubiquitous LM7805 linear voltage regulator offers internal current limiting (1.5A) and thermal shutdown. I’ve wondered for a long time if I could use this element as a heater. It’s TO-220 package is quite convenient to mount onto enclosures. To see how quickly it heats up and what temperature it rests at, screwed a LM7805 directly to the LM75A breakout board (with a dab of thermal compound). I soldered the output pin to ground (!!!) and recorded temperature while it was plugged in.

Power (12V) was applied to the LM7805 over the red-shaded region. It looks like it took about 2 minutes to reach maximum temperature, and settled around 225F. After disconnecting power, it settled back to room temperature after about 5 minutes. I’m curious if this type of power dissipation is sustainable long term…

Update: Reading LM75A values directly into an AVR

This topic probably doesn’t belong inside this post, but it didn’t fit anywhere else and I don’t want to make it its own post. Now that I have this I2C sensor mounted where I want it, I want a microcontroller to read its value and send it (along with some other data) via serial USART to an FT232 (USB serial adapter). Ultimately I want to take advantage of its comparator thermostat function so I can have a USB-interfaced PC-controllable heater with multiple LM75A ICs providing temperature readings at different sites in my project. To do this, I had to write code to interface my microcontroller to the LM75A. I am using an ATMega328 (ATMega328P) with AVR-GCC (not Arduino). Although there are multiple LM75A senor libraries for Arduino [link] [link] [link] I couldn’t find any examples which didn’t rely on Arduino libraries. I ended up writing functions around g4lvanix’s L2C-master-lib.

Here’s a relevant code snippit. See the full code (with compile notes) on this GitHub page:

uint8_t data[2]; // prepare variable to hold sensor data
uint8_t address=0x91; // this is the i2c address of the sensor
i2c_receive(address,data,2); // read and store two bytes
temperature=(data[0]*256+data[1])/32; // convert two bytes to temperature

 

 

This project lives on my growing GitHub page for microcontroller projects:

https://github.com/swharden/AVR-projects/

 


     

1 Rotary Encoder, 3 Pins, 6 Inputs

Rotary encoders are a convenient way to add complex input functionality to small hardware projects with a single component. Rotary encoders (sometimes called shaft encoders, or rotary shaft encoders) can spin infinitely in both directions and many of them can be pressed like a button. The volume knob on your car radio is probably a rotary encoder.

With a single component and 3 microcontroller pins I can get six types of user input: turn right, turn left, press-and-turn right, press-and-turn left, press and release,  and press and hold. Let’s pretend “press and hold and turn” is not a thing…

A few years ago I [posted a video] on YouTube discussing how rotary shaft encoders work and how to interface them with microcontrollers. Although I’m happy it has over 13,000 views, I’m disappointed I never posted the code or schematics on my website (despite the fact I said on the video I would). A few years later I couldn’t find the original code anymore, and now that I’m working on a project using these devices I decided to document a simple case usage of this component. This post is intended to be a resource for future me just as much as it is anyone who finds it via Google or YouTube. This project will permanently live in a “rotary encoder” folder of my AVR projects GitHub page: AVR-projects. For good measure, I made a follow-up YouTube video which describes a more simple rotary encoder example and that has working links to this code.

At about $.50 each, rotary encoders are certainly more expensive than other switches (such as momentary switches). A quick eBay search reveals these components can be purchased from china in packs of 10 for $3.99 with free shipping. On Mouser similar components are about $0.80 individually, cut below $0.50 in quantities of 200. The depressible kind have two pins which are shorted when the button is pressed. The rotary part has 3 pins, which are all open in the normal state. Assuming the center pin is grounded, spinning the knob in one direction or the other will temporarily short both of the other pins to ground, but slightly staggered from each other. The order of this stagger indicates which direction the encoder was rotated.

I typically pull these all high through 10k series resistors (debounced with a 0.1uF capacitor to ground to reduce accidental readings) and sense their state directly with a microcontroller. Although capacitors were placed where they are to facilitate a rapid fall time and slower rise time, their ultimate goal is high-speed integration of voltage on the line as a decoupling capacitor for potential RF noise which may otherwise get into the line. Extra hardware debouching could be achieved by adding an additional series resistor immediately before the rotary encoder switch. For my simple application, I feel okay omitting these. If you want to be really thorough, you may benefit from adding a Schmidt trigger between the output and the microcontroller as well. Note that I can easily applying time-dependent debouncing via software as well.

Quick Code Notes

Setting-up PWM on ATTiny2313

I chose to use the 16-bit Timer/Counter to generate the PWM. 16-bits of duty control feels excessive for controlling an LED brightness, but my ultimate application will use a rotary encoder to finely and coarsely adjust a radio frequency, so there is some advantage to having this fine level of control. To round things out to a simple value, I’m capping the duty at 10,000 rather than the full 65,535. This way I can set the duty to 50% easily by setting OCR1A to 5,000. Similarly, spinning left/right can adjust duty by 100, and push-and-turn can adjust by 1,000.

void setupPWM_16bit(){
    DDRB|=(1<<PB3); // enable 16-bit PWM output on PB3
	TCCR1A|=(1<<COM1A1); // Clear OC1A/OC1B on Compare Match
	TCCR1B|=(1<<WGM13); // enable "PWM, phase and frequency correct"
	TCCR1B|=(1<<CS10); // enable output with the fastest clock (no prescaling)
	ICR1=10000; // set the top value (could be up to 2^16)
	OCR1A=5000; // set PWM pulse width (starts at 50% duty)
}

Simple (spin only) Rotary Encoder Polling

void poll_encoder_v1(){
	// polls for turns only
	if (~PINB&(1<<PB2)) {
		if (~PINB&(1<<PB1)){
			// left turn
			duty_decrease(100);
		} else {
			// right turn
			duty_increase(100);
		}			
		_delay_ms(2); // force a little down time before continuing 
		while (~PINB&(1<<PB2)){} // wait until R1 comes back high
	}
}

Simple (spin only) Rotary Encoder Polling

void poll_encoder_v2(){
	// polls for turns as well as push+turns
	if (~PINB&(1<<PB2)) {
		if (~PINB&(1<<PB1)){
			if (PINB&(1<<PB0)){
				// left turn
				duty_decrease(100);
			} else {
				// left press and turn
				duty_decrease(1000);
			}
		} else {
			if (PINB&(1<<PB0)){
				// right turn
				duty_increase(100);
			} else {
				// right press and turn
				duty_increase(1000);
			}
		}			
		_delay_ms(2); // force a little down time before continuing 
		while (~PINB&(1<<PB2)){} // wait until R1 comes back high
	}
}

What about an interrupt-based method?

A good compromise between continuous polling and reading pins only when we need to is to take advantage of the pin change interrupts. Briefly, we import avr/interrupt.h, set GIMSK, EIFR, and PCMSK (definitely read the datasheet) to trigger a hardware interrupt when a pin state change is detected on any of the 3 inputs. Then we run sei(); to enable global interrupts, and our functionality is identical without having to continuously call our polling function!

// run this only when pin state changes
ISR(PCINT_vect){poll_encoder_v2();}

int main(void){
	setupPWM_16bit();
	
	// set up pin change interrupts
	GIMSK=(1<<PCIE); // Pin Change Interrupt Enable 
	EIFR=(1<<PCIF); // Pin Change Interrupt Flag
	PCMSK=(1<<PCINT1)|(1<<PCINT2)|(1<<PCINT3); // watch these pins
	sei(); // enable global interrupts
	
	for(;;){} //forever
}

All code for this project is available on the GitHub:

https://github.com/swharden/AVR-projects

 


     

Raspberry Pi RF Frequency Counter

I build a lot of RF circuits, and often it’s convenient to measure and log frequency with a computer. Previously I’ve built standalone frequency counters, frequency counters with a PC interface, and even hacked a classic frequency counter to add USB interface (twice, actually). My latest device uses only 2 microchips to provide a Raspberry Pi with RF frequency measurement capabilities. The RF signal clocks a 32-bit counter SN74LV8154 ($1.04 on Mouser) connected to a 16-bit IO expander MCP23017 ($1.26 on Mouser) accessable to the Raspberry Pi (via I²C) to provide real-time frequency measurements from a python script for $2.30 in components! Well, plus the cost of the Raspberry Pi. All files for this project are on my GitHub page.

img_8773

The entire circuit is only two microchips! I have a few passives to clean up the RF signal (the RF input is loaded with a 1k resistor to ground, decoupled through a series 100 nF capacitor, and balanced at VCC/2 through a voltage divider of two 47k resistors), but if the measured signal is already a strong square wave they could be omitted. The circuit requires a gate pulse which typically will be 1 pulse per second (1PPS) and can be generated by dividing-down a 32.768kHz oscillator, a spare pin on a microcontroller, a fancy 1PPS time reference, or like in my case a GPS module (Neo-6M) with 1PPS output to provide an extremely accurate gate.

schem

The connections are intuitive! The I2C address is 0x20 when A0, A1, and A2 are grounded. GPB(1-4) control the register select of the counter, and GPA(0-7) reads each bit of the selected register. The whole thing is controlled from Python, but could be trivially written in any language.

img_8777

Here’s a quick summary describing how the code works: First I send bytes to address 0 and 1 to set all pins of GPIO A as inputs, and GPIO B as outputs. Note that only 4 of 8 pins are used for the output, so technically 4 extra pins could be used for things like blinking LEDs or controlling other devices. I then set the register select pins by sending a value to 0x13 (GPIO B), and read the entire GPIO A bus (INTCAPB, 0x18). For address details, consult the datasheet. I do this 4 times (1 for each byte of the 32-bit counter), do a little math to turn it into a frequency value, and compare the current value with the last value and take the difference to display as the measured frequency.

screenshot

An advantage of this continuously running mode is that no clock cycles are lost, so a gate which accidentally fires a bit early due to jitter and cuts-off a cycle will compensate for it on a subsequent read. This is shown above, as a very stable 10MHz frequency reference is measured with this method. A “slow” 1PPS clock tick causes a reading slightly higher, compensated-for by the next reading being slightly lower. In this way, clock sources which are extremely accurate but suffer from low precision (like GPS time sources) are able to maximize the long-term measurement of frequency. Combining this frequency measurement technique with the ability to generate an analog voltage with a Raspberry Pi will allow me to perform some interesting experiments with a voltage controlled crystal oscillator.

Useful Links:

 


     

Generating Analog Voltage with Raspberry Pi

I recently had the need to generate analog voltages from the Raspberry PI, which has rich GPIO digital outputs but no analog outputs. I looked into the RPi.GPIO project which can create PWM (which I wanted to smooth using a low pass filter to create the analog voltage), but its output on the oscilloscope looked terrible! It stuttered all over the place, likely because the duty is continuously under software control. I ended up solving my problem with a MCP4921 12-bit DAC chip (about $1.50 on eBay). It’s controlled via SPI, and although I could have written a python program to bit-bang its protocol with RPi.GPIO I realized I could write directly to the Raspberry Pi SPI device using the echo command. Dividing 3.3V into 12-bits (4096) means that I can control voltage in steps of less than 1mV each, right from the bash console!

img_8696

Video: The Problem (RPi PWM jitters)

Video: My Solution (SPI DAC)

Hardware Connection

There’s very little magic in how the microchip is connected to the Pi. It’s a straight shot to its SPI bus! Here’s a quick drawing showing which pins to connect. Check your device against the Raspberry Pi GPIO pinout diagram for different devices.

img_8701-1

Controlling the DAC with a Bus Pirate

Before I used a Raspberry Pi to control the DAC chip, I tested it out with a Bus Pirate. I don’t have a lot of pictures of the project, but I have a screenshot of a serial console used to send commands to the chip. One advantage of the Bus Pirate is that I can type bytes in binary, which helps to see the individual bits. I don’t have this ability when I’m working in the bash console.

serial

I’m less familiar with the Bus Pirate, but this was a good opportunity to get to know it a little better. It look me a long time (requiring I pull out the logic analyzer) to realize that I had to manually enable/disable the chip-select line, using the “[” and “]” commands. When I set up the SPI mode (command m5) I told it to use active low, but I wasn’t sure how to reverse the active level of the chip-select commands, so I just did ]this[ instead of [this] and it worked great.

frompi

This is the signal probed when it was controlled by the Raspberry Pi, but it looked essentially identical when values were sent via the Bus Pirate. The only difference is there was an appreciable delay between the “]” commands and each of the bytes. It worked fine though.

Controlling the DAC with Console Commands

Once the hardware was configured, the software was trivial. I could control analog voltages by sending two properly-formatted bytes to the SPI hardware device. Importantly, you must use raspi-config to enable SPI.

# set analog voltage to minimum value (about 0V)
echo -ne "\x30\x00" > /dev/spidev0.0 # minimum

# set analog voltage to something a little higher
echo -ne "\x30\xAB" > /dev/spidev0.0 

# set analog voltage to maximum value (about 3.3V)
echo -ne "\x3F\xFF" > /dev/spidev0.0

Helpful Links:

 


     

Hacking a Cheap Ammeter / Voltmeter to Provide a Bluetooth PC Interface

I love analyzing data, so any time I see a cool device to measure something I usually want to save its output. I’ve lately come to enjoy the cheap panel-mount volt meters and current meters on eBay, and figured it would be cool to hack one to provide PC logging capability. After getting a few of these devices for ~$8 each on eBay and probing around, I realized they didn’t output measurement data on any of the pins (not that I really expected they would), so I coded a microcontroller to watch the lines of the multiplexed 7-segment display and figure out what the screen is displaying (an odd technique I’ve done once or twice before), then send its value to a computer using the microcontroller’s UART capabilities. Rather than interfacing a traditional serial port (using a MAX232 level converter, or even a TTL-level USB serial adapter) I decided to go full-scale-cool and make it wireless! I succeeded using a HC-06 Bluetooth serial adapter which you can find on eBay for ~$3. Although I have previously used custom software to hack the output of a TENMA multimeter to let me log voltage or current displayed on the multimeter, now I can measure current and voltage at the same time (wirelessly no less) and this is a far less expensive option than dedicating a multimeter to the task! The result is pretty cool, so I took pictures and am sharing the build log with the world.

The video summarizes the project, and the rest of this page details the build log. All of the code used to program the microcontroller (AVR-GCC), interface the device with the Bluetooth serial adapter, and plot the data (Python) is available as part of a GitHub project.

img_8436

This is what one of these modules looks like, and how it is intended to be used. One of the connectors has 3 wires (black = ground, red = power to run the display (anything up to 30V), and yellow = voltage sense wire). The other connector is thicker and is the current sense circuit. The black wire is essentially short-circuited to ground, so unfortunately this can only be used for low-side current sensing.

img_8134

The side of the display indicates which model it is. Note that if you wish to buy your own panel mount meters, look carefully at their current measuring range. Most of them measure dozens of amps with 0.1 A resolution. There are a few which only measure <1 A, but down to 0.1 mA resolution. This is what I prefer, since I rarely build equipment which draws more than 1 A.

img_8138

On the back you can see all of the important components. There’s a large current shunt resistor on the right, solder globs where the through-hole 4 character 7-segment displays fits in, and the microcontroller embedded in this device is a STM8S003 8-Bit MCU. This chip has UART, SPI, and I2C built-in, so it may be technically possible to have the chip output voltage and current digitally without the need for a man-in-the-middle chip like I’m building for this project. However, I don’t feel like reverse-engineering the hardware and software which takes measurements of voltage and current (which is an art in itself) and also figure out how to drive the display, so I’m happy continuing on developing my device as planned! I did probe all the pins just to be sure, and nothing looked like it was outputting data I could intercept. That would have been too easy!

img_8142

I snapped the device out of its plastic frame to be able to access the pins more easily.

img_8154

I then soldered-on headers to help with reverse-engineering the signals. Note that this was part of my investigation phase, and that these header pins were not needed for the end product. I have multiple panel mount ammeter / voltmeter modules on hand, so I left this one permanently “pinned” like this so I could access the pins if I needed to. A quick check with the continuity tester confirmed that every segment of every character (of both displays) is continuous (wired together).

img_8443

These headers made it easy to attach my 16-channel logic analyzer. I’m using an off-brand Saleae compatible logic analyzer. Their software is open source and very simple and easy to use. Saleae sells their official logic analyzers (which are well made and company supported) on their website, but they are expensive (although probably worth it). I purchased an eBay knock-off logic analyzer ($40) which “looks” like a Saleae device to the computer and works with the same open source software. If I were really serious about building professional products, I would certainly invest in an official Saleae product. For now, this is a good option for me and my hobby-level needs. An 8-channel version if as low as $10 on eBay, and $149 from Saleae.

img_8447

Connections are straightforward. I began probing only a single display. This is a good time to mention that an understanding of display multiplexing is critical to understanding how I’m reading this display! If you don’t know what a multiplexed display is, read up on the subject then come back here. It’s an important concept. While you’re at it, do you know what charlieplexing is?

img_8460

After gazing at the screen of squiggly lines, I was able to piece together which signals represented characters (due to their regularity) and which represented segments (which changed faster, and were more sporadic).

img_8463

I’ll be honest and say that I cheated a bit, using a very high value current limiting resistor and applying current (backwards) into the pins when the device was unplugged. I manged to illuminate individual segments of specific characters in the LCD. This supported what I recorded from the logic analyzer, and in reality could have been used to entirely determine which pins went to which characters/segments.

img_8594-1

Here’s what I came up with! It’s not that complicated: 16 pins control all the signals. The microcontroller raises all lines “high” to only one character at a time, then selectively grounds the segments (A-H) to pass current through only the LEDs intended to be illuminated. Characters are numbers and segments are letters. Note that “A” of the top display (voltage) is connected to the “A” of the second display (current), so both rows of 4 characters make 8 characters as far as the logic is concerned. The transistor isn’t really a discrete transistor, it’s probably the microcontroller sinking current. I used this diagram to conceptualize the directionality of the signals. The sample site of letters is high when a letter is illuminated, and the sample site of a segment is low when that segment is illuminated (the sample site of the segment is the base of the imaginary transistor).

logic

Knowing this, I can intentionally probe a few segments of a single character. Here is the logic analyzer output probing the second character (top), and two representative segments of that character (bottom). You can see the segments go nuts (flipping up and down) as other segments are illuminated (not shown). If you look closely at the blue annotations, you can see that each character is illuminated for about 1 ms and repeats every 13 ms.

img_8471

Now it was time to make my device! I started with a new panel meter and an empty project box. By this point I had reverse-engineered the device and concluded it would take 16 inputs of a microcontroller to read. I chose an ATMega328 which was perfect for the job (plenty of IO) although I could have used a much less powerful microcontroller if I wanted to interface an IO expander. The MCP23017 16-bit IO expander may have been perfect for the job! Anyway, I drilled a few circular holes in the back with a step-bit and cut-away a large square hole in the front with a nibbler so everything would snap-in nicely.

img_8482

I soldered wires to intercept the signal as it left the device’s microcontroller and went into the LED display.

img_8517

I then soldered the wires directly to my microcontroller. I also have an extra header available for programming (seen at the bottom) which I was able to remove once the software was complete. The red clip is clamping the serial Tx pin of the microcontroller and capturing the output into a USB serial adapter. Initially I debugged this circuit using the microcontroller’s on-board RC oscillator (1MHz) transmitting at 600 baud. I later realized that the serial bluetooth module requires 9600 baud. Although I could hack this with the internal RC clock, it was very unstable and garbage characters kept coming through. Luckily I designed around the potential of using an external crystal (pins 9 and 10 were unused) so it was an easy fix to later drop in a 11.0592 MHz crystal to allow stable transmission at 9600 baud.

img_8548

Now you can see the power regulation (LM7805) providing power to the MCU and wireless bluetooth module. Here’s the HC-06 datasheet (which is similar to HC-05) and another web page demonstrating how to use the breakout board. Also, I added a switch on the back which switches the voltage sense wire between the power supply and a sense connector which is on the back of the project box (red plastic banana jack).

img_8554

The bluetooth adapter expects 3.3V signals, so I added a quick and easy zener diode shunt regulator. I could have accomplished this by running my MCU on 3.3V (I didn’t have 3.3V regulators on hand though, and even so the module wants >3.6V to power the wireless transmitter) or perhaps a voltage divider on the output. On second thought, why did I use a zener ($!) over a resistor? Maybe my brain is stuck thinking about USB protocol standards.

img_8557

Since the chip was unstable transmitting 9600 baud, I tightened it up using a 11.0592 MHz crystal. The advantage of making your entire circuit look sketchy is that bodge jobs like this blend in perfectly and are unrecognizable!

img_8563

A quick reprogram to set the AVR fuses to switch from internal clock to external full-swing crystal was easy thanks to the female header I was able to pop out. I only recently started soldering-on headers like this with ribbon cable, but it’s my new favorite thing! It makes programming so easy.

img_8559

I packed it all in then added hot glue around the primary components (not shown). Again, if this were a production product I would have designed the hardware very differently. Since it’s a one-off job, I’m happy with it exactly like it is! It works, and it withstands bumps and shakes, so it’s good enough for me.

img_8581

I tested on a big piece of electrical equipent I’m building on the other side of the room. This device has its own 13.8V regulated power supply (and its own shelf!), so the wireless capability is fantastic to have. I just dropped this device between the power supply and the device under test. Rather than record the power supply voltage (which would always be a boring 13.8V) I decided to record a voltage test point of interest: the point just downstream of an LM7809 voltage regulator. I expected this voltage to swing wildly as current draw was high, and was very interested to know the voltage of this test point with respect to current draw. Although I have previously used custom software to hack the output of a TENMA multimeter to let me log voltage or current of this exact circuit, now I can measure both at the same time! Additionally, this is a far less expensive option than dedicating a multimeter to the task.

img_8584

I’m using RealTerm to access the serial port and log its output to a text file.

img_8590

A quick python script lets me graph the voltage/current relationship with respect to time. The (short) code to do this is on the GitHub page, and is demonstrated in the YouTube video.

demo

Here’s some data which shows the relationship between voltage (red trace) probed just downstream of an LM7809 voltage regulator and the total current draw of the system (blue trace). This data was recorded in real time, wirelessly, from across the room! This is exactly the type of interesting reading I was hoping to see.

img_8599

Now that it’s all together, I’m very happy with the result! This little device is happy serving as a simple voltage/current display (which is convenient in itself), but has the added benefit of continuously being available as a Bluetooth device. If I ever want to run an experiment to log/graph data, I just wirelessly connect to it and start recording the data. This build was a one-off device and is quite a hack (coding and construction wise). If I were interested in making a product out of this design, construction would greatly benefit from surface mount components and a PCB, and perhaps not necessitate super glue. For what it is, I’m happy how it came out, pleased to see it as a Bluetooth device I can connect to whenever I want, and I won’t tell anyone there’s super glue inside if you don’t.

Code used for this project is available at GitHub


     

Adding ADC to Microcontrollers without ADC

I recently had the need to carefully measure a voltage with a microcontroller which lacks an analog-to-digital converter (ADC), and I hacked together a quick and dirty method to do just this using a comparator, two transistors, and a few passives. The purpose of this project is to make a crystal oven controller at absolute minimal cost with minimal complexity. Absolute voltage accuracy is not of high concern (i.e., holding temperature to 50.00 C) but precision is the primary goal (i.e., hold it within 0.01 C of an arbitrary target I set somewhere around 50 C). Voltage measurement is usually a balance of a few factors: precision, accuracy, cost, simplicity, and speed. The method I demonstrate here maximizes precision and simplicity while minimizing cost. High speed operation is not of interest (1-2 measurements per second is fine), and as mentioned before accuracy is not a chief concern as long as precision is maximized. I would feel neglectful if I didn’t give a shout out to a few alternatives to this method: Using the 10-bit ADC built into most AVR microcontrollers (my go-to for ATMega328 at ATTiny85, but the ATTiny2313 doesn’t have any) often combined with an op-amp like this, using an IC like the MCP3208 8-channel 12-bit ADC (very expensive at $3.66 on mouser) are a good option, and fancy alternative dual slope methods as described in this really good youtube video and even mentioned nicely in the digital volt meter (DVM) / LCD driver ICL1706 datasheet. Those addressed, my quick and dirty idea uses only a couple cents of components and 3 pins of a microcontroller. There is much room for improvement (see my notes about a 555 timer, voltage reference, and operational amplifiers at the bottom) but this is a good minimal case starting point. This type of measurement is perfect for high precision temperature measuring using things like an LM335, LM35, or thermistor.

circuit

The concept behind this method is simple: use a current-limiting circuit to charge a capacitor at a constant rate so voltage rises linearly with time (rather than forming an exponential RC curve), and time how long that voltage takes to cross your test voltage.

A circuit which compares two voltages and outputs high when one voltage surpasses the other is called a comparator, and many microcontrollers (including ATMEL AVRs) have analog comparators built in (which compare AIN0 and AIN1, the result of which accessable by accessing the ACSR&(1<<ACO)) bit value (at least for the ATMega328, according to the datasheet). I can use the AVR’s comparator to time how long it takes a capacitor to charge to the test voltage, and output to that to the serial port. Note that I designed this entire circuit to use the most common transistor/resistors I could think of. It can be fine-tuned to increase speed or increase precision, but this is a great starting point. To generate a constant current I need a PNP transistor (I had a 2N2907 on hand) with a voltage divider on the base and a current limiting resistor above the transistor for good measure (in retrospect, with a more carefully chosen set of values this may not be needed). This is all that’s needed to charge the capacitor linearly and generate a positive ramp.

testrig-1

My test setup is a mess, but it demonstrates this idea works well, and is stable enough to run some experiments. In the frame you can see the ATMega328 microcontroller (big microchip), LM335 temperature sensor (the TO-92 closest to the MCU), a TTL FTDI serial/USB adapter (red board, top), and my USBTiny AVR programmer (blue board, right), and oscilloscope probes.

scope

To prevent this linear charger from charging forever, I make the microcontroller read the comparator which compares my test voltage with that of the ramp. If the test voltage is reached, or if the ramp reaches a cutoff voltage first (meaning the test voltage is too high to be measured), the count (time between last reset and now) is sent to the computer via serial port, and the capacitor is discharged through a PNP resistor. In the schematic, this is the “reset” pin. Note that the “measure” pin is AVR AC0, and AC1 is the test voltage. When all this is assembled, you can see how the linear ramps are created every time the reset transistor shuts off. Note that every 10th ramp is higher than the rest (shown here as the one left from center). This is because every 10th reading the data is summed and sent to the serial port, causing a little extra time before it is reset again. While the time value has been recorded of the comparator match of the test voltage and the ramp voltage, the capacitor is allowed to continue charging until the next cycle.

Interestingly, this method is largely insensitive to power supply noise. I’m using an extremely noisy environment (breadboard, DIP power regulator) but the recordings are rock solid. I suspect this is because the ramps are timed based on constant current, not abbsolute voltage, and that the ramps are fast enough to not be sensitive to slow changes in voltage. In reality, I don’t think I can adequately explain why the readings are so good when the supply is so shaky (the positive voltage rail is all over the place). It works, so I’m happy with it, and I’ll keep pushing forward.

miniterm

Lately I’ve been using RealTerm as a feature-rich alternative to HyperTerminal and a more convenient method than requiring custom python scripts be written every time I want to interact with the serial port in a way that involves debugging or logging or other advanced features. Here you can see the real time output of this device logging time to comparator match as it also logs to disk in a text file. This is great for simultaneously logging data (from RealTerm) and graphing it (from custom python scripts).

data_touch

This is what happens when I touch the temperature sensor for about 30s. I’m recording the time to voltage crossing of an LM335, so the number decreases as temperature increases. Also each data point is the average (actually the sum) of 10 points. It would be trivial to create some voltage test points, create a calibration curve, and infer the voltages involved, but this is more than enough already to prove that this method is robust and clean and precise and I couldn’t be more satisfied with the results! With a pair or capacitors and a few passives, this is totally implementable virtually anywhere. Considering my room is about 78F and my finger is about 98F, this 20F spread is about 1500 data points. That means each degree F is about 75 points, so I can resolve better 0.02 F (about 0.01 C) with this crude setup.

data_ac

If I let it run for about an hour, I catch my air conditioning coming on and off. Warmer temperature is higher voltage which means less time to charge, so the downslopes are my AC cooling my home and the up slope is my home passively warming. The fluctuations are only about 100 units which I (backwards calculate) assume are about 1-2 F.

These numbers seem so arbitrary! How can we calibrate this? This opens up a Pandora’s box of possible improvements. I’ll close by saying that this project works great exactly how it is to meet my needs. However, some modifications could be made to change the behavior of this device:

  • Slowing things down: A larger capacitor value (or higher resistor value) would increase the time or charging, lengthen the time to comparator threshold crossing, and increase precision. The readings would be slower (and more susceptible to noise), but it’s an option.
  • Self-calibration: Components (Rs and Cs) are sensitive to temperature and charge time can fluctuate with age, wear, temperature, etc. To self-calibrate with each sweep, add an additional comparator step which compares voltages between a precision voltage reference and your ramp would be a way to self-calibrate your ramp charge rate with each sweep. Optimally do this with two voltage references (3.3V and 1.8V are common) but comparing 0V to a single voltage reference would be a great step.
  • Don’t have the microcontroller gate: A 555 is perfectably capable of generating pulses to reset the ramp every so often, and frees up a pin of the microcontroller.
  • Use an op-amp for constant current charging. It seems like a lateral move, but if your deign already has an op-amp chances are there may be some unused amps, so eliminate a transistor for this purpose! Check out the constant current source section from TIs handbook on operational amplifier applications.
  • Use an op-amp for the comparator(s). The microcontroller’s comparator is handy, but if yours doesn’t have one (or you don’t feel like using one) configuring an unused op-amp stage as a comparator is a good option. The digital output could also trigger an interrupt on the digital input of a MCU pin as well!
  • Use timer and counters to measure time while using an external interrupt to gate the count. Your microcontroller’s on-board counter is likely extremely powerful so utilize it! This example doesn’t use it actually, but using it would let you count up to the CPU clock’s frequency of ticks between ramp starts and the comparator match.
  • Eliminate the microcontroller. Yeah, you heard me. If you use an op-amp keep resetting the ramps, and op-amp comparators to generate digital outputs of threshold crossings, you can use a standard counter (configured to latch then clear when the reset event is engaged by the 555 which induces resetting of the ramp by draining the capacitor), just use a counter IC to capture the value. You can clock it as fast as you want! You could even have it output its value directly to LED or LCD displays. In fact, this is how some digital volt meters work without the need for a microcontroller.

All code used in this project is available on its GitHub page