Project Logs

Collapse

• IO Expansion

  Christoph Tack • 05/21/2020 at 14:25 • 0 comments

  There's more IO needed than the BluePill can handle.  The following user IO needs to be controlled :
  

  • 11 LEDs
  • 11 buttons
  • 1 switch
  • 1 rotary encoder

  So we need an IO expander.  The original setup used a MCP23017 to handle all of this.  It turned out to be not such a good idea.  See below.

  Other options exist:

  1. Use 2 SX1509 boards
  2. 1x SX1509 for button inputs + 1x TLC59711 or TLC5940 for LEDs (expensive)
  3. GPIO for inputs ( 3 rows + 4 columns). 1 TLC59711 for LED output

  16 channel SPI LED driver, non-dimmable : TLC5925 | STP16CPC26 | CAT4016 | MAX6969: 16 outputs, integrated current source, Digikey €1/pce.  Advantage of using these is that charlieplexing is no longer needed.
  

  • SX1509
    • More than just an I²C IO-expander
      • Keypad scan engine on Rows : IO0 - IO7, Columns :  IO8 - IO15
      • PWM dimming engine for GPIOs
    • Sparkfun
      
      • pin spacing : 0.6" x 10pin (2x12p DIP socket could be used : Digikey ED3051-5-ND)
      • I²C connection could be made through 1x5pin socket strip
    • power supply : 1V8, 3V3 (not 5V, but IO is 5V tolerant)
    • LED driving capability
      • 15mA sinking/pin
      • 8mA sourcing/pin
    • QFN package, so breakout board needed.
    • I²C
  • MCP23017
    • basic I²C IO-expander, not much extra functionality
    • AliExpress module : €0.76
    • Used by Adafruit
    • 25mA push/pull
    • I²C
    • 2 IRQ pins
    • level triggered interrupts, edge triggered interrupts (no distinction between falling and rising edge interrupts)
    • 16bit
    • 3V3, 5V compatible
    • available in SOIC28
    • This device has no internal keypad scanning engine, unlike the SX1509.  So you'll have to drive the rows & columns yourself when using a matrix keypad.  This causes a lot of I²C interrupts and overhead for the MCU.
    • The IO expander also controls the 15 LEDs in this setup using charlieplexing.  A charlieplexing library for the MCP23017 has been developed, loosely based on the Chaplex library. 
    • Key scanning combined with charlie plexing the LEDs, causes flickering on the LEDs.  Moreover the intensity of the LEDs can't be controlled.

  • MCP23S17
    • SPI version of the MCP23017
  • PCF8574
    • 8bit

  • PCF8575
    • 16bit
    • asymmetrical output driver: 1mA source, 25mA sink

  PCB Design

  The I²C expansion board can control 8LED/switches.  The LED/switch-PCBs have four LED/switches.  If the I²C expander would be incorporated on the LED/switch PCB, effectively 50% of the I²C expander would be wasted.

  All FFC connectors are the same to reduce BoM-cost and lead time.  For the 3V3/I²C-connector, this means that some pins are not really used.

• LED-switch panel

  Christoph Tack • 05/21/2020 at 12:43 • 0 comments

  Design

  The PCB has been designed using EasyEDA.  You can find the design there.
  

  The design needed to be compact, so that multiple of these boards can be connected next to each other while keeping the same spacing between the buttons.  The rightmost button can be snapped off, if needed.

  EMC precautions : resistors and capacitors for filtering, ground plane for signal return.  One third of the FFC pins is GND.  The ground plane on the rightmost switch is poorly connected to the other ground plane.  We'll see what it brings.
  

  Component choices

  Interconnection

  Requirements for the connectors : cheap, small, easy to solder.  Which brings to to the 1.0mm pitch FFC cable.

  A little search on AliExpress "ffc 14 pin", brings up a source that can deliver the cables for €0.10/pce.	The matching connectors can be had for about €0.05/pce.	For debugging, AliExpress also provides breakout boards at €0.91/pce. Remark that the breakout comes in dual row, so it's not breadboard friendly.

  Buttons

  Requirements : big buttons, should be able to handle lots of operations, visual feedback (through LED).  The selection can be found on a separate page.
  

  Bonus facts

  The I²C-expander uses matrix keyboard scanning.  This requires 8 IO lines for 15 switches.  If not enough IO lines had been available, there would be an alternative approach as described by Mosaic Industries:
  

  This technique would require only 5 IO lines for 15 switches (max. 20 switches).  The downside is that 5 diodes need to be added.  This probably won't work with an SX1509 or similar which has a built in keypad scanning engine.
  

• Baseboard

  Christoph Tack • 05/21/2020 at 12:35 • 0 comments

  Diagram

  The diagram above gives an overview of the baseboard.  The yellow background represents the baseboard and the functionality present onto it.  The other modules (speaker, rotary encoder, sensors, displays) are external modules.
  

  • Microcontroller platform selection
  • Audio module
  • User Input Control
  • Ambient light sensing
  • Movement detection sensor
  • Main display
  • Alarm time display

  Old data:
  Rev.0: IO Panel : LCDs, switches, IO expansion, LED array, light sensor all integrated on one PCB

  Verifying interfacing to peripherals

  • ☑ MP3 module
  • ☑ Serial debug port
  • ☑ Main display 32x16 LEDs
  • ☑ 2x I²C expander port : Unfortunately, both connectors should have been 14pins FFC instead of 10pins FFC now.  I solved it by cutting off three tracks of the FFC-connector.  It would have been a cleaner workaround to replace the FFC-connectors on the I²C-expansion boards by 10pin types.  But then I would also have to buy 10pin FFC-cables.
  • ☑ EEPROM
  • ☑ Rotary encoder
  • ☑ Time sync port
  • ☑ PIR sensor
  • ☑ Alarm time display
  • ☑ Light sensor
  • ☑ Sunrise simulation

• Firmware

  Christoph Tack • 01/26/2019 at 21:10 • 0 comments

  Test hardware

  This setup has the BluePill on the breadboard, connected to a Nucleo debugger (white PCB on the left).  The 7segment module on top connects to two GPIOs of the BluePill.  The BluePill receives its DCF-pulse info from a DCF-receiver on the breadboard.  That DCF-receiver is connected to a DCF-antenna, attached with rubber bands to the breadboard.

  DCF-signals are very weak.  Making breadboard connections with relatively long wires as done here, makes the BluePill a fairly good RF noise radiator.  So with the setup as shown here, it's nearly impossible to get a correct DCF-reading.  The picture was actually taken around 9PM.  The DCF-library makes a best guess of the data it receives.  There's no reliability threshold, so wrong data might be shown in the event of a useless signal.  Most of the times, even with a bad signal, the clock recovers and finds the correct time.
  

  I faced the same problems with the BluePill when trying to receive FM-RDS.  Using a Nucleo, results were a little bit better, even with the same mess of wires.

  Code can be found on Github.

  DCF signal quality for the same setup the next day, on Sunday, about the same time of day was much better.  No easy explanation about what caused such an interference on Saturday night.

  ---

  Parameter setup

  This design has quite a lot of parameters that can be set.  It's challenging to find a way to set these all up easily using only a 32x16 display.  The display can only show 10 characters at a time.

  Menu input would be limited to using a rotary encoder or a few switches.

  I have tried a few menu libraries for Arduino : Arduino Menu System, Arduino_LCD_Menu and ArduinoMenu.  They all suffer from the same problems:

  • Requiring a lot of key presses for setting up parameters
  • Difficulty of displaying the menu hierarchy in a way that is simple to grasp.  E.g.  setting the weekdays for the alarms are four levels deep in the menu.  Without a menu overview, it will be hard to keep track of what you're setting up.

  Let's take the Google clock app for Android as a reference.  This app allows to setup more or less the same parameters as listed below.  Using the app, an alarm can be setup in less than 30s.  This would be nearly impossible to do with a 32x16 LED array and four buttons.

  Solution

  Parameter setup will be done using physical switches.  A combination of DIP switches and rotary switches will be used. 

  Advantages:

  • Each parameter can be set without having to browse through the other parameters, speeding up the parameter setup process
  • Parameter status is indicated by the physical state of the switch.  This doesn't require power and doesn't cause undesirable light output at night.
  • Depending on the target audience, the switch functionality can be implemented using DIP-switches and rotary switches or with panel mount toggle switches and rotary switches that allow for older people to control the device.

  Disadvantages:

  • Adds cost, because it requires a lot of extra hardware
  • No other means for setup of parameters (e.g. through bluetooth) because the switches would no longer reflect the current settings.  A way to circumvent this, would be to replace the rotary switches by rotary encoders & 7segment displays and replacing the dip switches by momentary switches and LEDs.

  The input panel will be implemented on a separate PCB, allow for upgrade, changes in a future stage.

  Menu structure

  • Alarm 1 (Alarm 2 has the same menu structure).  The menu item is dynamic.  An icon shows if the alarm is enabled or not, followed by the alarm time
    • Activation
      • On
      • Off
    • Time
      • HH:mm
        • minutes can only be set in 5min. steps.  It allows to set up time faster.  Nobody wants to get up at...

  Read more »

• Housing

  Christoph Tack • 01/20/2019 at 11:50 • 0 comments

  Main housing

  There are multiple options for creating a housing for an electronics project.  I don't have interest in learning a decent program like Fusion360 for creating laser cut drawings.  The one-time cost of a standard box doesn't weigh up against all the hours invested in learning a new tool.  So I'll settle for a standard housing.  The OKW P/4.9-AL (drawing) might be a good choice:

  • cheap (around €10)
  • dimensions are ok (215 x 130 x 77mm).  Wide enough, so that it doesn't easily fall over.
  • front panel can be replaced by a PCB, or laser cut panel.
    • The front panel is a thin aluminium sheet.  I milled the needed slots in it using a Dremel multifunctional rotary tool, but the result isn't very clean.
    • The original front panel is too flexible now with all the cutouts.

  To be fair, it would have been better if the housing was 10mm higher and 10mm narrower.  That would have allowed for a nicer placement of the alarm LCD.
  

  Material

  Because of the sunrise simulation, the housing, or a least a major part of it must be translucent or transparent.  Translucent materials are preferred as the diffusion of the light will make the housing light up as a whole.  It's an option to use transparent materials for the housing and stick a translucent film onto it.
  

  • Acrylic : can be laser cut

  Box design

  Online file generators for laser cutting

  • T-slot Box
    
    • Easy
  • Lattice Hinges
    • Looks nicer, but will need some experimenting

  Manufacturers

  • Seeedstudio laser cutting
  • Ponoko : online laser cutting
  • Buda::lab : makerspace
    
  • Elecrow

  Antenna enclosure

  There's a chance that the electronics of the generate too much EM-noise, so that the antenna and receiver need to be placed in a separate housing.

  A suitable housing is the Hammond 1593NBK (Farnell €2.75).
  

  If you want to build a really cheap enclosure for your antenna with hot glue and some scrap piping, have a look here.

  Further reading

  Adafruit : Laser cut enclosure design

  Front panel

  This still leaves options for the front panel.  An aluminium front panel is nice, but expensive (Front Panel Designer).  An alternative would be to create the front panel instead as a (non functional) PCB.

  Open source CAM tools

  https://github.com/geobruce/FabLab/blob/master/CAM-tools-CNC.md

• Power supply

  Christoph Tack • 06/28/2018 at 11:41 • 0 comments

  Requirements

  If no sunrise LED is needed, then the electronics could be powered from a 5W USB wall charger.  The sunrise LED will probably need about an extra 10W.

  Cable

  Not all USB cables are made equal.  I read about it in the past and I ran into the same problem.  The cable I originally used, had an AWG28 for power.  I lost more than 1V over it when the light of the circuit was on.  Only 3.13V remained at the 5V signal (the power supply also has internal resistance).  This resulted in unstable behavior with the MP3 player.
  Switching it with a AWG24 cable, which had a much lower internal resistance fixed my problem.  USB Cables of AWG22 and AWG20 also exist, but these tend to be longer or much more expensive.
  

  Backup

  The common solution is to use a goldcap as back up power supply.  The limitations of goldcaps are:

  • limited energy storage capacity
  • high internal resistance
  • voltage decreases linearly with respect to the current being drawn.

  All of this makes that in case of power outage the goldcap can't be used to sound the alarm to wake the user.

  An alternative approach will be used here.  A cheap alkaline AA-cell will power a MCP1640 boost converter that will step up the voltage to the required 3V3.  As long as the normal power is supplied, the MCP1640 can be disconnected from the AA-cell.

  Filtering

  DCF-communication is susceptible to noise.  Proper filtering is needed to reduce the risk of bad signal reception.

  • ANP002: The Protection of USB 2.0 Applications
  • AliExpress 32707561851 : USB EMI filter
  • USB Hardware Design Guidelines for FTDI ICs

  Connector

  USB-C connector

  USB-C allows for a maximum power transfer of 20V/5A.  This would be more than enough for this project.

  The advantage of using a USB-C connector would be that a power adapter with USB-C could be used.  These will probably be quite standard in the near future.  At the time of writing (2018) these are still expensive.

  There are more drawbacks: to get the 20V/5A, the device delivering the power and the device sinking the power must communicate with each other using BMC-coding.  This is implemented in the Richtek RT1715 and the TI TUSB32x family.  These devices are still rather expensive.  The SMD packages are small and hard to solder by hand.

  Using this connector would be a typical case of over engineering.  There's no real advantage is using USB-C only for powering a device.  The exception could be for powering an electric razor.  You could use your phone charger to power it.  It saves you one power adapter in your luggage.

  DC jack

  Cheap, readily available, reliable...  The drawback is that there's no guarantee about the voltage level and polarity.  The user can connect whatever (s)he likes.  It's up to the powered device to cope with this.

• Sunrise LED

  Christoph Tack • 06/24/2018 at 16:12 • 0 comments

  Background

  A gradually brighter light can mimic the sun rise.  It will allow for gentle wake-up in the morning.  

  Experience with the Philips HF3463 showed me that you shouldn't count on it to get up on time.  I sometimes find myself waking up by the alarm sound and facing the alarm clock in full brightness.  When you're sleeping on your side and the light is behind you, it won't wake you either.

  LED driver modules

  The plan is to power the clock from a 5V USB power adapter.  The LEDs require about 12V/500mA.  So we need a boosting LED module, not a buck module.

  If we decide to make the module ourselves, we need to select an IC:

  • easy to solder : SOT23 is ok, QFN isn't
  • over voltage protection (when LED is not connected or broken).  Proper solutions directly sense the output voltage using an extra pin.  This brings the minimum number of pins for the IC to 6 : gnd, vin, vout, sw, fb, en.
  • VIN min < 5V
  • VIN max > 5V
  • VOUT min > 12V
  • IOUT > 1A (we need 500mA).  Don't confuse this with the switch current.  The input current of a boost converter is essentially the inductor current.  When the boost converter switch is conducting, the input current is equal to the switch current.  If we need 6W at the output, we need at least 8W at the input.  The switch current should be at least 2A to be safe.
  • low power.  Remark that the output LEDs continue to draw current when the LED module is in shutdown.  External circuitry is needed to fully shut down the LED module.
  • dimming capability (analog & digital dimming are explained in the MP3202 datasheet)

  Digikey selection (for 1A switch current): RT4533GJ6 and MP3202 (pin compatible), TP61165 (not pin compatible)

  In the hindsight, the LM3410 would have been a better option, because it allows for larger drive currents.
  

  Revision 1.0

  Details about this circuit can be found on EasyEDA.

  The circuitry above has some issues:

  • Thick wires must be used for VCC & GND because they have to carry up 1.1A.  These wires will probably too thick to fit in the JST-XH.  For debug, I soldered thicker wires on bottom side.
  • The current is set for 500mA, but only about 400mA is being drawn.  The feedback voltage never reached 200mV.  The 0.4ohm sense resistor combination had to be replaced by a 0.5ohm (2x 1ohm) resistor combination.
  • Value of the resistors for analog dimming are too big.  As the PCB is quite small, they pick up some noise.  Making them a factor of 10 smaller improves it a lot.

  The RT4533 announces itself as boost converter for 10W LEDs 10 WLEDs.  It's important where you put the space.  Richtek means you can drive 10 white LEDs.  It's easy to misread it as : "RT4533 can drive 10W LEDs".
  

  It may well be possible to drive LEDs, but not at 10W.  Let me explain:

  1. The maximum input voltage for the RT4533 is 5.5V
  2. The maximum switch current is 1.2A.  For a boost regulator, this is also the maximum current through the inductor.  The current through the inductor is essentially the input current to the circuit.
  3. So maximum input power is 5.5 * 1.2 = 6.6W
  4. Let's be optimistic and assume the output power is 80% of the input power : 5.3W, which is far from the advertized 10W.

  Measurement results

  The output current drops linearly with the control voltage, as expected.
  

  Efficiency is around 80%, as shown on the RT4533 datasheet.

  The module already saw a second use as a replacement driver for lighting the christmas tree. The original driver was specified for 32V/100mA. Setting the two sense resistors to 3.9ohm and you're done.  Power draw on the 5V power supply was about 800mA.

  Dimming the light using PWM

  Earlier experiments with dimming LEDs showed me that 8bits is not enough to generate smooth transitions. ...

  Read more »

• Inspirational clocks

  Christoph Tack • 05/10/2018 at 20:30 • 0 comments

  LED animations are nice, but they shouldn't stand in the way of functionality.  The DOTKLOK has very long time transition animations.  This will get annoying real soon.

  Some of the clocks show the time up to second resolution.  For a bed side alarm clock, I don't see a point in doing that.  Besides, how accurate is the clock anyway?  If the clock is not regularly synced to some time source, the shown seconds will be incorrect.  Second resolution is useful in timers (for cooking etc.), but an alarm clock is not a timer.
  

  O-Led Morphing clock

  Nice animations
  

  link

  Harifun's Morphing Digital Clock

  Nice animation, (too) big LED array, too bright, blue color will keep you awake
  

  Harifun's Morphing Digital Clock

  Craig Bonsignore's Open Clock Project

  Nice clock, but without a manual, you can't properly operate it.  Sometimes digits are green and sometimes red.
  

  Craig Bonsignore's Open Clock Project, also on Hackaday.io

  Xronos Clock

  Xronos Clock
  The idea of the arcade buttons is good, but the menu navigation is quite unusual.  The black button cycles through the menu items of the top hierarchy level.  The red button is used to go from top hierarchy to the level second level.  Then it cycles through the menu items of that second level.  The white button finally is used to change user settings.  As there are only three buttons, you can only cycle in one direction through the items.  For setting the year or the minutes of the hour, that's quite annoying.  A fourth button would have made menu navigation a lot more intuitive: up, down, select & back.

  See how long it takes and how many key presses are needed to set it all up?

  Nice housing
  

  Matrix Clock

  Nice animations
  

  Matrix Clock (technical data can be downloaded here)

  DOTKLOK

  Nice animations
  

  DOTKLOK

  Alpha Clock Five

  • Simple housing
  • Expensive LED display.
  • Acrylic housing (laser cut)

  Alpha Clock Five

  3-Way Display Alarm with 2.2" TFT

  • Has a brightness setting for day, night and allows setting brightness level for day/night transition.
  • It's personal taste, but :
    • I dislike the use of a different color for each digit.
    • Why design & 3D-print a RECTANGULAR housing?  Choose a COTS-housing, design your electronics to fit that housing, make cutouts in the housing where needed.

View all 8 project logs