An Update On AN8008 Multimeter

I wrote about the AN8008 multimeter.  I have been generally happy with the meter.  But more recently, I started to notice some problems: the meter sometimes does not start up correctly, I have to turn the rotary switch a few times; and more annoyingly is that the resistance measurements seem off: the shorted resistance can be 5 to 10 Ohms or more.  Initially, the meter also displays low-battery sign;  I measured the batteries; it seems that they should be OK.  But I changed batteries anyway, but the things did not improve.   I suspected that the rotary switch contacts might be the issue.  I opened it up and found solder flux residue on the PCB circular tracks that make contact with spring clip on the rotary switch.  I cleaned the board with alcohol; the flux was easy to clean.   After that the problem seems to go away; the probe short resistance is about 0.2 Ohm.  

It is interesting the flux residue that has migrated overtime since it was not a problem initially.

This type of sliding contact is cheap but unreliable.  Periodical cleaning is necessary. 

For comparison Fluke multimeters have a much more robust construction, 

More on Atomic Pi

I wrote about Atomic Pi about two years ago.  Recently, I started using it again as a desktop computer.  I like it for the quietness without a fan.  As a desktop computer, which it is not really intended for, it is under-powered.  The most severe limitation is the 2GB RAM (maximum possible for this Intel Atom processor), which is not enough to support web browsers' full functionality.   The SD card and the buit-in eMMC memory are relatively slow as a swapping device, especially writing (eMMC average  120MB/s read, 32MB/s write and 0.38ms seek).    A solid station disk on the USB 3 is a better alternative; the benchmark shows 417MB/s read, 396MB/s write and 0.24ms seek.  Initially I used a USB3 hub, but the disk is not working properly with the hub.  So the USB3 solid state disk is plugged into directly the lone USB3 port.  We still need a way to connect other USB devices, such as a keyboard and a mouse (although a Bluetooth keyboard/mouse is another option if no other USB devices are needed.). There is a USB2 port, but it is a 2mm JST connection.  So I had to make an adapter to connect with a USB hub.  The 5-pin connector pinout is GND, 5V, D-, D+, GND; I wired it a USB A female connector on a small proto board with a ribbon cable, with the shield grounded.  At least 2GB swap space is recommended.   Now it is usable as a desktop device.   I installed Lubuntu, the lightweight version of Ubuntu; it works adequately with multiple applications running.  I do not think Windows can run adequately.

Another device to explore is the XMOS processor that is used as an audio processor.  It appears programmable.  We'll see if we can do something with it next time.

Wireless Router

I have a couple of old routers that have been replaced by more advanced ones.  So it is time to hack them for other use.  A wireless router is an embedded computer with wireless connection; so it is good for IoT applications.  A lot of old inexpensive routers have 32MB RAM and 4MB Flash, which is rather limited for more advanced router application.  OpenWRT no longer supports this type routers.  It'll be good to branch off OpenWRT to support this type of routers for IoT.   We looked at one such router earlier.

Here we take a look at TP-Link TL-WR841N, Ver 9.2.  300Mbps wireless N router, with 4x 100Mbps ethernet ports.  The main components are

• Qualcomm Atheros QCA9533-AL3A SoC
• 24Kc MIPS processor with 64KB I-Cache and 32KB D-Cache at up to 650MHz
• 25MHz clock input
• 1.2V switching regulator for core voltage.  2.62V LDO for DDR1 and GPIO.
• RF 20/40MHz band 2.4GHz OFDM for 802.11b/g/n, 2x RX/TX
• Zentel A3S56D40GTP-50L 
• 256Mb (16Mx16) 2.5v DDR400 200MHz @CL3xTSOPII-66.  
• Spansion S25FL032P 
• 32-Mb 3V Flash memory QUAD I/O 80MHz clock 40MB/s.  
• BCD's AP3502E 
• 340KHz fixed frequency 2A synchronous buck converter
• 3.3V output

The PCB date code is 1434, about 2014 August time frame.  QCA9553 is more powerful and more integrated than AR9130 used on the other router.

J3 is a 4-pin UART connector,

1. TX (2.6V)
2. RX
3. GND
4. 3.3V

Note the I/Os are 2.6V.  TX has no problem driving 3.3V UART receiver; there appears to be a few resistors in series with the RX pin, so it could be driven with a 3.3V input, but it may be prudent to add another 1K resistor in series.

The last OpenWRT version for this router is openwrt-18.06.8.  We get the boot message from the UART, 

U-Boot 1.1.4 (Build from LSDK-9.5.3.16 at Nov 29 2013 - 10:46:36)

ap143 - Honey Bee 1.1

DRAM:   32 MB
Flash Manuf Id 0x1, DeviceId0 0x2, DeviceId1 0x15
Flash:  4 MB
Using default environment

## Booting image at 9f020000 ...

   Uncompressing Kernel Image ... OK

Starting kernel ...

[    0.000000] Linux version 4.9.214 (buildbot@0d27b924961c) (gcc version 7.3.0
(OpenWrt GCC 7.3.0 r7989-82fbd85747) ) #0 Thu Feb 27 21:25:59 2020
[    0.000000] bootconsole [early0] enabled
[    0.000000] CPU0 revision is: 00019374 (MIPS 24Kc)
[    0.000000] SoC: Qualcomm Atheros QCA9533 ver 1 rev 1

BusyBox v1.28.4 () built-in shell (ash)
  _______                     ________        __
 |       |.-----.-----.-----.|  |  |  |.----.|  |_
 |   -   ||  _  |  -__|     ||  |  |  ||   _||   _|
 |_______||   __|_____|__|__||________||__|  |____|
          |__| W I R E L E S S   F R E E D O M
 -----------------------------------------------------
 OpenWrt 18.06.8, r7989-82fbd85747
 ----------------------------------------------------- 

We can interact with it through the ash shell.  Next we'll try to rebuild the firmware from the source and start to tweak.

H632S-WC Weather Channel EZ Crank Radio Extreme

 

The crank is a 3-phase alternator driven by two-stage gears; the output is rectified by 6 diodes.   There is another diode between the bridge rectifiers and the output voltage.   The FM radio IC is CD9088CB, which can operate between 1.8V and 5V, drawing about 5mA.  The 8Ω 4W speak and earphone are driven by TDA2822M, a dual low-voltage power amplifier, which works down to 1.8V.  The rechargeable battery pack is a 3-cell NiMH for 3.6V.   Most other components are for the radio.   The rectified voltage is used for direct battery charging without regulation.  And the 5 parallel LEDs are powered directly with only small series resistors.  All the wire connections are fragile.   For a retail price of $15, we do not expect it to be a very robust construction.

The battery pack seems no longer can hold charge.  We'll replace it with a small 480mAh LiPo pack (1.2 x 1.0 x 0.3"), that costs about $6 and is a little smaller than the NiMH pack (1.2x1.2x0.42") .   We install a 2mm JST PH 2.0 2-Pin connector.

We bring out the voltage after the bridge rectifiers to the outside, so we can monitor the crank generator output and potentially use it to power other devices, such as other chargers.   The diode to the Lipo isolates the battery to these wires, so it is relatively safe.

When unloaded, the crank generates about 5-8V at a reasonable crank rate.  It can peak over 10V if cranked really fast.  When it is loaded with battery, the voltages does not exceed 4.5V; after diode drop, the voltage on the battery is only about 3.8V, so it is safety for the battery .  The generator can supply enough power to sustain the radio, but not the LED light.

We also connect the output to a boost regulator, that generates 5V.  We use the 5V to charge a Bluetooth earphone.   The generator puts out 4.4V/45mA, about 200mW.   The rectified output is not very stable.  If we take the output from the battery side, it's better.  If we use it to charge a smartphone, the crank cannot keep up and the internal battery drains first.

AC Termination

In the case of the RS422 circuit, the differential line impedance is around 100$\Omega$ and "Standard RS-422 drivers are guaranteed to source and sink a minimum 20 mA across a 100Ω load."  (The receiver sensitivity is 200mV.)  That's 40mW dissipated on the termination resistor, and it could be 100mW dissipation at the driver if powered from a 5V supply.  Since this power dissipation is constant regardless the data pattern, the AC termination should be considered.  The AC termination draws power only at the data transition.  How effective is the AC termination?  The AC termination is a resistor and a capacitor in series.  The resistor value matches the transmission line impedance, so about 100$\Omega$.   People seem to use a capacitor value from several hundreds picofarads to a few nanofarads.  How do you select the capacitor value?  If we look at the transmission ratio (the voltage at the termination vs the input voltage), it is $$2\frac{1+j\omega R C}{1+2j\omega RC}.$$If $\omega >> 1/RC$, the ratio is 1.  If we take $R=100$ and $C=1$nF, the frequency must be much greater than 1.6MHz.  This is much more than the usual baud rate we use.  

But the steady state analysis is not the right way to go.  The step response that we are interested in is a transient.  Consider a 1V step applied to the transmission line.  Initially, a 10mA current flows into the termination resistor and capacitor.  The voltage across the capacitor rises as it is charged; and the voltage starts overshoot and reflects back to source.  It can be shown that the reflected voltage is half of the voltage across the capacitor.   The overshoot rises as 2RC time constant and reaches the peak when the reflected voltage makes the round trip.   So the overshoot is depended on the length of the transmission line.  If the transmission line delay is 10ns (~6ft), there is about 10% overshoot with 100$\Omega$ and 1nF termination.

If the transmission line delay is 100ns, the overshoot is $1-\exp(-2\cdot100n/(2\cdot 100 \cdot 1n)) = 63\%$.

TI's AN-903 erroneously stated, ; then the overshoot would be 39% or larger.

But one thing that we have not considered is the source impedance.  The driver could have 10-20$\Omega$ output impedance or series resistors may be included.   When the source impedance is greater or equal to the termination resistance, there is no overshoot, instead there is just a step followed by exponential rise.  For this case, a small capacitor gives fast edges (necessary for high speed transmission).

Christmas LED Lights

I have a string of 60 LED Christmas lights with 6 distinct colors.  I measured the RMS current with a multimeter, and found to be 27mA, so it worked out to be about 3.2W or about 54mW each, which seems reasonable if each LED takes about 20mA.  The string appears to have 3 parallel segments, with 3 lights having 3 wire connections.  That means each LED string consumes 9mA, so the 54mW number seems high for each LED.  Four of LEDs (all blue color why?) are not working, but the rest of the string works, so the connection is maintained.  I was curious to take one off; that segment was out.  So it is the LED that provides the electrical connection.  I took it apart and measured the resistance an found it to be 90 Ohms both ways.  Is there a built-in parallel resistor for fault tolerance?   Each of these resistors would take about 10mW.  The LEDs are running directly off AC, so the current is measured as true RMS, but the voltage RMS is not known.  

Anti Sulfur Chip Resistors

 Many automotive grade chip film resistors have the  anti sulfur feature.  Bourns has a white paper, "Advantages of Sulfur-Resistant Film Resistors".  It describes in good detail the construction.   In short, sulfur based gasses (such as caused pollution) can react with Ag/AgPd resistor terminal that leads to crack the package.  A Palladium (Pd) protection layer is added to the Ag terminal to eliminate the formation of Silver Sulphide and hence increases the reliability.