Many people on Internet are talking and documenting "GNU radio" ability to create a SDR out of a USB DVB-T TV tuner. [I apologise for too many acronyms in one sentence.]
A 20US$ device (Ezcap EZTV668 in my case) is supposed to tune, display and decode transmissions between 65 and 1700 MHz. Sampling resolution is only 8-bit and there is no front-end filtering for this direct conversion receiver (can be added externally, of course), but this will let me have a peek at 23cm band.
Also it could be useful as a poor man's spectrum analyzer in my lab.
Who knows? Time - and experimentation - will tell (me).
Well, first it has to come from Far East.
29 May 2012
21 May 2012
Locating R1235 on FT-817
Two weeks ago I proposed a "play safe" procedure for testing the health FT-817's ACC port (Vcc line only). If R1235 is blown, what can be done? First let us find out where it is.
Remove all power sources to the FT-817, including the internal battery.
Unscrew the top cover (note, this procedure is the same of the "optional filter installation") and gently unplug the speaker cable: a shiny PCB full of SMD components is exposed. Well, that's not the correct side for R1235.
 |
| FT817ND without top cover. |
Remove the five screws that hold the board (red circles in the picture above), the flat cable near the DATA connector (yellow rectangle) and the two coax connectors (yellow circles). Now the PCB can be lifted and rotated 180° above the front panel.
Check out the video to understand where you have to look for R1235 on the bottom side of the top board.
 |
| Area of interest to locate R1235 |
A Vcc line shorted to ground though this 10 ohm resistor for a couple of seconds increased the resistance to about 7000 ohm. Voltage without load is the same of the supply, but as soon as some current is drawn the voltage reading drops.
Fix? Stay tuned.
10 May 2012
Checking FT-817 ACC port health
Probably not many FT-817 owners know that the rear ACC port is not much tolerant to misuse. This is particularly the case of the +V line protected only with a tiny 10 ohm 1/8W 1/16W resistor (R1235): if this line is shorted to ground the resistor blows and goes open circuit (I've already heard of two blown resistors).
So, even testing the health of the ACC port +V line can be dangerous. But since that line is always connected to +V of the power supply or internal battery, you may use the following safe trick.
Power off the FT817. Unplug the internal battery. Remove any external power supply but leave the power cord connected to the radio. Get hold of a 1.5V cell and find out how to use it to power the FT817 (button cells can be held in place with a clothes peg).
The RTX will obviously not power up, but an accidental short circuit on the ACC line will not damage immediately the internal 10 ohm resistor (P = V^2/R = (1.5)^2/10 = 0,225W, about 1/4W instead of 14W at 12V!).
Extract +V and GND lines from the ACC socket with two insulated wires and measure voltage across them with a voltmeter. If you read your battery's voltage out of the ACC port, then everything should be fine.
You may also try to guess the internal resistor value, in case you have doubts it was damaged before. How? Connect a 10 ohm resistor across your loose wires and measure the voltage: it should be half of the battery reading. Or use Ohm's law to do the reverse engineering.
This procedure only tests the health of the +V ACC port line, not the whole ACC socket health.
So, even testing the health of the ACC port +V line can be dangerous. But since that line is always connected to +V of the power supply or internal battery, you may use the following safe trick.
Power off the FT817. Unplug the internal battery. Remove any external power supply but leave the power cord connected to the radio. Get hold of a 1.5V cell and find out how to use it to power the FT817 (button cells can be held in place with a clothes peg).
The RTX will obviously not power up, but an accidental short circuit on the ACC line will not damage immediately the internal 10 ohm resistor (P = V^2/R = (1.5)^2/10 = 0,225W, about 1/4W instead of 14W at 12V!).
Extract +V and GND lines from the ACC socket with two insulated wires and measure voltage across them with a voltmeter. If you read your battery's voltage out of the ACC port, then everything should be fine.
You may also try to guess the internal resistor value, in case you have doubts it was damaged before. How? Connect a 10 ohm resistor across your loose wires and measure the voltage: it should be half of the battery reading. Or use Ohm's law to do the reverse engineering.
This procedure only tests the health of the +V ACC port line, not the whole ACC socket health.
11 April 2012
NiMH laptop battery from 2001
I got hold of an old laptop from 2001 with the battery dead: what a better opportunity to have a look inside? If possible I wanted to preserve the battery case as much as possible.A close inspection revealed that the battery case was composed of two parts very closely coupled together. Poking around with a small flathead screwdriver showed that the "weakest" point was the area around the battery connector. And from that point I started pushing and pulling. Few "crack" and some more "ziiipppp" were released before I could remove the lid, which was glued to elements.
The battery, being a 9.6V 4500mAh NiMH looked as shown in the picture: 8 cells in HR-4/3FAU format. All cells measure 1.265V each, which is a sign or them being still balanced somehow, and not dead. A small two-wire component was also found in close contact with two batteries. This could be a simple TC resistor or a more complex device which had also kept count of the number of recharge cycles.I recharged the battery in place and it got very hot to touch, but charging stopped normally (with the two-leaded component in close contact with two elements). Final voltage was about 11V.
Drawing some heavy current allowed to evaluate then internal resistance developed during the years. At 3A the voltage drop across battery terminals suggested an internal resistance of 3.3 ohm, that is about 0.4 ohm per cell. This is high resistance for a high capacity battery! I wonder if there is a way to reduce it with slow discharge/charge cycles?
One week after the full charge the pack still exhibits 10.6V, about where I have left it, so apparently no cell is dead or leaking current.
Besides being a good paper weight, this pack could be used to power some light-current load despite the internal resistance, with an acceptable and known voltage drop. I am thinking of some power LEDs in series. Other usage ideas welcome.
30 March 2012
FT-817 on Mac OSX Leopard, via Bluetooth
I am not a MAC user, but I wanted to test the FT-817 CAT-to-Bluetooth adapter on it.
I borrowed a PowerBook G4 with OSX 10.5.8. Then I searched for a CAT software, and the easier to install was flrig from the fldigi suite. I tried jLog but it complained that "radiocomm.jar are not properly installed" (no results on Google to help troubleshooting). And I could not compile grig either.
I attach seven screenshots of the pairing procedure and sample usage. firig was able to control my FT-817 through the serial port over bluetooth, I just had to use the proper configuration (see the 6th screenshot below).
Provided there is software for more recent MACs, the adapter should work on them too.
I borrowed a PowerBook G4 with OSX 10.5.8. Then I searched for a CAT software, and the easier to install was flrig from the fldigi suite. I tried jLog but it complained that "radiocomm.jar are not properly installed" (no results on Google to help troubleshooting). And I could not compile grig either.
I attach seven screenshots of the pairing procedure and sample usage. firig was able to control my FT-817 through the serial port over bluetooth, I just had to use the proper configuration (see the 6th screenshot below).
Provided there is software for more recent MACs, the adapter should work on them too.
 |
| Make sure to select the last option, "Any device" |
 |
| flrig in use and the configuration screen |
 |
| When you're done, disconnect the adapter |
21 March 2012
Yaesu CAT bluetooth interface - video
For the Curious and the Interested, here is a short video introducing the wireless (Bluetooth) CAT interface I made for Yaesu FT-817/FT-857&FT-897:
Contact me if you want one interface for your shack.
Contact me if you want one interface for your shack.
13 March 2012
Frequency Reader with Tuning Knob
The "Frequency Reader" is a stand-alone display with three push buttons that allows to read remotely the FT-817 display data and gives minimal user interaction ability. It is supposed to work on FT-857 and FT-897 too.
As of today, March 13th 2012, the "Frequency Reader" supports a tuning knob which lets you retune the radio remotely. The firmware has been tailored for users of microwave transverters, but nothing prevents using it for the usual V/U/HF operations. Three buttons do:
For interested builders, it should be noted that this new firmware works on a slightly different hardware circuit, on which few connections have been moved around to facilitate the design of PCB.
If a simple Frequency Reader is needed, just omit mounting the knob and step button leaving their terminals open.
Technical insight.
The Frequency Reader polls the radio every second for a frequency change, in case the operator touches the front panel dial knob. If the external knob is rotated, the VFO frequency is updated of a "step" amount every 100 ms. If the knob is rotated very fast, the step is increased accordingly in a linear fashion, allowing faster QSY's.
The knob component, a quadrature rotary encoder, must be selected with care. It should have no detents, or generate one step per detent. If it includes a pushbutton, it can be wired to any of the three required buttons mentioned above.
What is left to do.
As of today, March 13th 2012, the "Frequency Reader" supports a tuning knob which lets you retune the radio remotely. The firmware has been tailored for users of microwave transverters, but nothing prevents using it for the usual V/U/HF operations. Three buttons do:
- USB/CW mode toggle (no action if in a different mode)
- VFO A/B toggle
- tuning step selection
For interested builders, it should be noted that this new firmware works on a slightly different hardware circuit, on which few connections have been moved around to facilitate the design of PCB.
If a simple Frequency Reader is needed, just omit mounting the knob and step button leaving their terminals open.
Technical insight.
The Frequency Reader polls the radio every second for a frequency change, in case the operator touches the front panel dial knob. If the external knob is rotated, the VFO frequency is updated of a "step" amount every 100 ms. If the knob is rotated very fast, the step is increased accordingly in a linear fashion, allowing faster QSY's.
The knob component, a quadrature rotary encoder, must be selected with care. It should have no detents, or generate one step per detent. If it includes a pushbutton, it can be wired to any of the three required buttons mentioned above.
What is left to do.
- Shoot a short demo video
- Update the website with the new diagram and firmware
- Refine the firmware for general public release
07 March 2012
A coil or a lifelong supply of enameled wire
While the supercapacitor and white LED in the handshake flashlight have been used to build a proof-of-concept USB rechargeable flashlight, the coil is now calling for attention. A passive audio filter came to my mind.
On my DVM the coil measures ~22mH for 100 ohm of wire resistance. At 1kHz this means a coil Q of 1,3 (Q = 2 * pi * f * H / R).
I am not into passive audio filters but the Q sounds too low to be useful. The other option would be to use the coil as a lifelong source of enameled copper wire. According to the perceived diameter (not measured, but it is thin) I would have about 1km of wire!
On my DVM the coil measures ~22mH for 100 ohm of wire resistance. At 1kHz this means a coil Q of 1,3 (Q = 2 * pi * f * H / R).
I am not into passive audio filters but the Q sounds too low to be useful. The other option would be to use the coil as a lifelong source of enameled copper wire. According to the perceived diameter (not measured, but it is thin) I would have about 1km of wire!
Subscribe to:
Posts (Atom)
