28 August 2014

Virtual Audio Cable for Android (not!)

I want to be able to listen to the receiver in my shack while doing other things around the flat. I wanted to do it with a Bluetooth headset paired with an Android smartphone, so I need a simple "virtual audio cable" app that pipes the audio from the physical micrphone over to the BT headset. The app could also provide additional functionality, like squelch or CAT control or even the ability to reply and transmit.

As of today I have not been able to locate a suitable Android app. I tried to build one myself using MIT's App Inventor, but their Bluetooth library supports data transfer only, not audio (August 2014).

I will have to use a stand-alone headphone+base. They cost about 10€ in Far East but they transmit in the FM broadcast band, which is not license-free in Italy and it is too crowded for any low-power signal.

Case closed.

20 August 2014

Solar lamp experiments

My family owns a solar lamp bought at the popular Swedish furniture shop. Given the amount of light it can produce, all night long, I had to take some measurements to understand how much power the same hardware could deliver to a stand-alone circuit.

What am I thinking of? Something like an off-grid sensor that stores data for large amounts of time before any human intervention.

One doubt was: does the cover influence the amount of light reaching solar panels? If so, how much?

The twin solar panel is used to charge a 1.2 V NiMH battery. I decided to measure the current flow through the battery. The experiment was carried on with the help of my daughter: connect crocodile clips, read the ammeter, put the cover on, ...



Discharge.
The discharge current is just above 16 mA, and the lamp with 4x 3 mm white LEDs is very bright. There is a nice step-up circuit in there. That's 19mW out. A fully charged 2000 mAh battery could keep the light on almost 100 hours.

Recharge.
Having never measured a live solar system, I did interesting discoveries.
Without the cover, in full sunshine and oriented right at the Sun, the circuit pushes 100 mA into the battery (120 mW). Put the cover and current goes down to 80 mA. I didn't expect those small solar panels could produce so much current! But that's the optimum condition and requires something to track the Sun: could it be energetically worth?

What happens on a cloudy day? Recharges at 5 mA without cover, down to 3.7 mA with the lamp cover. That's 6 mW instant power, a huge difference from a bright day!

Another interesting discovery about the lamp was that it keeps charging even if the switch is OFF, meaning that you can let the battery top-up when you don't need the light at night.


In the next episode. Considering a 70% charging efficiency for NiMH technology, how much power can the stand-alone device drain for prolonged use? Would it be worth to track the Sun?

14 August 2014

70 MHz progress - August 2014

Thanks to some amazing Es openings last July 2014, my 5W and Moxon beam allowed me to reach 7 DXCC (I, G, GM, ES, 9A, OH, YO) and 8 different squares.

Few more countries should be possible, even though I probably need to move the antenna to the balcony facing South to work EA and CT. Or build/use a second antenna and run a coax through the whole apartment.

Now that the Es peak is gone I can concentrate in working neighbouring squares, like JN34 and JN45. The only station I heard from JN45 is a big gun and we will arrange a JT65 QSO, possibly with the help of aircraft scatter, because 2x SSB was not successful.

09 August 2014

Arduino CW keyer - current consumption

Before boxing my new CW keyer based on K3NG work, I measured the current consumption since I have not been able to find it online. With an ATmega32U4 (a Leonardo clone) it drains 32 mA, regardless it is powered at 5V or through the supplied regulator (RAW pin). If the power save mode is enabled in the firmware, current in my specimen drops to 3 mA, but the keyer is not functional. I had hoped for lower current.

The embedded power LED could be removed to save few mA.

Conclusion: an on/off button must be installed.

02 August 2014

How to cut potentiometer shaft

I have always been very bad at finishing my projects with a proper enclosure. But something meant to stay like a CW keyer needs a good home. That's the case of my reproduction of the excellent K3NG Arduino keyer. The main challenge for me is drilling holes, and the keyer needs at least 6. I had an unused L-C tuner and its box was pre-drilled too!

When everything was almost inside I noticed something I had forgot: the long pot shaft! The knob on the very top doesn't look too nice, does it?!


No way I would take everything apart, and a quick web search revealed few methods to cut it out. The simplest solution was to use a pipe cutter: no electric tool! I passed over the requirement for a vice, crossed my fingers and started turning around:


At this point I held the lower part with pliers and snapped away the top length of plastic:


The final result is nice and I have a CW speed potentiometer, that is much more practical and immediate than "tap here n'there" methods.

Then another problem arose: the 3.5mm panel plug is too short for the box thickness and I cannot put the nut on!