The Joule Thief

I had it ready for months, but never finalized. This Xmas 2007 we needed a simple way to light the Nativity scene, so I took the chance to get it working.

I used the joule thief in its simplest form, as probably published November 1999 issue of EPE (Everyday Practical Electronics). It's a NPN transistor, one resistor, a bifilar transformer and a LED (any color). I found the diagram on the web and worked immediately.

Lighted using a dead saline battery.

Lighted using a dead saline battery. No flash.

The Nativity scene, with room lights ON.

The Nativity scene, with room lights OFF.

How did I wind the transformer? I saw many questions about it around the web. First of all, you need some luck since you're probably winding it on a core with unknown characteristics. I took some length of enameled copper wire (2 metres), bent in half and wound about 8 turns through the core. Windings are spread about on 80% of the core. The core is a ferrite bead recovered from WhoKnowsWhereLand. Try it first with a fresh battery. If it doesn't lit, add some turns and try again. If at 20 turns you still don't get light, then:

• check your LED is wired properly
• change the core

Don't forget to scrape off the insulation from the enameled wire at solder points!

Before someone asks, no, I have no spare cores to send around.

With the saline battery my joule thief produced noticeable light (in darkness) for 36 uninterrupted hours. I will measure how much current it draws at the next battery change.

Driving a LED from TTL/CMOS

It's the first thing you learn about TTL/CMOS chips: they cannot (should not) drive a LED directly. A transistor driver is needed, and is shown at the right in its simplest form.
When the gate output goes high the BJT saturates and current flows through your LED. A limiting resistor is needed to control the LED current.

The table below shows R values vs supply voltage vs LED current.

Vcc	R ohm	R_ohm	R_ohm
3,6	45	30	22,5
4,8	105	70	52,5
6	165	110	82,5
7,2	225	150	112,5
8,4	285	190	142,5
9,6	345	230	172,5
10,8	405	270	202,5
12	465	310	232,5
13,2	525	350	262,5
I_LED [A]	0,02	0,03	0,04
V_LED [V]	2	2	2
V_CE [V]	0,7	0,7	0,7

I will need these in my hi-brightness red LED optical MCW transmitter.

CCD as infrared sensors

This might be obvious to many of you, but it wasn't to me.

CCD, the digital camera sensor, is sensitive to infrared light, such as the one emitted by all kinds of remote controls (if you don't consider prehistoric ultrasound models).

Webcams use CCD as well, so here's a screenshot taken with an old one to prove that it actually works.

The IR generator is a handheld device running the pre-installed IR remote control software.
You can see the lighted IR-LED, a reflection on the left (double-bounce?) and another spot near the top border of the picture.

You can reproduce this experiment aiming a remote control to a digital camera lens. Then press a key while looking at the display.

Perhaps a simple way to explain some physics to kids?

Antenne Yagi per usi portatili - Q&A

Data la visibilita' che ha questo blog tramite alcuni motori di ricerca (e' una semplice constatazione), postero' qui in forma anonima le domande e risposte che ricevo sul mio articolo pubblicato su Radiokit Elettronica di novembre 2007.

La prima domanda riguarda il fissaggio degli elementi: c'e' un accorgimento particolare per evitare che ruotino attorno all'asse del boom?

No. L'incastro e' molto stretto, al punto che e' quasi impossibile inserire l'elemento a scatto senza farsi male, ma si deve farlo scorrere sul boom fino al punto previsto.

Avevo il problema della rotazione con i giunti a T forati, che dopo un po' perdevano di robustezza, ma con questi ad incastro gli elementi rimangono dove li metti. Il problema e' invece riuscire ad allinearli sullo stesso piano.

Bisogna tenere presente che comunque si tratta di antenne per installazioni temporanee, dove l'importante e' la rapidita' di s/montaggio e il peso. Il resto e' grasso che cola.

Eventualmente, se i propri incastri sono troppo laschi, si puo' montare l'antenna a testa in giu', con il boom in alto e gli elementi appesi come dei pipistrelli.

Multiband beam (70 MHz)

I have no experience on this matter, so this is a purely theoretical post.

If looking for a directional antenna that covers 50 MHz, 70 MHz and 144 MHz, why not consider a log-periodic? With a boom of 4.5m, max width of 3.6m and 11 elements, three bands are covered. A matching system is needed.

Another version, probably with a better impedance range but lower gain, counts 15 elements over a 5m boom.

So, there's plenty of room to play with. Just need to put to work some simulation software and then cut lots of pipes...

Antenna tuners are reversible

I'm back on posting after a few weeks without any shack or lab activity.

I had few minutes to spare, so I wanted to do a measurement on my homebrew LC switched antenna tuner. I wanted to see what kind of impedance range it can match. How? Read on...

The LC circuit is a bidirectional quadripole, so it can be used in any direction: close one end on 50 ohm and measure R-jX on the other.

I used the MFJ-259B to do measurements, so I have no idea if "X" is inductive or capacitive. I closed the antenna side on a simple 50 ohm resistor (2x100 ohm in parallel) and flipped switches all around to see what the instrument showed.

On all HF I could read low R with highX (probably the equivalent of a short whip) as well as moderate R&X (a long non resonant, non half-wave, wire). I was concerned the capacitance swing was somehow wrong, but my feelings were wrong and the tuner actually does the job.

Closed for holidays

About to take a short vacation, with lots of open projects:

• FT-817 hands free adapter
• Finishing 4m TX #1
• Rebuilding 4m TX #2 (it has a short Vcc-GND on the ICS512 adapter)
• Put into an enclosure the NiMH constant-current charger
• Adapt two relays for 7A13 Tek plugin
• Put into an enclosure the Cisco 7960 headset adapter

That's it for the circuits laying on the workbench. Pending circuits list is much longer...

Controlling canned oscillators

For a number of projects it might be interesting to control the frequency of a canned oscillator. Some of them have a control pin, others an enable pin, others... no option.

There are two possibilities, one of which should always work:

1. controlling the voltage supply
2. controlling the load capacitance

The latter works only if the module does not include an output buffer. Most recent canned oscillators should be insensitive to load C variations.

The voltage supply control apparently works on all canned oscillators. You'll need to map your oscillator behavior vs. voltage, but a few Hz shift can be achieved, say 200Hz at 30 MHz, that is 0.06%.

YMMV.