I needed one of these headband magnifiers for some needlework recently so went out and bought the Jaycar version catalogue No. QM3511. It cost NZ$35.90. This model has one built-in set of lenses, with another free set that can be slid into place behind the first. There is also a 7x loupe that can be swung into position over the right eye for real close-up work. The build is a bit rough with some very sharp plastic corners and lips that leave red weals across one's forehead, but it works. A LED illuminator module slots into place on top of the headset, which uses a pair of high brightness white LED's run by two AAA cells (not included). The LED lamp was faulty and this is the bit that was interesting.
When I first switched on the LED lamp, the LED's operated for about 3 seconds then stopped. Switching on and off a few times produced a blip of light, then nothing. Leaving it for a while before trying again, I got the short burst of light again, then nothing. Since I didn't really care about the light feature, I pulled it apart to have a look, thinking there was a dodgy switch or poor battery connection. Instead I find the circuit is wired so that it could NEVER have worked.
Using a multimeter, I checked the battery voltage. When the light was off, the voltage was 2.85V; a little low considering they were two new AAA cells. When the light was switched on, the battery voltage dropped to 1.8 volts. After leaving it on for perhaps half a minute, an interesting smell became apparent, so I switched off before any smoke got out. Then I repeated the exercise while monitoring the current from the batteries. At first switch-on, the drain from the batteries was 250mA for the brief period when the LED's were lit. When the LED's stopped, the current drain went UP to 800mA. No wonder the batteries were starting to feel the pain.
Tracing out the LED driver circuit, which is on a very tiny pcb, this is how it was wired: Don't try and build this circuit; it is faulty.
Notice how the LED's are more or less directly across the supply with only the resistance of the small choke in series. I don't know why the LED's even gave a brief burst of light, because fractionally after switch-on, the NPN transistor will switch on, shorting out the LED's and placing a near short-circuit across the supply. In fact the current is only limited by the small (1 ohm) resistance of the choke, a small length of wire, and the internal resistance of the batteries. The designers of the circuit and pcb got it wrong, or perhaps they also sell AAA batteries, because this will go through batteries very rapidly without producing any light for your investment.
Some of you will ask why is there 'any' circuitry to run a simple low power LED. In fact you can put say 30 ohms in series with the paralleled LED's and it will work very well. And, indeed it does, because that is what I tried. The current drops back to a respectable 18mA and the glow from the LED's is as expected. However, my curiosity was piqued and I determined how it is meant to work.
Instead of the cathode end of the LED's connecting to the negative rail, they should be connected to the collector of the NPN transistor and the present connection from the collector to the LED anodes is removed. The circuit then works at a reasonable 20mA of current. The operation looks like some form of relaxation oscillator. The only reason I can think of for the need for a design like this, is that the duty cycle increases a little as the voltage goes down, maintaining the brightness constant as the batteries age. Scarcely necessary, one would have thought. Still, the bottom line is the designers built a faulty pc board, FORGOT TO CHECK IT, then most likely produced them in the thousands.
I am a bit wary of the quality of Jaycar's gadgets. Prior to this incident, I have bought a stud finder which is as useless as a 2-bob watch, a USB stick on which the connector fell apart and a soldering iron which issues an alarming burning smell every time I use it, as well as having a thick and inflexible power cord which makes it awkward to use anyway. I know Jaycar has thousands of gadgets on their shelves but I suspect a large percentage are sourced from China and without a full quality audit trail. However, Jaycar's electronic components are fine. I have bought many electronic components such as semiconductors, resistors, capacitors and the like over many years and have not had any quality issues with them.
There are no comments yet
Make a comment using the form on the contact page