I just can’t help myself, so I grabbed another kit off the pile and decided to build it. This is yet another “cheap” kit from eBay intended for educational use. This particular kit is an EQKit HBL-22 “breathing” LED heart lamp kit with green LEDs. There are a few variations of “breathing” LED projects with similar setup, but this one appeared the most professional, at a cost of just AU$1.75 posted.
Consistent with other Chinese kits, this came inside a zip lock bag with no instructions. A green dot sticker is applied to the outside to confirm the colour of the LEDs included in the kit.
The PCB is a fibreglass type single-sided board with white silkscreening on the front and green solder mask and white silkscreen on the rear. Values are clearly denoted, as is the polarity. The text code on the rear is 18122F_P228-171227, suggesting this one was made 27th December 2017. The board is lovely in its finish with a smooth glossy front, and the shape is very distinctive and well cut.
The other components supplied include the LEDs, which are a less-common frosted variety for better appearance, resistors, SS8050 transistor, LM358 dual-opamp, trimpot, switch and capacitor. No connector for power is supplied, although I did have some header pins left from another project which I used. There is no IC socket either, which is probably recommended for beginners in case they get it the wrong way around or overheat the chip.
Construction and Testing
The project requires the following number of joints to be made:
22 LEDs x 2 joints 1 Trimpot x 3 joints 1 Capacitor x 2 joints 1 Transistor x 3 joints 6 Resistors x 2 joints 1 IC x 8 joints 1 Switch x 6 joints 1 Power x 2 joints TOTAL - 80 joints
Construction was straightforward and relatively trouble free, although slightly tedious due to the repetitive nature of LED installation, taking approximately 35 minutes. Many of the pads were fairly small, making it a little more difficult to make the joints, although being tinned did help.
The small joints have an added side effect of making accidental solder bridges nearly impossible with the solder resist, so I suspect this is a “feature”. The additional copper area in some places may help prevent lifting of traces, but thermal bridges have been placed to avoid sucking too much heat from the iron.
Some spare LEDs were included, a thoughtful addition. The effect is a green heart that increases and decreases in brightness in a cyclical fashion, with speed/depth adjusted by the trimpot VR1. The diffused LEDs make it quite appealing and less glary to look at.
I traced out the schematic from the board by overlaying a flipped image of the underside with the silkscreen. Because of the sophisticated design, it was a little more difficult to trace out the circuit, but you appreciate the clever routing a little more once you try to trace it.
Because of the sophistication of the circuit and my curiousity about how it worked, I decided that it would be best to draw it in a simulator rather than by hand. For this, I used Analog Devices (formerly Linear Technologies) LTspice. A few component substitutions were made to use the inbuilt SPICE models but this shouldn’t affect the results too much. If you want to try simulating it yourself, you can download my model here.
The dual opamp is configured as an integrator (top) and Schmitt trigger (bottom), with R1/R2 forming a voltage divider that produces a virtual ground at around half the input voltage. The transistor is used to drive the LEDs (of which there are 22 in parallel, only one has been modelled).
As expected, the voltage into the base of the transistor and the current through the LED varies cyclically … hence the breathing effect. The speed and depth of modulation is dependent on the setting of VR1.
In this graph, the green trace shows the voltage on the capacitor (referenced to 0v, whereas it is actually held with the other leg at around virtual ground 2.5v), blue trace represents the output of the Schmitt trigger and red trace represents the LED current. With the Schmitt trigger off, the capacitor charges until it reaches the trigger point and the LED brightness increases. Once the Schmitt trigger fires, the capacitor discharges until it reaches the opposite trigger point, resulting in the LED brightness decreasing.
That being said, I still don’t think I completely understand the finer points of the circuit, but it was some work to even get this far after many years of not actually using a SPICE simulator.
This kit was actually quite well put together, even without considering its very low price. Despite not having any instructions, the PCB design and finish was very good, the component quality was evidently quite good and construction was trouble-free. Of course, it could have been made better if a power connector and IC socket was supplied, but that would have been a bit much to ask given the AU$1.75 price tag.
Despite having only one IC and being seemingly deceptively simple, it proved to be quite an educational experience to trace out the schematic and try simulating it in LTspice. After all, it’s been a long time since I had to drag out any SPICE simulator, so it was a good experience just to refresh my skills. More than that, it reminds me of how difficult analysis of analog electronics can be – even after spending some time with the simulator, I’m sure that I still don’t understand the finer points of how the circuit works despite understanding the overarching concept. For this alone, I think I got a lot more than my money’s worth.