I am sharing my thoughts on the construction project of a ridiculously simple and cheap Touchless Hand Sanitizer Dispenser. The basics behind this build are simple, but I have seen individuals over complicate it! I hope you have a chance to give this do it yourself project a try.
The schematic diagram below consists of a reflective optical sensor linked to a comparator circuit that generates the proximity signal. The comparator circuit drives a medium power transistor to control one small dc water pump to deliver a liquid preparation having an antiseptic or medicinal action when applied to the hand. Construction is extremely simple. This is easily within the capabilities of any novice who know how to grip a charged soldering iron by the right end!
The power supply of the circuit is very simple. It is powered by a ‘plug-top’
USB travel charger that needs to provide a DC voltage of 5V at 750mA minimum. Smartphone USB power bank utilization is also good, since the circuit draws relatively little current in its standby state.
As you can see, the circuit shared here is simple, straight forward, and self-explanatory. The TCRT5000 (SENS-1) is a compact reflective sensor which includes a 950nm infrared emitter and phototransistor in a leaded package which blocks visible light. According to the datasheet its peak operating distance is 2.5mm, and my configuration offers a hand proximity detection range of 80mm maximum (50mm typical). Note that the multiturn trimpot (P1) must be aligned carefully to set the correct hand proximity detection threshold.
The water pump is a cheap electric motor driven micro submersible water pump rated for 6VDC operation. It will work smoothly from 5VDC power supply (2.5VDC minimum) with near 250mA current consumption. In this circuit, the BD139 driver transistor (T1) with the 470Ω base resistor (R4) can drive the water pump safely and smoothly. It is a good idea to look at the water pump’s datasheet before ‘tuning’ the associated driver electronics. Further, it is substantive to cool the driver transistor (T1) as it is important to employ a power driver transistor which is mounted on a small aluminum carrier.
The BD139 is a silicon epitaxial planar NPN transistor designed for audio amplifiers and drivers utilizing complementary or quasi-complementary circuits. See its absolute maximum ratings:
In principle, you could also use another transistor in place of the BD139 which on the surface may appear suitable but be prepared for some homework. The values of prime interest in the transistors’ datasheet are: hFE, Vbe, Ib, Vce (sat), and Ic. I suggest taking a piece of paper and checking everything is as calculated and makes sense.
A side note: If you find that a transistor in a circuit is becoming too hot to touch it surely needs a heat sink (heat is generated in a transistor due to the current flowing through it). The heat is not a serious problem if IC is small or if the transistor is used as a simple switch because when ‘full on’ VCE is almost zero. Usually, the thermal power is determined by the collector current IC and the voltage VCE across the transistor: P = IC × VCE. And, a heat sink is rated by its thermal resistance (Rth) in °C/W (a lower thermal resistance means a better heat sink).
The napkin calculation: Presume that your driver transistor is handling 1A current and the VCE is 1V. If so the power (P) is 1×1=1W. Assuming the maximum operating temperature (Tmax) for that transistor is 100°C, and the maximum ambient temperature (Tair) is 25°C, the maximum thermal resistance (Rth) for the heat sink is (100-25)/1 = 75°C/W (Rth = (Tmax – Tair)/P). Okay, pick a heat sink with a thermal resistance which is less than the calculated value!
Below you can see some snaps of my initial ‘breadboarded’ experimental setup. The idea worked well in practice (indoor and outdoor)!
The simple way to enhance the basic design is add an indicator that lights up when the tank is fully empty. Detecting the empty tank condition and preventing the water pump dry run is not difficult as we have a ‘spare’ op-amp in the LM358 IC. For a simple solution, a tricky dry/wet sensor is glued at the top of the water pump (see below).
And then, I modified the circuit so that it can light up an “empty tank” fault indicator if the tank is in an empty state.
The updated complete schematic is shown below. In this improvement, the infrared emitter of the TCRT5000 reflective sensor also works as a simple voltage reference (circa 1.2V) for the second comparator – see J1 wire link in the new schematic.
What is the point of having an LM358 for a sanitizer dispenser project? I have chosen the LM358 because the IC already includes two op-amps, and it works in a wide operating voltage range. Note an important point, theLM358 has a maximum allowable input voltage of less than Vcc by as much as 1.5V at 25°C or as much as 2V across the whole temperature range (else the output may be ambiguous). But it does not matter here in this circuit configuration (do probe various circuit points in your working prototype and find out why)!
Of course, the mechanical construction of a compact enclosure is more complicated than the build of the core electronics described here. With a bit of practice, you can easily rig up your own model of the automatic sanitizer dispenser. Following the rough layout art (side view) given below it is possible to make a good artifact. This is where you get to start using your own imagination and creativity to create something salient. It is all up to your imagination and ingenuity. You should be able to get the enclosure made up in a maker space or 3D printing lab, assuming you don’t want the bother of fabricating it yourself.
If you are a trifle fainéant maker like me (ha ha), then note that you can get rid of the ‘enclosure nightmare’ simply by using a big mason jar. The only thing that is missing is the professional look (the consequence is that the build time and effort is somewhat reduced). Look at the rough artwork shown below!
Let off the hook
Across the web, there are countless designs for automatic hand sanitizer dispenser machines. Many are very canonical, employing nothing more than a handful of discrete components. Others are microcontroller driven and very complex. There is also an endless fascination by some makers to build the smallest and cheapest versions possible for everyday use. This is not a clever design or a perfect project as there are so many limitations, but I worked (on a budget) to derogate those at the expense of a few minor advantages.
I would like to say that I have been designing or been trying to design an advanced version of this project with a couple of other deluxe features. Whatever your design idea for a feature-rich automatic hand sanitizer dispenser, I would love to know about it.