Week 10: Output device

Vibrating motors

For this week assignment I considered moving steps forward my final project development, getting to know output devices world.
Starting from the list Neil provided us, I decided to start with DC vibration motors. I decided not to follow "replicating Neil's job" path, instead I preferred to start from output devices' characteristics to understand components I would have needed for a minimal circuit.
I really wanted to make acquaintance with vibrating motors because I'll soon have to work with it for my final project. Moreover, I wanted to know how can you control working processes of such a tiny motor without using a motor driver.

Minimal Arduino circuit

In my lab I found Groove starter kit vibrating motor module and I simply desoldered it so as to use it for my PCB.
To start with, I had a look at some tutorials about this kind of output device and this one really seemed clear and complete to me. The example was a minimal circuit made for Arduino and it included:

• Arduino board;
• a vibration motor;
• 1N4001 Diode;
• 0.1µF ceramic capacitor;
• 1KΩ Resistor;
• 2N2222 NPN Transistor;
• USB connector

I highlighted the main components cause each one has a fundamental role that is of primary importance to solve components parallel and series connections and my first intention was to understand WHY those components were necessary and were placed precisely like that :

1- TRANSISTOR: this component's importance stays in that it has to provide the motor a proper current, not too high, not too low; in fact, the vibration motor needs a minimum of current to work ( around 75mA ) but too much current could damage it seriously, therefore, the transistor takes Vin (input voltage) from Arduino output pin (G ¹ line of the transistor) and sends it to the motor (D ² line) or to ground (S ³ line) if it is too high.

2- DIODE:the diode, together with the capacitor, is linked in parallel to the motor and they both act as "security measures"; in fact, the diode lets the current flow in a direction only and this prevents the passage to possible current flows generated by motor working process. Therefore, it better to use an higher DC-blocking-voltage diode than a lower one (which risks to be burnt by high voltages - remember: I ∝ V from OHM's law).

3- CAPACITOR:this component accumulates charges of current coming from the output pin and gradually hands it to the motor; consequently, it better not to use a high-capacity capacitor because it would loose current off too slowly to make the motor working.

From Arduino to ATtiny 44 circuit

Anyway I had to substitute some components it with ATtiny44 to which I also had to link a AVRISP and FTDI connectors.
As for the diode and the transistor, it required me more time to understand how could I replace them with components I had in my lab.

DIODE

The circuit I took as an example used a 1N4001 diode but to know its main characteristics I had a look at its datasheet. The most important one I had to care for was V_RRM which regarded reverse current that the motor may produce. Looking into the drawers around me I could only find 30V V_RRM Despite being the example diode V_RRM higher, using a 30V one wouldn't compromise my board functioning (it is really unlikely that reverse motor current can reach up to 30V) and I could take 1A diode because my coin motor needed around just 75 mA.

TRANSISTOR & MOSFET

As for the transistor, things got complicated a bit, also due to my little electronics knowledge. In the example, they used 2N2222 transistor but I had none in my lab...actually I couldn't find any transistor at all! I spent something like an hour in seeking for any kind of that device without results, just because I had no clear ideas about the difference between MOSFETs and TRANSISTORs.
Transistors look like (usually) 3-pin black tiny bricks and are divided in two big families: BJT (Bipolar Junction Transistors) and MOSFETs. They both work following the same principles but MOSFETs gave greater control abilities whereas BJT are almost used like ON / OFF devices.
BJT BJP are current amplifiers: according to I_B value (current flowing in the base, the current in C (collector) and E (emittor) is higher.

MOSFETs
First of all. let's start from the name:M O S F E T stands for metal–oxide–semiconductor field-effect transistor. So being a FET transistor it uses an electric field to control the electrical behaviour of the device. This component, believe it or not, it's really close to a sandwich: The P-silicon (either it can be N) level is a semiconductor which has been deprived /furnished of electrons to increase conductivity. So, basically it is a transistor with improved skills and despite working on a reference current it considers G (gate)'s voltage. ( Here you can find an example of MOSFET datasheet and here others for many kind of MOSFET).
In my lab I found Now that I had a clearer idea about what I had around me on the table, I could solve my problem using a mosfet. I was then ready to draw my board on EAGLE

BRAINO Drawing

I had already used EAGLE before (see this link for basic tools explanation) but I was running out of time to meet with a new software and also I wanted to make a more complicated board, so I wanted to walk on already-blazed trails.
I easily added and link the components below:

• ATtiny44
• Mosfet
• Diode
• Capacitor
• Resisitor

NB: to link the ATtiny44 to other components, I looked at pinout configuration on the datasheet.

While linking all the pins I needed, I saw an obscure PWM acronym on the example circuit and I wanted to discover what it could mean.

PWM: Pulse Width Modulation

Pulse Width Modulation is a "method" to control some output device, deciding the percentage of time we want the signal to be high or low, so basically it consists in set the duty cycle as desired. From microcontrollers' point of view all this system refers to something like a timer, because they are told (via code) when to set a pin high.
Little more about PWM and SoftwareSerial here

At this point, I wanted to make my board sewable, like Lilypad Arduino or Flora boards, but I didn't know how to add holes in my schematic and I looked up in the internet at this, this, this and this tutorial but I didn't end up with a nice HOW-TO but I thought I could simply look for Lilypad hole device in another library (fab.lbr hasn't them).
NB: notice that all devices in holes.lbr are mounting holes and they're not conductive.

I then found this GitHub repo with two libraries and I found what I needed. Starting from Lilypad schematic I found info about pads I was looking for: ..and I searched for them in Lilypad library: Seeming them too big for my board, I chose another type: I add one of them for each ATtiny pin: I then added an ISP and FTDI connector ( I forgot I wasn't using an Arduino ;) ) ...and also two through-all pads for my vibration motor I found it difficult to give SCK and PA5 pads a position but I realized that one of them (SCK) was linked to the ISP connector and, since its function is related to microcontroller clock, I needn't an outer pad for it and the other (PA5) was already connected to the motor. So i simply deleted them: I selected Top, Pads and Vias layers and I exported it as Image (monochrome, 1000dpi). Now it was milling time...and the best is yet to come ;)

MILLING or LASERCUTTING? That's the question

As you'll soon see, the question received quick answer, I decided to use my lab's Trotec 400 Speedy Flex to cut my board, because I never used for this before but also because Roland SRM-20 was playing against my self control.
I had to define my board edges, being the shape quite strange, I caught the opportunity to imagine something it could look like in real world.....and I saw it looked like an human brain! Hence, I gave it the name of Braino.
I used Illustrator instead of GIMP because free drawing is not the strong arm of my abilities: I copied and pasted the whole sketch in three different files so as to take out from them 3 images:

• traces: where I had the traces, text and pads with no hole - to be engraved with 1/64 of inch mill;
• holes: where I just had the holes I had to cut (black infill) - to be cut with 1/32 of inch mill;
• outline:where I just had the outer profile.

I then exported them as .png files and send them to fab modules (see week4 for detailed settings). Once I loaded outline file I saw the path of the mill would have been the following: NB: notice that what is in black would be cut out and what in white is left. In my case anyway it didn't make much difference because the inner path was still out of traces range.

Now, I know it's rarely machine's fault if you fail, but I believe there was also an ill omen running against me:
First attempt: the PCB board didn't stick enough to the plate and it moved while milling...what came out was to be thrown away but fortunately I didn't break the mill.
Second attempt: I forgot to pull up the mill before moving and I broke it.
Third attempt: I changed the broken mill with a defective one and it didn't cut anymore after a while. My mind went like...

Looking up at that demon that was mocking at me, I saw the lasercutter and I decided she could be my saviour.
I so edited my exportation from .png ( now preparing files for milling looked so much easier! ) and I made it vectorial (see week2 for how to vectorize an image on Illustrator) paying attention that everything that had to be engraved was in black, cuts in red and what to be left in white: ( That thin line is 0.001mm thick in red and there are also red circles inside pads)
I so repeated printing procedure from Illustrator and set parameters on JobControl Thanks to Luca Giacolini for parameters setting.
Anyway, first result hadn't holes cut properly and so I sacrificed that board to find right cutting setting From 80 W power and 0.6 speed I moved to 100 W power and 0.3 speed for cutting but still I had bad results because holes didn't detach and I damaged pads in pushing them out. Since the outline had been cut finely, holes problem's could come from the fact that the cutting line was precisely the hole border and so there was copper that hadn't been scrabbled away that may prevent the laser from passing the whole material thickness. Consequently, I shrunk circles so as they didn't touch copper layer. And....... Tadaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaan!
My Braino came out well!!!

Braino Soldering

For soldering tips see week 4.
I wrote my BOM I soldered some components and then I flayed coin motor wires: NB:to know the direction of the diode, check the side that has one or two lines (they sign the line of diode symbol: -->| ); notice that in my circuit, diode is mount the other way up because it has to prevent current from passing. Then I soldered the motor: Since one of the wires broke, I decided to use pads properly ( ;) ) and I re-arranged my board: NB: so as not to get confuse with blue and red of motor wires, I looked up at this datasheet to check which was + and which was - .

After soldering, I checked my circuit with multimeter and it seemed to be work fine.

Experiments

Grove shield

I couldn't help testing it immediately, still I was afraid to damage it somehow. Therefore, I first tested another Grove vibration module I had with an Arduino Uno usig the code I found here:

int MoPin = 9; // vibrator Grove connected to digital pin 9
void setup() {
pinMode( MoPin, OUTPUT );
}
void loop() {
digitalWrite(MoPin, HIGH);
delay(1000);
digitalWrite(MoPin, LOW);
delay(1000);
}
}

NB: check which pin you are linking the module to ( since I had a previous version of Grove shield I hadn't pin 9, I linked to D7 pin and changed it in the code ).
everything went ok just by copying and pasting the code on Arduino IDE.