Fullscreen HTML5Med Res (??MB)HTML4

Max’s Cool Beans By Max the Magnificent Arduino Bootcamp – Part 19 D o you know what you were doing at 5:16:20 pm on Saturday, 23 November 1963? I do. I was six years old, and I was standing bravely (cowering) behind the sofa with my teddy bear captivated by the sound of the electronic opening music accompanying the very first episode of the British science fiction television series, Doctor Who. I’ve been a devoted fan ever since. Why did this just pop into (what I laughingly refer to as) my mind? Well… A gaggle of techno-geeks There is no official collective noun for a group of engineers, but two terms I favour are ‘a sprocket of engineers’ and ‘a rube of engineers,’ where the latter derives from the American cartoonist, sculptor, author, engineer and inventor, Reuben Garrett Lucius Goldberg (a.k.a. Rube Goldberg). Rube is best known for his cartoon depictions of complicated gadgets performing simple tasks in indirect and convoluted ways. Fig.1. Electronic component artwork for the Doctor Who TARDIS (Source: Paul Parry). 50 (a) Flat cut (b) Bevel cut Fig.2. A cut above the rest. As I’ve mentioned before, whenever I return to England to visit my dear old mom, on the Friday before I return to America, a gaggle of my techno-geek friends flock from across the country. We all meet at the house of my brother, Andrew, where we spend a happy day and evening showing off our latest and greatest creations and telling tall tales of derring-do. Why do I mention this here? Well… Electronic component art One of the guys who attends our geeky gatherings is Paul Parry, who is the founder and owner of Bad Dog Designs (https://bit.ly/3QsB8kw). Over the past decade, Paul has captivated and cultivated a worldwide following for his Nixie-tube-based timepieces. More recently, Paul has started a new line in electronic component-based artworks (https://bit.ly/4dryK7t). I love all these creations, but the one that really caught my eye was that of Doctor Who’s TARDIS (Fig.1). I have one on order for when I visit the UK this summer. Meanwhile, back in America, I meet up with a group of friends on Tuesday evenings. We have a light supper and then watch two episodes of Doctor Who (we started with the 2005 reboot starring Christopher Eccleston). These guys are going to be blown away when I flaunt my new acquisition. A handy-dandy tip In my previous column (PE, June 2024), I remarked on how my friend Joe Farr has a vast amount of practical knowledge. I also commented on the kits of pre-stripped, pre-formed, solid core jumper wires you can find on Amazon (https://bit.ly/3vDs2KK). While we were chatting about ‘this and that,’ Joe dropped a little nugget of knowledge into the conversation. He noted that he always cuts his solid core jumper wires at a 45° angle if he’s using them with breadboards (Fig.2). This is (a) Without insulating tape (b) With insulating tape Fig.3. The old insulating tape trick. because (a) they slide in more easily and (b) square ends have the potential to damage the breadboard’s sockets. I tell you; I learn something new every day! Seeing red Joe also taught me another tidbit of trivia. He says he usually covers the front of his red light-emitting diode (LED) 7-segment displays with a piece of red insulating tape. I must admit to being surprised. ‘I’m surprised,’ I said (so it must be true). I told Joe I’d assumed insulating tape was opaque. He replied that although the tape does dim the displays a little, it also makes the active segments ‘pop.’ I just tried this myself and I see what Joe means. I’ve tried to replicate this effect in Fig.3 (only partially successfully, I fear). The situation without insulating tape is shown in Fig.3a. In this case, the active segments are a little brighter, but the inactive segments still manage to stand out. By comparison, when the insulating tape is applied as depicted in Fig.3b, even though the active segments are a tad dimmer, the inactive segments become almost invisible, thereby making their active counterparts more prominent. Once again, I learn something new every day! Resistance is futile! Toward the end of my previous column, I showed a picture of my own breadboard setup with four multiplexed 7-segment displays presenting the hours and minutes in HH:MM format. At that time, I directed your attention to a rinky-dinky Practical Electronics | July | 2024 Fig.4. Replacing free-standing resistors with a BOB. little board in the upper right-hand corner of my breadboards. As you may recall, at one stage in our experiments we had two groups of eight 150Ω current-limiting resistors (this was before we realised that we needed only one such group). We started off using free-standing resistors, as shown in the upper portion of Fig.4. In addition to looking untidy, one problem with having a bunch of resistors in proximity like this is that it’s easy for the leads of adjacent resistors to make unwanted connections with each other. This can result in annoying issues, especially if the connections are intermittent. One solution is to replace the freestanding resistors with dual in-line (DIL) resistor packages. An example is provided by the Bourns 4100R Series (https://bit.ly/4biAmPH). These are available in different-sized packages with different numbers and configurations of resistors. For our purposes here, we would like a 16-pin package containing eight isolated 150Ω resistors. The main problem with this approach is that the range of resistance values supported by the Bourns 4100R Series (and similar offerings from other manufacturers) is limited. A cheap-and-cheerful alternative is to build our own breakout board (BOB) carrying eight discrete 150Ω resistors, as shown in the middle of Fig.4. In addition to preventing any leads shorting, using BOBs like this makes it easy to replace one group of resistors with another, which is something we may well be doing a lot of in the not-so-distant future (I feel like an oracle… but where are we going to find one at this time of the day?). Building a BOB As an aside, I have more than my fair share of friends called Bob. When I’m Practical Electronics | July | 2024 talking to my wife (Gina the Gorgeous), I need to qualify which Bob is the topic of discussion, so I call them Bible Bob, Carpenter Bob, Sparkly Bob... I recently ran (a) Saw between holes (b) Score and snap through holes across a science fiction book called We Are Legion (We Are Bob). I had to buy this book Fig.5. Preparing the BOB board. just for the title. It was only board to the header pins at the same time. after it had arrived that I discovered it’s Fight this temptation with all your might. an awesome story. And I’m not the only If you do solder them simultaneously, one who thinks so because this tome there’s more than a batting chance the has 4.5 stars with 24,000+ ratings on solder won’t wick (get sucked through) Amazon (https://bit.ly/3wsnoiQ). the via all the way to the other side. In But we digress. To build our BOB, turn, when you eventually plug your we will need to start with a piece of BOB into your project’s breadboard, board. One contender is stripboard the header pins may separate from the (https://bit.ly/3JLtbTJ), but this would BOB (sad face). require us to cut tracks using a knife or Thus, our first task is to solder the a special stripboard track cutting tool board to the header pins (Fig.6a). Use (https://bit.ly/3UprIYh). In this case, a hot soldering iron and hold it on each however, I opted for a piece of prototyping joint long enough for the solder to wick board (https://bit.ly/3UMkcZ1). all the way through the via (remember Now we have another choice. We can we talked about soldering, along with either use a fine-toothed hacksaw to cut Alan Winstanley’s handy-dandy Basic our board between the holes (Fig.5a), Soldering Handbook, in PE, March 2024). or we can use a utility knife to score When it comes to attaching the lines through the holes and then snap resistors, you have another decision to the board along these lines (Fig.5b). In make because there are various ways to my case, I opted for the latter solution do this. Joe comes from a prototyping because I happened to have a utility background where things often need to knife in my pocket whilst my hacksaw be reworked. Based on this, he favours was hiding in the tool chest in the garage simply bending the resistor leads at 45° (thus are our decisions made). angles, soldering them to the pins, and We’re also going to need cut two cutting off any excess lead (Fig.6b). The 8-pin pieces of the same long-tailed advantage of this approach is that it’s (0.1-inch pitch) header pins we used easy to desolder things if required later. to create the breadboard power LED By comparison, I tend to come from assemblies in our previous column a more mission-critical and safety(https://bit.ly/3U1Vp2z). Stick these 8-pin critical background where ‘Failure is header pin strips into an old breadboard not an option.’ Based on this, I tend to mounted four holes apart. The reason overengineer things, such as twisting the for using an old breadboard is that we leads through a full 360° circle (Fig.6c). are about to do some soldering. Bits of The downsides to this approach are solder and fragments of wire can fall that it’s easier to create solder bridges into the breadboard’s holes, so we will between adjacent pins (because the joints want to reserve this board for soldering are bigger) and it’s harder to rework the and similar non-active activities in the BOB later. The upside is that these solder future. This means it would be a good joints are not going to fail anytime soon. idea to mark ‘Do Not Use’ on this board Just to make sure we’re all up to date, for future reference. I’ve added this BOB to the drawing Now, this next part is important, so of the latest and greatest incarnation sit up and pay attention. You may be of our prototyping platform (file tempted to solder the resistors and the CB-Jul24-01.pdf). As usual, all the files (a) Solder header pins (b) Bend legs at 45° (c) Twist legs through 360° (a step too far? ) Fig.6. Adding the resistors to the BOB board.. 51 No connection C onnection D3 Display segments and pin numbers G round (0V ) D2 Set New Segment Value → Turn Transistor T0 On → Wait 2ms → Turn Transistor T0 Off Set New Segment Value → Turn Transistor T1 On → Wait 2ms → Turn Transistor T1 Off Set New Segment Value → Turn Transistor T2 On → Wait 2ms → Turn Transistor T2 Off Set New Segment Value → Turn Transistor T3 On → Wait 2ms → Turn Transistor T3 Off : (b) Set New Segment Value → Set 2:4 Inputs to 00 → Wait 2ms Set New Segment Value → Set 2:4 Inputs to 01 → Wait 2ms Set New Segment Value → Set 2:4 Inputs to 10 → Wait 2ms Set New Segment Value → Set 2:4 Inputs to 11 → Wait 2ms : May result in “Ghosting” on old display (c) Set 2:4 Inputs to 00 → Set New Segment Value → Wait 2ms Set 2:4 Inputs to 01 → Set New Segment Value → Wait 2ms Set 2:4 Inputs to 10 → Set New Segment Value → Wait 2ms Set 2:4 Inputs to 11 → Set New Segment Value → Wait 2ms : May result in “Ghosting” on new display (d) en = 0 → Set 2:4 Inputs to 00 → Set New Segment Value → en = 1 → Wait 2ms en = 0 → Set 2:4 Inputs to 01 → Set New Segment Value → en = 1 → Wait 2ms en = 0 → Set 2:4 Inputs to 10 → Set New Segment Value → en = 1 → Wait 2ms en = 0 → Set 2:4 Inputs to 11 → Set New Segment Value → en = 1 → Wait 2ms : (e) Turn All Segments Off → Set 2:4 Inputs to 00 → Set New Segment Value → Wait 2ms Turn All Segments Off → Set 2:4 Inputs to 01 → Set New Segment Value → Wait 2ms Turn All Segments Off → Set 2:4 Inputs to 10 → Set New Segment Value → Wait 2ms Turn All Segments Off → Set 2:4 Inputs to 11 → Set New Segment Value → Wait 2ms : 5 10 9 1 2 4 6 7 D0 D1 (a) DP G F E D C B A 3 3 3 3 C C C C B B B E E 13 B E 11 12 E 10 9 8 7 6 5 4 3 2 Arduino pin numbers Fig.7. Current circuit with four 7-segment displays. In a crunchy nutshell, the question posed by James the inquisitor can be summarised as follows: ‘Why don’t we use a 2:4 decoder to drive the transistors, thereby freeing up two of the Arduino’s digital I/Os?’ That’s a good question. Remembering that I’m only ever an email away, I’m surprised you haven’t presented this poser yourself. Using a 2:4 decoder (we read this as ‘two-to-four decoder’) is certainly a possibility, although there are some considerations to… well, consider. On the off chance this function is new to you, let’s start by considering what we mean by a 2:4 decoder in the first place. The idea is to create a logic function that has two inputs and four outputs. The inputs, which we will drive using our Arduino, have 2 2 = 4 possible combinations: 00, 01, 10, and 11. We want to use each of these input combinations to select one of our four possible outputs, which we will use to control the transistors that control our displays. Furthermore, in the case of our current circuit, we require ‘active high’ outputs, which means their active state is 1 and their inactive state is 0. This is due to the way we have things wired, whereby a 1 will turn the corresponding transistor on. Remember, we want only one transistor (and thus one display) to be active at a time. It may (or may not) be worth noting that our hypothetical 2:4 decoder can be thought of as implementing a ‘one hot’ mentioned in this column are available from the July 2024 page of the PE website: https://bit.ly/pe-downloads Why don’t we… I received an interesting email from a reader we will call James (because that’s his name). James had a question regarding our current circuit (Fig.7). As you will recall, in the present incarnation of our clock, we are using eight of our Arduino’s digital input/output (I/O) pins to drive the display segments, with all four displays connected in parallel (sideby-side). Also, we are using four digital I/O pins to control the transistors we use to activate the displays. This means we are using 12 digital I/Os in all. As we’ve previously noted, the Arduino Uno has only 14 digital I/O pins (remember that we aren’t considering the 6 analogue pins, which can also be used as digital I/Os, because we are holding them in reserve for a later date). Also, we aren’t using digital I/Os 0 and 1 because these are reserved for communications between the Arduino and the host computer. So, the situation we find ourselves in is (a) the Arduino Uno has 12 digital I/Os available to us and (b) we are using all of them to drive our four displays. This means we don’t have any digital I/Os free to implement other functions, like driving our piezoelectric buzzer, for example (we introduced this component in PE, December 2023). s1 s0 y3 y2 y1 y0 0 0 0 0 0 1 0 1 0 0 1 0 1 0 0 1 0 0 1 1 1 0 0 (a) Truth table Fig.9. Alternative control scenarios. s1 s1 y3 F rom s1 Arduino s0 2:1 y2 y1 y0 0 (b) Symbol s0 T o transistors s0 y3 & y2 & y1 & y0 (c) G ate-level implementation Fig.8. 2:4 decoder with active-high outputs (driving 4 displays). 52 & code in that only one of the four outputs is ‘hot’ (active) at any time. Based on the functional specification we just established for our 2:4 decoder, we can create a truth table (Fig.8a), a symbol (Fig.8b), and a gate-level implementation (Fig.8c), where s stands for ‘select’ and y stands for ‘output.’ In this case, we are using two NOT (inverter) gates and four 2-input AND gates. You can find a quick refresher on the truth tables and functioning of NOT, AND, NAND and XOR gates in my Cool Beans Blog on the C/C++ bitwise operators (https://bit.ly/3UqT9B6). A more in-depth discussion is provided in my book Bebop to the Boolean Boogie (https://bit.ly/4a6wDDm). So, in the case of our current fourdisplay clock, we now have a choice between not using a 2:4 decoder and having no free digital I/Os or using a 2:4 decoder and having two free digital I/Os. As a slight aside, if we ever decide we want to drive anywhere from five to eight displays, the way we are currently implementing things would leave us with no choice but to use a 3:8 decoder (three inputs and eight outputs), which would still leave us with one free digital I/O. Seeing ghosts Remember that we are just spitballing ideas here. We aren’t really going to deploy a 2:4 decoder, but it’s well worth taking the time to cogitate and ruminate over the implications if we were to do so. Your knee-jerk reaction might be that using a 2:4 decoder as discussed above would be easy peasy lemon squeezy. Hold hard! Not so fast! Remember that I said there were considerations to consider. Take a moment to think about (a) our gate-level implementation and (b) how we might use the Arduino to drive the inputs to the decoder. Can you envisage any potential problems? Practical Electronics | July | 2024 s y1 y0 0 0 1 1 1 0 s 1:2 y1 y0 (b) Symbol (a) Truth table s y0 d0 d1 y1 y0 s (c) Gates y1 d0 d1 (d) Waveforms Fig.10. 1:2 decoder with active-high outputs (driving 2 displays). Hold onto your hat because I’m about to make your head spin and your brain wobble on its gimbals. Let’s start by reminding ourselves that the way our program is driving our existing circuit can be summarised as shown in Fig.9a (observe we are using the twomillisecond (2ms) on/active time for our displays that we settled on in our previous column). The key point with respect to our existing implementation is that it’s possible for us to turn all the transistors off at the same time. Now consider our 2:4 decoder-based solution. In this case, there will always be one of our displays that’s active. In the context of this column, the term ‘ghosting’ refers to our trying to display two values too closely together on the same display, one for a longer time (the desired value) and one for shorter duration (the fainter, ‘ghost’ value). For the moment, let’s assume that we are always driving our segments with on/off values associated with one of the digits 0 through 9. In this case, we have two options. We could set the segments to their new values and then set the 2:4 decoder’s inputs to their new values (Fig.9b), but this may result in ghosting on the old display. Alternatively, we could set the 2:4 decoder’s inputs to their new values before setting the segments to their new values (Fig.9c), but this may result in ghosting on the new display. Could things get worse? You betcha! One at a time The way we’ve been thinking about things based on the truth table in Fig.8a and the sequences discussed in Fig.9b and Fig.9c is that the inputs to our 2:4 decoder will transition as follows: 00, 01, 10, 11, 00… In turn, this means we are activating our displays in the order D0, D1, D2, D3, D0… Unfortunately, this leads us to fall into the trap of thinking that both inputs to our 2:4 decoder will change simultaneously. Thus far, however, we’ve learned how to change only one of our Arduino’s digital outputs at a time. Practical Electronics | July | 2024 What are the implications of this? Well, suppose that s1 = 0 and s0 = 1, which means display D1 is active. Now suppose we wish to deactivate display D1 and activate display D2. This means we want s1 = 1 and s0 = 0. To put this another way, we want the s1-s0 inputs to our 2:4 decoder to transition from 01 to 10. One way to achieve this would be to use the following statements: digitalWrite(S0, 0); digitalWrite(S1, 1); Unfortunately, this means the inputs to our 2:4 decoder would now follow the sequence 01, 00, 10, which means we would be driving our displays in the order D1, D0, D2. Alternatively, we could swap the order of our statements to read as follows: digitalWrite(S1, 1); digitalWrite(S0, 0); In this case, the inputs to our 2:4 decoder would follow the sequence 01, 11, 10, which means we would be driving our displays in the order D1, D3, D2. Either way, we may now experience ghosting on two displays (let’s call this ‘double ghosting’) rather than one (our original ‘single ghosting’ scenario), which means things are going from bad to worse. Happily, there are two ways around this problem that immediately spring to mind. The first relies on the fact that it doesn’t really matter in which order we activate our displays, just so long as we follow the same sequence repeatedly. So, rather than feeding the inputs of our 2:1 decoder with the binary sequence 00, 01, 10, 11, 00… (D0, D1, D2, D3, D0…), in which multiple bits change on the 01 to 10 and 11 to 00 transitions, we could instead use a Gray code sequence (named after Frank Gray), such as 00, 01, 11, 10, 00… (D0, D1, D3, D2, D0…), in which only one bit changes at a time. An alternative solution relies on the fact that there is a way by which we can change multiple Arduino outputs simultaneously (although we don’t want to get into that here). However, we would still have our original ‘single ghosting’ problem, and neither of these techniques would protect us from another potential ‘gotcha’ resulting from gate-level delays. Don’t delay We often fall into another trap – that of thinking our primitive logic gates switch instantaneously. In the real world, however, there’s always some element of delay. Just to make things a little simpler, let’s perform a thought experiment. Let’s imagine we have only two displays and consider the signals associated with a 1:2 decoder implemented using two NOT gates (Fig.10). As we see, due to the delays associated with the NOT gates, there’s a period when both the output signals are 1, which means both the displays will be activated at the same time. The same thing will apply to our 2:4 decoder (and a 3:8 decoder if we decided to use one); that is, there will be short periods where multiple displays are active simultaneously. With respect to our earlier ghosting examples (both ‘single ghosting’ and ‘double ghosting’), only one display was active at any time. In this case, however, we really do have multiple displays active simultaneously. This means that, in addition to a new form of ghosting, our Arduino’s pins will have to drive twice the expected current. Since we are talking about multiple outputs being simultaneously active for a duration of only a few nanoseconds due to this gate delay scenario, would either of the associated ghosting and current issues really be a problem? Probably not. On the other hand, I think it behooves us to engineer things in such a way that such a situation never occurs in the first place. Adding an enable There are two relatively simple solutions that will address both our sequencing and our gate-delay-induced problems. First, we could add an en (‘enable’) input to our 2:4 decoder. This signal would be controlled by our Arduino. In this case, let’s say that if en is 0, all our outputs will be forced to their 0 (inactive) states. It’s only when en is 1 that the selected output will be 1. The truth table, symbol, and gate-level implementation of this new incarnation of our 2:4 decoder function are shown in Fig.11. In the case of the truth table (Fig.11a), X represents a ‘Don’t Care’ state. Thus, we read the first row of this table as: ‘When en = 0, we don’t care what values are presented to s1 and s0 because we are going to force all the outputs to 0.’ We read the remaining rows as: ‘It’s only when en = 1 that we use the values on s1 and s0 to activate one of our outputs.’ Having this enable signal allows us to use the control sequence presented in Fig.9d. This solution addresses all the ghosting and over-current issues we introduced earlier. However, there’s a fly in the soup and an elephant in the room (I never metaphor I didn’t like). Remember that we started off using 4 of the Arduino’s pins to directly drive our transistors. Then we moved to using only 2 of the Arduino’s pins to drive our 2:4 decoder, which freed up 2 pins, but caused us to run into ghosting and other issues. By adding the enable to 53 implementation shown in Fig.8c. To do this, we will need a hex inverter y3 & s1 in the form of an SN7404 (Fig.13a) 0 X X 0 0 0 0 y3 y2 & and a quadruple 2-input AND in 1 0 0 y2 0 0 0 1 F rom s 1 To s0 2:1 y1 & Arduino s 0 the form of a SN7408 (Fig.13b). 1 0 1 0 0 1 0 y1 T ransisto rs s0 Both these devices are available in 1 1 0 0 1 0 0 y0 y0 & 14-pin breadboard-mountable dual 1 1 1 1 0 0 0 en en in-line (DIL) packages. You can also (a) T ruth Table (c) G ate-level implementation (b) S ymbol access downloadable data sheets: 7404 (https://bit.ly/44Rs3bb) and Fig.11. 2:4 decoder with active-high enable and active-high outputs. 7408 (https://bit.ly/3UwuCKV) on the interweb. You can order these chips individually means we can turn them all off before the 2:4 decoder, we are now using 3 of if you wish, or you could opt for a we modify the values on the inputs to the Arduino’s pins, leaving only 1 pin small kit containing two each of the our 2:4 decoder. When all the segments free, which prompts us to ask, ‘Is all ten most popular 74LS devices, such as are off, it doesn’t matter if we transition this worth the effort?’ the Bridgold set on Amazon (https://bit. through intermediate input values or if The answer to this question is, ‘It ly/3wqgzyv). Note that the ‘LS’ portion the decoder’s gate delays cause multiple all depends.’ If having a free pin is of their monikers stands for ‘low-power transistors to be activated, because all we mission-critical and this is the only Schottky,’ but that’s not something we are displaying is… a blank. This means way to get it, then the answer would need to concern ourselves with here. that if we were to add a 2:4 decoder to be ‘Yes.’ Otherwise, we might have to The plastic packages typically have our current circuit, we would employ think about living without the decoder. a small semicircular notch at one end the sequence depicted in Fig.9e. between pins 1 and 14 as shown in Now, you may be thinking that our Drawing a blank Fig.13. Alternatively, they may have a rambling 2:4 decoder discussions serve Are you drawing a blank with respect dimple next to pin 1. Either way makes it no purpose, to which I would respond: to the second solution that will address easy to determine which pins are which. (a) it’s all curious James’s fault for asking both our sequencing and our gate-delayN e x t , l e t ’s u s e t h e g a t e - l e v e l the question in the first place and (b) induced problems? If so, that would implementation from Fig.8c and our you may be surprised how the problems be ironic, because ‘drawing a blank’ knowledge of the chip pins from Fig.13 and solutions we’ve talked about here is the solution. to create a schematic (Fig.14). pop up again in the future. Let us now return to our current It’s common to use the character ‘U’ implementation, but we’ll assume it’s to indicate an integrated circuit in a What the heck! been augmented with a simple 2:4 schematic (we are saying that U1 is our I fear that I can no longer be trusted. decoder (without an enable) as illustrated SN7404 and U2 is our 7408). In the case Why? Well, earlier I said that we were in Fig.12 (observe that we now enjoy of this sort of logic, sometimes we might just spitballing ideas and we weren’t two free pins). show the entire package as a single really going to deploy a 2:4 decoder. Do you remember earlier when we said: symbol. Other times we may show the However, looking at Fig.12 has given me ‘For the moment, let’s assume that we are individual gates, in which case we add a an uncontrollable urge to do just that. always driving the segments associated number or letter to indicate which gate I just glanced at my trusty Texas with one of the digits 0 through 9.’ I’m in the package we are talking about (eg, Instruments TTL Data Book (Second afraid this was a bit of a hareng rouge U2.1 is the gate in the 7408 with signals Edition) from 1981 – a classic, do yourself (‘red herring’). 1A, 1B, and 1Y; U2.2 is the gate with a favour and get one. The SN74139 As we know, we currently have signals 2A, 2B, and 2Y; etc.). The pin is a 16-pin device containing two 2:4 total control of all the segments. This numbers provide a further clue. decoders, and they Now, let’s add these devices to our even have enables Displ ay segments No connection G round (0V ) C onnection breadboard-based clock (Fig.15) and wire if we wanted to use and pin numbers them up as per our schematic (Fig.14). them, but they have I’ve updated our master drawing to reflect active low outputs, these changes (file CB-Jul24-02.pdf). which is opposite to We will talk about the four unexpected what we want. 5 10 9 1 2 4 6 7 D3 D2 D0 D1 resistors in a moment, but first… Also, it will be Sometimes when you look at a gatemore fun to replicate DP G F E D C B A level schematic of a circuit board like the gate-level s1 en s 1 s 0 y3 y2 y1 y0 3 3 3 3 C C C C B B E B E B E (+5V ) (+5V ) V CC V6A CC 6A 6Y 6Y 5A 5A 5Y 5Y 4A 4A 4Y 4Y (+5V ) (+5V ) V CC V4B CC 4B 4A 4A 4Y 4Y 3B 3B 3A 3A 3Y 3Y 14 13 12 12 11 11 10 10 9 98 8 14 13 12 12 11 11 10 10 9 98 8 & & & & 14 13 14 13 E & y3 y2 y1 y0 & & & 2:1 s1 s0 F ree 13 12 9 8 7 6 5 4 3 2 11 10 Arduino pin numbers Fig.12. Current circuit augmented by 2:4 decoder. 54 1 12 23 34 45 56 1A 1A 1Y 1Y 2A 2A 2Y 2Y 3A 3A 3Y (a) SN7404 (a) SN7404 7 1 12 23 34 45 56 G3Y ND G ND (0V ) (0V ) 67 1A 1A 1B 1B 1Y 1Y 2A 2A 2B 2B 2Y 67 7 G2Y ND G ND (0V ) (0V ) (b) SN7408 (b) SN7408 Fig.13. Chip packages. Practical Electronics | July | 2024 4 Connection 6 & 5 No connection y3 U2.2 1 From Arduino 3 & 2 U2.1 5 12 6 y2 To transistors 11 & 13 U1.3 s0 P in 10 1 U1.1 s1 Pin 11 2 y1 U2.4 10 U1 = 7404 U2 = 7408 8 & 9 y0 U2.3 Fig.14. Gate-level schematic for 2:4 decoder. D3 D2 D1 To displays D0 7408 7404 11 (s1) 10 (s0) From Arduino Listing 3a. Modifications to use 2:4 decoder. Fig.14, you wonder as to the rationale behind the way in which the designer selected which gates in which chips were to be connected to which wires. Why did I use U2.3, U2.4, U2.1, and U2.2 to drive signals y0, y1, y2, and y3, respectively, for example? Oftentimes, the answer is more pragmatic than you might expect – it facilitates the physical layout in some way. In this case, I wanted to be able to draw the wires associated with our chips in such a way that they didn’t cross over, which would have made things harder for you to comprehend. Feeling unconnected? One final point worth noting before we fire up our new implementation is that we are using only two of the six inverters (NOT gates) in our SN74LS04 package. In a real-world design intended for production, we would typically use pullup or pull-down resisters to connect any unused input pins to power or ground, respectively (Fig.16). Why? Well, there’s a faint possibility that an unconnected input, which may ‘float’ up and down, could result in that NOT gate undergoing uncontrolled oscillations, and an oscillating gate consumes power. Also, even though Practical Electronics | July | 2024 Fig.15. Wiring up the chips on the breadboard. the outputs on our unused gates aren’t connected to anything via wires, such oscillations may result in noise in the form of electromagnetic interference (EMI) that can affect other signals nearby. Happily, the chances of this being a problem are extremely slight in our case, not least because of the SN74LSxx technology we are using. On the other hand, I’m trying to teach you good practices, and it’s a good idea to get into the habit of tying off unconnected inputs, so that’s what we are going to do. In a scenario where gates are totally unused, as in our example, it’s often up to the designer to decide between attaching pull-up or pull-down resistors to their inputs (remember that they are the same resistors; the ‘up’ and ‘down’ qualifiers are used to remind ourselves what we are doing with them). In this case, I favour using pull-down resistors to ground. Also, it’s usually the designer’s choice as to the value of resistors to use. Anything 1kΩ or higher is typically good. I personally favour 10kΩ resistors (brown-black-orange). I have friends (seriously) who opt for 2.2kΩ (red-redred) or 4.7kΩ (yellow-purple-red). In the case of a cheap-and-cheerful prototype, you may end up using whatever you have to hand. In some situations, we may be using a gate or a more complex function, but not employing all its inputs. In such cases (unless stated otherwise in the data sheet), it’s imperative that we apply appropriate resistors to place these pins in appropriate states. If we are using only two inputs to a 3-input AND gate, for example, we would use a pull-up resistor on the unused input (a pulldown would force the AND’s output to (+5V ) V CC 14 6Y 13 12 1 2 3 1A 1Y s1 s1 5Y 11 10 4Y 9 8 4 5 6 7 2Y 3A 3Y G ND (0V ) s0 s0 Fig.16. Unconnected inputs connected to ground via pull-down resistors. 55 0). Alternatively, if we are using only two inputs to a 3-input OR gate, we would use a pull-down resistor on the unused input (a pull-up would force the OR’s output to 1). Let’s rock and roll! I just forked the latest and greatest program from our previous column (PE, June 2024), edited it, and made it available for your perusing pleasure (file CB-Jul24-03.txt). Let’s skim through the minor changes before we look at the tasty part. We’ve removed our TRAN_ON and TRAN_OFF definitions because we are no longer driving the transistors directly. Similarly, we’ve replaced our PinsTrans[] array in which we specified the four pins used to drive our transistors with two independent definitions for our decoder select controls: int PinS0 = 10; int PinS1 = 11; There are a few more minor tweaks of this ilk, but the big change takes place in our loop() function, as illustrated in Listing 3a. To be honest, apart from the line numbers themselves, the statements associated with Lines 71 through 89 remain unchanged. The only new part takes place when we come to output our HH:MM data to the displays in Lines 92 through 99. What we are doing is implementing the sequence shown in Fig.9e: Turn all segments off (Line 94) → Set 2:4 inputs (Lines 95 and 96) → Set new segment values (Line 97) → Delay 2ms (Line 98). Most of this is achieved by reusing and reordering statements we’ve seen before. The only new part occurs on Lines 95 and 96 when we set the 2:4 input signals. To understand this, we first need to remind ourselves that we previously made the decision to activate our displays in the sequence D0, D1, D2, D3 (we could have done it the other way round, but this was easier). To achieve this with our new circuit, we want the s1-s0 signals driving the inputs to our 2:4 decoder to follow the sequence 00, 01, 10, 11 (Fig.8a, Fig.14, and Fig.15). As part of our for() loop declaration on Line 92, our integer iDisp variable cycles through the values 0, 1, 2 and 3. In the case of the Arduino Uno, variables of type int (integer) are 16 bits wide. Let’s consider the binary values of the leastsignificant eight bits in the context of our counting sequence (the most-significant eight bits will all be 0s). iDisp 0 ≡ 1 ≡ 2 ≡ 3 ≡ Binary 00000000 00000001 00000010 00000011 The math symbol ≡ means ‘equivalent to’ or ‘identical to.’ We discussed how to use the bitwise operators as masks in a previous Cool Beans (PE, March 2021). These are the techniques we are using here. Consider the following portion of Line 95: iDisp & B0000001 Components from Part 1 LEDs (assorted colours) https://amzn.to/3E7VAQE Resistors (assorted values) https://amzn.to/3O4RvBt Solderless breadboard https://amzn.to/3O2L3e8 Multicore jumper wires (male-male) https://amzn.to/3O4hnxk Components from Part 2 7-segment display(s) https://amzn.to/3Afm8yu Components from Part 5 Momentary pushbutton switches https://amzn.to/3Tk7Q87 Components from Part 6 Passive piezoelectric buzzer https://amzn.to/3KmxjcX Components for Part 9 SW-18010P vibration switch https://bit.ly/46SfDA4 Components for Part 10 Breadboard mounting trimpots https://bit.ly/3QAuz04 Components for Part 12 Light-Dependent Resistor https://bit.ly/3S2430m Components for Part 13 BC337 NPN Transistor https://bit.ly/40xAgyS Components for Part 14 HC-SR04 Ultrasonic Sensor https://bit.ly/49AMBq4 Components for Part 15 Real-Time Clock (RTC) https://bit.ly/3S9OjHl Components for Part 18 Long tailed (0.1-inch pitch) header pins https://bit.ly/3U1Vp2z Components for Part 19 Prototyping boards Kit of popular SN74LS00 chips https://bit.ly/3UMkcZ1 https://bit.ly/3wqgzyv What this means is that this operation will return 00000000 (0) when iDisp is 0 or 2, and 00000001 (1) when iDisp is 1 or 3. Remembering that 0 and 1 are equivalent to LOW and HIGH, respectively, these are the values we are writing to our 2:4 decoder’s s0 input. Now consider the following portion of Line 96: (iDisp >> 1) & B0000001 This starts by using the >> bitwise shift right operator to shift the value from iDisp one bit to the right (again, this doesn’t modify the value stored in iDisp because we aren’t assigning anything to iDisp). The original least-significant bit ‘falls off the end’ and a 0 is shifted into the new most-significant bit. What this means is that, following the >> operation, we have the following intermediate value to play with: iDisp 0 ≡ 1 ≡ 2 ≡ 3 ≡ Binary after >> 1 00000000 00000000 00000001 00000001 Since we are using a bitwise & (AND) operator, any 0s in our When we subsequently perform the bitwise & (AND) operation, Bxxxxxxxx value will return 0s in the corresponding positions we will see 00000000 (0) when iDisp is 0 or 1, and 00000001 (1) in the result. By comparison, any 1s in our Bxxxxxxxx value will return whatever 0 or 1 value is in the corresponding bit of Cool bean Max Maxfield (Hawaiian shirt, on the right) is emperor of all he iDisp (note that this doesn’t surveys at CliveMaxfield.com – the go-to site for the latest and greatest modify the contents of iDisp in technological geekdom. because we aren’t assigning Comments or questions? Email Max at: max<at>CliveMaxfield.com anything to iDisp). 56 Practical Electronics | July | 2024 when iDisp is 2 or 3, and these are the values we are writing to our 2:4 decoder’s s1 input. There are a couple of things to note here. The first is that the following two statement return identical results: (iDisp >> 1) & B0000001 iDisp >> 1 & B0000001 This is because the >> (bitwise shift) operator has a higher precedence than the & (bitwise AND) operator (for further details on this, I recommend C/C++ Operator Precedence at: https://bit.ly/3wH8w06). Personally, I prefer to use ( ) parentheses to (a) force the precedence I want and (b) to make it clear what I expect to occur to anyone reading my code (including me) in the future. The second point is that we don’t really need to perform the shift at all. For example, although the following statements return different results, these values will result in the same action in the context of our program: (iDisp >> 1) & B0000001 iDisp & B0000010 This is because the digitalWrite() function will set the output to 0 (off, LOW) if it is presented with a value of 0 (zero), and it will set the output to 1 (on, HIGH) if it is presented with any other (non-zero) value of 1 or more. I just ran our new program, and… ran into a WTW (‘what the what?’) problem with one of my breadboards. I’ll tell you about this in my next column. Suffice it to say that, once I’d sorted my wayward breadboard, our 2:4 decoder-based clock performed like a dream (happy dance). Online resources For the purposes of this series, I’m going to assume that you are already familiar with fundamental concepts like voltage, current and resistance. If not, you might want to start by perusing and pondering a short series of articles I penned on these very topics – see: https://bit.ly/3EguiJh Similarly, I’ll assume you are no stranger to solderless breadboards. Having said this, even if you’ve used these little scamps before, there are some aspects to them that can trap the unwary, so may I suggest you feast your orbs on a column I wrote just for you – see: https://bit.ly/3NZ70uF Last, but not least, you will find a treasure trove of resources at the Arduino.cc website, including example programs and reference documentation. Next time… I’m afraid I’ve told two untruths (my mother won’t be happy). As you may recall, at the end of our previous column, I said: ‘In our next column, we really and truly are going to refresh our minds as to the concept of binary-coded decimal (BCD) and then investigate various deployments of BCD-to7-segment decoders in our clock (otherwise my name’s not Max the Magnificent). The first untruth is that I failed in my BCD quest (we’ll cover that next time). The second untruth is that my name is indeed Max the Magnificent, because that’s what appears as my moniker on Zoom calls and name badges when I attend technical conferences, and you can’t argue with logic like that! Until next time, have a good one! STEWART OF READING 17A King Street, Mortimer, near Reading, RG7 3RS Telephone: 0118 933 1111 Fax: 0118 933 2375 USED ELECTRONIC TEST EQUIPMENT Check website www.stewart-of-reading.co.uk Fluke/Philips PM3092 Oscilloscope 2+2 Channel 200MHz Delay TB, Autoset etc – £250 LAMBDA GENESYS LAMBDA GENESYS IFR 2025 IFR 2948B IFR 6843 R&S APN62 Agilent 8712ET HP8903A/B HP8757D HP3325A HP3561A HP6032A HP6622A HP6624A HP6632B HP6644A HP6654A HP8341A HP83630A HP83624A HP8484A HP8560E HP8563A HP8566B HP8662A Marconi 2022E Marconi 2024 Marconi 2030 Marconi 2023A PSU GEN100-15 100V 15A Boxed As New £400 PSU GEN50-30 50V 30A £400 Signal Generator 9kHz – 2.51GHz Opt 04/11 £900 Communication Service Monitor Opts 03/25 Avionics POA Microwave Systems Analyser 10MHz – 20GHz POA Syn Function Generator 1Hz – 260kHz £295 RF Network Analyser 300kHz – 1300MHz POA Audio Analyser £750 – £950 Scaler Network Analyser POA Synthesised Function Generator £195 Dynamic Signal Analyser £650 PSU 0-60V 0-50A 1000W £750 PSU 0-20V 4A Twice or 0-50V 2A Twice £350 PSU 4 Outputs £400 PSU 0-20V 0-5A £195 PSU 0-60V 3.5A £400 PSU 0-60V 0-9A £500 Synthesised Sweep Generator 10MHz – 20GHz £2,000 Synthesised Sweeper 10MHz – 26.5 GHz POA Synthesised Sweeper 2 – 20GHz POA Power Sensor 0.01-18GHz 3nW-10µW £75 Spectrum Analyser Synthesised 30Hz – 2.9GHz £1,750 Spectrum Analyser Synthesised 9kHz – 22GHz £2,250 Spectrum Analsyer 100Hz – 22GHz £1,200 RF Generator 10kHz – 1280MHz £750 Synthesised AM/FM Signal Generator 10kHz – 1.01GHz £325 Synthesised Signal Generator 9kHz – 2.4GHz £800 Synthesised Signal Generator 10kHz – 1.35GHz £750 Signal Generator 9kHz – 1.2GHz £700 HP/Agilent HP 34401A Digital Multimeter 6½ Digit £325 – £375 HP 54600B Oscilloscope Analogue/Digital Dual Trace 100MHz Only £75, with accessories £125 (ALL PRICES PLUS CARRIAGE & VAT) Please check availability before ordering or calling in HP33120A HP53131A HP53131A Audio Precision Datron 4708 Druck DPI 515 Datron 1081 ENI 325LA Keithley 228 Time 9818 Practical Electronics | July | 2024 Marconi 2305 Marconi 2440 Marconi 2945/A/B Marconi 2955 Marconi 2955A Marconi 2955B Marconi 6200 Marconi 6200A Marconi 6200B Marconi 6960B Tektronix TDS3052B Tektronix TDS3032 Tektronix TDS3012 Tektronix 2430A Tektronix 2465B Farnell AP60/50 Farnell XA35/2T Farnell AP100-90 Farnell LF1 Racal 1991 Racal 2101 Racal 9300 Racal 9300B Solartron 7150/PLUS Solatron 1253 Solartron SI 1255 Tasakago TM035-2 Thurlby PL320QMD Thurlby TG210 Modulation Meter £250 Counter 20GHz £295 Communications Test Set Various Options POA Radio Communications Test Set £595 Radio Communications Test Set £725 Radio Communications Test Set £800 Microwave Test Set £1,500 Microwave Test Set 10MHz – 20GHz £1,950 Microwave Test Set £2,300 Power Meter with 6910 sensor £295 Oscilloscope 500MHz 2.5GS/s £1,250 Oscilloscope 300MHz 2.5GS/s £995 Oscilloscope 2 Channel 100MHz 1.25GS/s £450 Oscilloscope Dual Trace 150MHz 100MS/s £350 Oscilloscope 4 Channel 400MHz £600 PSU 0-60V 0-50A 1kW Switch Mode £300 PSU 0-35V 0-2A Twice Digital £75 Power Supply 100V 90A £900 Sine/Sq Oscillator 10Hz – 1MHz £45 Counter/Timer 160MHz 9 Digit £150 Counter 20GHz LED £295 True RMS Millivoltmeter 5Hz – 20MHz etc £45 As 9300 £75 6½ Digit DMM True RMS IEEE £65/£75 Gain Phase Analyser 1mHz – 20kHz £600 HF Frequency Response Analyser POA PSU 0-35V 0-2A 2 Meters £30 PSU 0-30V 0-2A Twice £160 – £200 Function Generator 0.002-2MHz TTL etc Kenwood Badged £65 Function Generator 100 microHz – 15MHz Universal Counter 3GHz Boxed unused Universal Counter 225MHz SYS2712 Audio Analyser – in original box Autocal Multifunction Standard Pressure Calibrator/Controller Autocal Standards Multimeter RF Power Amplifier 250kHz – 150MHz 25W 50dB Voltage/Current Source DC Current & Voltage Calibrator £350 £600 £350 POA POA £400 POA POA POA POA Marconi 2955B Radio Communications Test Set – £800 57