Full Stack Electronics · Stage Four · Experiment 21

Experiment 21: First Power-Up

The last step from the breadboard to a printed circuit board. Put the board in place of the NE555 of Experiment 1, switch it on behind a current limit, and measure it as you measured the chip.

Experiment 20 ended with 46 parts on the board, and each joint passed the inspection. No current has gone through the board yet. Some faults show only when current flows. Examples are a short under a part, a joint that does not conduct, and a PNP in an NPN position. A short across the supply can burn a part or a track in a few seconds. So this experiment checks the supply pins with a multimeter, and then switches the board on with a current limit. The circuit is the blinker of Experiment 1, with the board in place of the NE555. The oscilloscope then measures the board as it measured the chip: the period and the levels of the output. Last, the experiment tests RESET, because the blinker does not use it.

You will need

Check the supply pins

The lead named a short across the supply as the fault that does the most damage. A multimeter finds that short before any voltage goes on the board. In continuity mode, the meter beeps when the resistance between its probes is low.

  1. Set the multimeter to continuity. Touch the two probes together. The meter beeps.
  2. Put one probe on the pin marked +5V and the other probe on the pin marked GND. These are pins 8 and 1.
  3. Listen. Expect no beep. A beep means a short between +5 V and ground. Do not connect the supply. Find the short under the microscope, as in Experiment 20, and remove it. Then do this check again.

Put the board in place of the NE555

The supply pins are not shorted, so the board can go into a circuit. The circuit is the blinker of Experiment 1. J1 gives each pin of the board the number of the same pin on the NE555, as Experiment 19 checked. So a lead from each board pin goes into the hole of the NE555 leg with the same number. All the other parts stay where they are. Figure 21-1 gives the circuit.

Schematic. The board is a pale rectangle with eight numbered pins. Pin 8 goes up to the +5 V line and pin 1 down to the ground line. The bench supply, 5.00 V with a limit of 0.05 A, is on the left, between the two lines. RESET, pin 4, goes to the +5 V line. CONTROL, pin 5, goes through a 0.01 µF capacitor to ground. On the right, RA, 10 kΩ, goes from +5 V to DISCHARGE, pin 7. RB, 10 kΩ, goes from DISCHARGE to the node CAP. THRESHOLD, pin 6, and TRIGGER, pin 2, both connect to CAP. C, 10 µF, goes from CAP to ground. OUTPUT, pin 3, goes over the RA wire to the LED at the very right, and the LED goes through 220 Ω to ground.
Figure 21-1: The board in the blinker of Experiment 1. Outside the board are the parts of Experiment 1. Click to enlarge.

This page calls the timing parts RA, RB and C, as Experiment 12 did. Experiment 1 called them R1, R2 and C1, but the board uses those names for its own parts. Figure 21-2 shows the breadboard without the NE555. A green marker is on each hole where a leg of the chip was. The number on the marker is the pin number.

The half-size breadboard of Experiment 1 with the NE555 removed. A dashed outline shows where the chip was, across the centre channel at columns 4 to 7. Green numbered markers sit on the eight holes of the chip's legs: 8, 7, 6 and 5 above the channel in row e, from column 4 to column 7, and 1, 2, 3 and 4 below the channel in row f, from column 4 to column 7. All the other parts are in place: the two 10 kΩ resistors, the 220 Ω, the 0.01 µF capacitor, the LED, the 10 µF capacitor, the green wire over the chip position and the red, blue and yellow wires.
Figure 21-2: The breadboard of Experiment 1 without the NE555. The lead from each board pin goes into the hole with the same number. Click to enlarge.
  1. Disable the supply output. Remove the NE555 from the breadboard.
  2. Push the socket end of a lead onto each pin of the board.
  3. Put the pin end of each lead into its hole, as the table gives. Read the pin name on the board and the number on the marker in Figure 21-2.
  4. Check each lead against the table before you continue.
PinName on the boardHoleConnects to
1GND4fThe ground rail, through the blue wire
2TRIG5fC, and pin 6 through the green wire
3OUT6fThe LED, through the yellow wire
4RST7fThe +5 V rail, through the red wire
5CTRL7eThe 0.01 µF capacitor
6THR6eRB, and pin 2 through the green wire
7DIS5eRA and RB
8+5V4eThe +5 V rail, through the red wire
Photo: the assembled black board on the bench beside a breadboard. Each of its eight upright pins carries a jumper lead with a black socket housing. The leads run in several colours into the breadboard, where the red 10 µF capacitor, resistors and wires of the blinker are visible. The silkscreen names the pins: GND, TRIG, OUT and RST on one side, +5V, DIS, THR and CTRL on the other.
Figure 21-3: My board with a lead on each of its eight pins. The other end of each lead is in the hole of the NE555 leg with the same number.

Switch on with a current limit

Each lead is in its hole. The supply goes on last, with a limit on its current. In constant voltage mode, CV, the supply holds its output at the set voltage. If the circuit takes more than the limit, the supply changes to constant current mode, CC. It then holds the current at the limit and lets the voltage fall. So a short gets no more than the limit, and the voltage across it falls.

  1. With the output disabled, set output 1 to 5.00 V and its current limit to 0.05 A, as in Experiment 1. The supply leads stay on the rails of the breadboard.
  2. Look at the current display. Enable the output.
  3. Read the current. Expect a small current that does not change, about 0.01 A. My board took 0.01 A. I had the limit at 0.10 A.
  4. Check that the display shows CV and 5.00 V. Then look at the LED. Expect it to blink about five times a second.

If the display shows CC, or the voltage falls below 5.00 V, disable the output at once. Do not increase the limit. The limit stopped a fault, and more current can burn the part that has the fault. Find the short under the microscope. Then do the checks of this experiment again from the start.

Measure the board

The supply holds 5.00 V and the LED blinks. Now the oscilloscope measures the board, with the probes and the settings of Experiment 1. Figure 21-4 shows the probe points.

The breadboard of Figure 21-2 in pale grey, with four probe points in colour. The channel 1 tip, yellow, is on hole 6g, in the column of pin 3. The channel 2 tip, magenta, is on hole 5h, in the column of pin 2 and the 10 µF capacitor. The two ground clips, blue, are on the bottom ground rail at columns 10 and 16.
Figure 21-4: The probe points: channel 1 on OUT, channel 2 on CAP, both ground clips on the ground rail. Click to enlarge.
Oscilloscope channelTip or hookGround clip
Channel 1: OUTHole 6g, the column of pin 3The ground rail
Channel 2: CAPHole 5h, the column of pin 2 and CThe same ground rail
  1. Disable the supply output. Attach the probes as the table gives.
  2. Set each probe and each channel to 10X, with DC coupling. Set both channels to 1 V/div and the timebase to 50 ms/div. Put both ground markers near the bottom of the screen.
  3. Set the trigger to Edge, source CH1, rising slope, Auto mode, level 1.3 V. Enable the supply output.
  4. In the Measure menu, add Period for channel 1.
  5. Set the cursor source to CH1. Put Y1 on the low level of OUT and Y2 on its high level. Read both.

The board copies the NE555, so expect the readings of Experiment 1. By calculation, RA, RB and C give a period of 208 ms. OUT is high while C charges and low while C discharges, as on pin 3 of the NE555. Experiment 14 moved Q19's base from Q to Q̄ on the board for that reason.

Oscilloscope screen. Channel 2, magenta, shows CAP: a slow rise and a faster fall. Channel 1, yellow, shows OUT: a square wave that is high during each rise of CAP and low during each fall. The cursor box reads delta X 209 ms, 1 over delta X 4.785 Hz, Y2 1.82 V, Y1 0.06 V. The Period measurement at the bottom reads 210.11 ms.
Figure 21-5: My board at its first power-up. OUT (yellow) is high while CAP (magenta) rises. The Period measurement reads 210.11 ms. OUT is 0.06 V low and 1.82 V high. Click to enlarge.
MeasurementCalculationNE555, Experiment 1My board
Period208 ms200.46 ms210.11 ms
Frequency4.81 Hz4.99 Hz4.76 Hz
OUT while C chargesHighHigh, about 3.7 VHigh, 1.82 V
OUT while C dischargesLowNear 0 V0.06 V

The board runs at 210.11 ms, 1.0 % from the calculation. The phase of OUT is that of pin 3. The CAP trace in Figure 21-5 also jumps by about 1 V at each switch, and the NE555 trace of Experiment 1 does not. I have not found the cause of these jumps yet, so this page does not take the turning points of CAP from this capture. One reading is far from the NE555: the high level of OUT, 1.82 V against about 3.7 V.

Why OUT is high at only 1.82 V

The last section found OUT high at 1.82 V. A bad joint on Q19 could be the cause, but the circuit gives a simpler cause. Q19 can pull OUT down to 0 V, and it cannot pull it up. On the board, R21 pulls OUT up, as Experiment 14 explained. Figure 21-6 shows the output stage while C charges, with the LED on OUT.

Schematic in two pale regions. On the board: Q-bar, at 0 V, goes through R14, 10 kΩ, to the base of Q19, which is off. Q19's emitter goes to ground. R21, 4.7 kΩ, goes from +5 V, 5.00 V, to Q19's collector, which is OUTPUT, at 1.82 V. A green arrow beside R21 shows 0.68 mA. OUTPUT leaves the board at pin 3. On the breadboard: the LED, 1.67 V, and the 220 Ω, 0.15 V, go from OUTPUT to ground, with a green arrow beside them. Green markers 1 to 4 match the steps in the text.
Figure 21-6: The output stage while C charges, with the LED of Experiment 1 on OUT. The numbers match the steps below. Click to enlarge.
  1. Q̄ is 0 while C charges, so Q19 is off. No current flows into its collector.
  2. Current flows from +5 V through R21 into OUT. R21 has 5.00 V − 1.82 V = 3.18 V across it. By Ohm's law it passes 3.18 V / 4.7 kΩ = 0.68 mA.
  3. Q19 is off, so all of that current flows out of pin 3, through the LED and the 220 Ω to ground. The 220 Ω has 0.68 mA × 220 Ω = 0.15 V across it. The LED has the remaining 1.67 V.
  4. OUT is at 1.82 V. An LED conducts from about this voltage, and its voltage changes little when its current changes. So the LED holds OUT at about 1.8 V, and R21 takes the rest of the 5 V.

The NE555 drives pin 3 high through its own transistors. Experiment 14 gave up to 200 mA for pin 3, from the TI datasheet. With that much current available, the LED did not hold pin 3 down, and it stayed at about 3.7 V. R21 gives at most 1.06 mA, the current with OUT at 0 V. So the steps give a prediction. Remove the LED, and nothing takes the current of R21 except the probe. Then OUT rises to almost 5 V.

  1. Disable the supply output. Pull the anode lead of the LED out of the breadboard.
  2. Enable the supply output. Read Y1 and Y2 on channel 1 again, and the Period.
  3. Disable the output, and put the LED back.
Oscilloscope screen with the LED out. Channel 1, yellow, OUT, now switches between the bottom of the screen and near the top. The cursor box reads delta Y 4.98 V, Y2 5.04 V, Y1 0.06 V. The Period measurement reads 210.25 ms. Channel 2, magenta, CAP, has the same shape as before.
Figure 21-7: My board with the LED out. OUT is 0.06 V low and 5.04 V high. The Period measurement reads 210.25 ms. Click to enlarge.

Without the LED, OUT is high at 5.04 V, the supply voltage as this oscilloscope reads it. The low level stays at 0.06 V, where Q19 holds OUT. The period moved from 210.11 ms to 210.25 ms, 0.07 %. So the LED set the high level, and the output stage works. The LED does not change the timing.

The high level of an open collector

On an open collector output, the pull-up resistor and the load set the high level. The resistor and the load are in series between +5 V and ground, so the same current flows through both. The high level is the supply voltage minus the voltage across the resistor:

VOUT = 5.00 V − I × 4.7 kΩ

With no load, I is almost 0 and OUT is at the supply. With the LED, I is 0.68 mA and OUT is at 1.82 V. With OUT shorted to ground, I is 1.06 mA. A smaller pull-up gives more current to the load. It also gives more current into Q19 while Q19 holds OUT low.

Test the reset

The output stage works. One block of the board has done nothing yet: the reset block. In the blinker, the red wire holds RESET at +5 V, so the timer runs and the reset block stays off. A blink that works does not show that reset works. So the test moves RESET to 0 V.

  1. With the LED blinking, pull the end of the red wire of column 7 out of the +5 V rail at the bottom. Push it into the ground rail beside it. RESET is now at 0 V.
  2. Look at the LED. Expect it to go off and to stay off.
  3. Move the end of the red wire back into the +5 V rail. Expect the LED to blink again.

Experiment 14 gave the steps of a reset on the board. Q22 holds Q at 0 and Q23 holds S at 0. Q̄ is then 1, so Q19 holds OUT low and the LED is off. Q20 empties C. In the simulation of Experiment 16, OUT stayed at 0.04 V during the reset, and the first charge after it took 203.8 ms. On my board the LED went off and stayed off while RESET was at 0 V. It blinked again after the release. I checked this by eye only, with no capture. To see the reset on the oscilloscope, expect OUT near 0 V and CAP falling towards 0 V while RESET is low.

Five builds, one timer

The reset worked, so each block of the board has now worked. The comparators, the divider, the latch and the output work in each blink. The discharge switch works in each fall of CAP, and the reset block worked in the test. This board is the last of five builds of the same timer in this course. Figure 21-8 gives the period of each build. Each build used 10 kΩ, 10 kΩ and 10 µF.

Dot chart on one period axis from 188 to 214 ms, with a dashed line at the calculation, 208 ms. Simulation, Experiment 16: 209.3 ms, +0.6 %. NE555 chip, Experiment 1: 200.46 ms, −3.6 %. IC timer, Experiment 2: 199 ms, −4.3 %. Transistor breadboard, Experiment 12: 192.59 ms, −7.4 %. The board, this experiment, in green: 210.11 ms, +1.0 %.
Figure 21-8: The period of each build against the calculation, 208 ms. The board is in green. Click to enlarge.

All five builds are within 8 % of the calculation. Part tolerances can move a period by that much. A typical electrolytic capacitor has a tolerance of ±20 %. The board is nearest to the calculation, at 1.0 %, and 0.4 % from the simulation of its own schematic. The board and the NE555 ran on the same breadboard with the same RA, RB and C. Their periods differ by 4.8 %. These captures alone do not show the cause.

Photo: four builds side by side on a grey surface, each under a card with a number. 1: a breadboard with the NE555, a capacitor, an LED and a few parts. 2: a breadboard with two ICs and some transistors, the timer from IC blocks. 3: a full breadboard with many transistors and wires, the timer from transistors. 4: the black board with its eight pins, the silkscreen names of the blocks, and DISCRETE 555 in the middle.
Figure 21-9: My four builds of the timer on the bench: the NE555 (1), the IC timer (2), the transistor breadboard (3) and the board (4). The simulation is the fifth build.

Quick check

Three questions on this experiment. They ask what to do when the limit stops the current, why OUT is low on the board, and why the reset needs its own test.

1. At switch-on the supply shows CC and the voltage falls. What do you do?

CC means the circuit takes more than the limit. The limit stopped the fault. More current can burn the part that has the fault, so disable the output and find the short.

2. With the LED on OUT, the high level of OUT is 1.82 V. Without the LED, it is 5.04 V. What is the cause?

Q19 cannot pull OUT up. R21 does, through 4.7 kΩ. The LED conducts from about 1.8 V and takes all the current of R21. Without the LED, nothing takes that current, and OUT rises to the supply.

3. Why does RESET need its own test?

The blink uses each block except reset. Only RESET at 0 V shows that Q21 to Q23 hold the timer.

Before moving on

The board passed each test. The supply pins have no short, and the board takes 0.01 A. It blinks the LED of Experiment 1 at 210.11 ms, with OUT high while C charges, and RESET stops it. That is the last step from the breadboard to the board. The Epilogue compares the result with the aims of the Introduction, and gives what comes next.

Photo from above: the black board at the top, with eight jumper leads in loops down to the breadboard of Experiment 1. On the breadboard there is no NE555, and the red LED is lit. Two probe leads, yellow and purple, run from the breadboard to the clips at the bottom left.
Figure 21-10: The breadboard of Experiment 1 with no NE555 on it. The board runs the blinker through its eight leads, and the LED is lit.
Contents All experiments