The L19 Lab puts Lesson 19 on the bench as a fox hunt. A taped firing line runs along the table, and numbered boxes sit downrange of it at positions I measured and posted. Every box holds an HB100, powered or not, so weight gives nothing away; one of them is live.

Your team moves its kit along the firing line, stopping at the station marks taped every 0.75 m. At each station the kit faces straight downrange, square to the tape, so every station shares one heading and the only thing that changes is where along the line you stand. At each station you read the beam sweep’s Angle readout (set to Peak) and Peak Gain. Each angle is a line from that station toward the fox; the lines cross at the fox, and the box nearest the crossing is your call. Peak Gain plays a smaller part: when two boxes sit close to the crossing, the pattern of levels along the line says which one fits.

Steps 1 and 2 happen before the fox goes on: the bring-up on your own HB100, then the reference shot that fixes which end of the tape positive Angle points to. Steps 3 and 4 are the hunt: the same measurement at every station mark, then the fix. Step 5 is mine, and the Part 3 writeup starts from it.

The range fits a classroom. The firing line runs 0.5 m past the box field at each end, so a box near the edge still sits within about ±45° of several stations, the edge of what the PHASER can track. The nearest boxes are 1.0 m out, past the 0.88 m far-field distance of Lesson 17, and the farthest are 2.0 m out, where the tone is still well above the floor.

These sizes come from a simulation of the hunt with a bearing error of 2.5°, a bearing off by 10° one time in seven, and an HB100 aimed only roughly at the line: the call names the right box about 86% of the time, and the median fix lands 0.15 m from the fox. A wider room helps less than you might expect; a deeper field hurts more, because depth is the direction a single firing line measures worst.

The Pre-Lab is Part 1 of the packet, done before the lab period and listed at the end of Lesson 19. What you need at the bench is the notebook, L19_foxhunt.ipynb: it holds the practice set, draws every bearing line over the boxes, and prints the results. Write its four functions, run its test cell until every check passes, and then run it on the practice set, which has one bad station in it. The error budget and station plan in Part 1 tell you what to expect: use the stations farthest apart that still detect the fox, because the distance between them sets how well the fix is pinned down in depth.

I tape the firing line and mark station 0 at its middle, with stations at \(x = -1.5, -0.75, 0, +0.75, +1.5\ \text{m}\), and post the measured center of every box in the same frame. Place the middle of the array, between E4 and E5, on the station mark, and square the board’s patch face to the tape with a straightedge or a square, so broadside points straight downrange.

Because every station faces the same way, there is no heading to work out. Est. Angle \(\theta_i\) at station \(x_i\) says the fox lies on the line \(x = x_i + y\tan\theta_i\), with \(\theta_i\) positive toward the \(+x\) end of the tape. Whether positive Angle points to \(+x\) depends on which way round the board sits on its stand, which is what the reference shot checks.

The sweep jitters a little from one pass to the next, so read Angle on ten consecutive sweeps from the Tracking tab and use their mean. The spread of those ten, largest minus smallest, should be a degree or two.

An HB100 radiates mainly forward, so expect two to four of the stations to detect it; a station far off its axis may see a weak tone or none, and one that reads beyond ±45° is past what the PHASER tracks. Skip those and keep going. A good bearing passes within about 0.15 m of the fix; one that misses by far more than the rest is a bad bearing, usually a reflection or a person in the path: retake it, or let your notebook drop it. The fix should land 0.15 to 0.2 m from the live box, and the boxes are at least 0.75 m apart, so that is close enough to name the right one.

Peak Gain is useful only when Rx Gain (dB) is the same at every station. Your notebook compares the levels along the line with the shape a source in each candidate box would produce, free-space spreading plus the element roll-off you measured in the L18 Lab, with the overall level fitted out. It cannot give a range on its own: the fox’s aim and the room’s reflections move the level by several dB.

Part 3 is the writeup, done as a team after the reveal: how far your fix landed from the live box, the error of each bearing, and what the baseline and the depth did to the fix.

Lesson 20 returns to shaping and steering the beam, starting from the closed-form array factor: where the nulls fall, why the first sidelobe sits 12.8 dB down, and how the beamwidth depends on the number of elements.