KAREN'S BED // REV 1.0 // ON-SITE EXECUTION CHECKLIST
0 / 6 STEPSH1 · 95–98% CONFIDENCE
Working hypothesis: Pigtail solder joint at the HJH5 PCB cracked or went high-resistance in transit. Joint passes continuity (~1 mA) but can't pull the controller input low enough under a real pull-up. Both Claude + ChatGPT independently converge.Fix is a reflow + strain relief.
Tools // tap to check off
Steps // tap header to expand · tap box to mark done
01
Fast Isolation Test 30 seconds · decides everything
▸
At the relay terminals, take a short jumper wire and manually short COM1 directly to NO1 — relay bypassed entirely.
✓ BED MOVESRelay or power issue. Skip to Step 3 reflow only if the symptom returns. First try: swap USB brick, re-check Momentary time in Smart Life.
✗ NO MOVEMENTConfirms the pigtail / PCB path is broken. Continue to Step 2 or jump straight to Step 3 reflow.
Do this before opening the wand. Zero tools needed beyond a jumper.
02
Voltage Sanity Check optional · diagnostic gold
▸
Meter in DC Volts mode. Probe negative on GND / Pin 3, positive on the HU pigtail.
Idle: ~5 V (or whatever VDD is on this controller)
During relay trigger: should snap to ~0 V
Smoking gun: if trigger drops HU to 1–3 V instead of 0 V, confirms high-resistance joint. Controller can't see LOW, ignores the press. Go reflow.
If it hits 0 V cleanly, the joint is fine and the problem is elsewhere — but this is unlikely given symptoms.
03
Reflow the Three Joints HU pad · HD pad · GND pad
▸
Open the wand shell. Hit each of the 3 pigtail joints with flux + fresh 60/40 solder. Wick off the old joint first if it looks dull or grey.
Look for cold joints (dull grey, lumpy)
Look for hairline cracks around the wire root
Look for pad lift (pad separated from PCB)
If pad lifted: don't reuse it. Scrape solder mask off the trace a few mm back and solder to the bare trace instead. Anchor with hot glue.
04
Validate with Resistance Mode Ω · NOT continuity · target < 0.5 Ω
▸
Meter in resistance (Ω) mode — the beep mode lies here. Probe from the far end of each pigtail to its PCB pad.
Reading
Verdict
< 0.5 Ω
✓ Good. Move on.
0.5 – 2 Ω
⚠ Marginal. Reflow again.
> 2 Ω
✗ Bad. Redo from scratch.
Short your probes together first and subtract that reading — cheap meters show ~0.2–0.5 Ω of probe resistance.
05
Strain Relief the thing that actually prevents a repeat
▸
Hot glue or 2-part epoxy over:
Each of the 3 solder joints
The first ~1 cm of wire immediately after the joint
Also: tie down the cable bundle inside the wand shell so flex doesn't transfer to the pads.
This is the step that prevents a second trip. The joints don't fail on their own — they fail because the wire flexes and snaps them. Lock the flex out.
06
Final Test app → bed moves
▸
Close up the wand, reconnect to bed, plug relay into USB.
Tap Up in Smart Life → bed raises briefly → releases
Tap Down in Smart Life → bed lowers briefly → releases
Press manual HU / HD buttons → still works normally
All three pass = mission complete. Have Karen test from her phone before you leave.
If joints fail again // future-proof option
REV 3.0 — cable-side tap. Instead of soldering onto fragile SMD button pads, splice into the HU, HD, and GND wires in the cable itself (thicker insulation, stronger joints, no tiny pads). Tradeoff is a visible splice location. Not needed unless this reflow fails. Ping Claude if you want the HTML wire map built out.