Printer troubleshooting technician providing live chat help

Joe B.

Expert ★★★★★ 27,853 satisfied customers

Specialities: 3D printers, PSU sizing, heated bed wiring, brown-out resets, Marlin/Klipper/RRF tuning

3D printer reboots on heat — brown-out, undervoltage, voltage drop under load, bed/heater inrush & EMI

Technician avatar Technician
Hey! Sudden restarts when the heated bed or hotend fires usually mean a power supply sag (undervoltage/brown-out) or a wiring/connector issue causing a drop. Quick triage:
• Printer & board: model, 12V or 24V, PSU wattage/brand (e.g., Mean Well 24V/350W).
• What reboots first — the screen, the USB host (OctoPrint), or the whole MCU? Any “brown-out reset” or watchdog in logs?
• Does it happen only when the bed heats, or also when the hotend cycles (PID)? Any smell of hot plastic at the bed terminals / XT60 / screw blocks?
Customer
Ender-3 on 24V. Stock PSU says 24V/240W. It reboots the moment the heatbed starts. With nozzle-only heating it’s fine. LCD blanks, fans stop, then it comes back. Happens mid-print too.
Technician avatar Technician
Classic VIN dip. The bed pulls 8–12A on 24V (even more on 12V systems). If the PSU can’t deliver or you have resistance at terminals (loose screws, burnt connector, skinny wire), input voltage sags and the board’s brown-out detection resets the MCU. We’ll run a PSU sag test, then fix wiring/PSU sizing/PID to stop it.
Customer
How do I do that sag test safely? I’ve got a basic multimeter and OctoPrint.
Technician avatar Technician
Step 1 — PSU sag test (no printing, bed only)
1) Safety: Power off, remove mains if you need to access terminals. Set your DMM to DC volts.
2) Clip probes to the controller VIN (board + and –) or the PSU 24V output. Ensure the probes are secure — no slips.
3) Power on. In your host (screen/OctoPrint), preheat bed only to 60–70 °C. Watch the meter at heater switch-on and during PID pulsing.
4) Healthy 24V systems should stay near 23.5–24.2 V with tiny ripple. Dips below ~22.5 V (or below the MCU’s BOD threshold) indicate PSU/wiring problems.
5) Repeat with hotend only. If hotend is stable but bed isn’t, the bed circuit is the culprit (PSU current headroom, bed wiring, terminals, MOSFET/SSR).
Customer
OK — I see 24.1 V idle, but when the bed turns on it drops to about 21.9 V then the screen dies.
Technician avatar Technician
That’s a definite undervoltage/brown-out. Next, fix the highest-impact, lowest-risk items:
• Terminals: Power off, inspect the PSU and mainboard screw blocks, bed terminals, XT60s. Look for browning, melted plastic, or loose screws. Reseat and re-torque gently.
• Wire gauge & length: Use at least AWG16–14 for the bed on 24V (AWG14–12 on long runs/12V). Replace crispy or tinned wire ends with ferrules.
• Bed input connector: Many boards burn these over time. Consider an external MOSFET module or SSR to offload current from the mainboard if the connector shows heat damage.
• PSU capacity: Bed (e.g., 220–300 W) + hotend + motors + electronics should be below ~80% of PSU rated wattage for headroom.
Customer
The PSU label says 240W. Bed sticker says 220W. That’s already close. Would a 350W PSU fix it?
Technician avatar Technician
Likely yes. Rule of thumb: sum loads × 1.25 (20–30% headroom). For a 220 W bed + 40 W hotend + 40–60 W for motors/electronics, a 24V/350W (14.6A) quality PSU gives breathing room and better regulation. Alternatives:
• Run the bed from a dedicated PSU or AC silicone bed + SSR (observe mains safety).
• If staying DC, ensure line voltage selector (115/230V) is correct and the PSU isn’t a counterfeit/weak unit.
Customer
Could firmware cause this? It started after I enabled PID for the bed.
Technician avatar Technician
Firmware can expose a weak PSU. Bed PID pulses the load rapidly; marginal supplies or bad connectors dip more vs. slow bang-bang. Do this:
• Temporarily switch bed to bang-bang and re-test. If resets stop, address PSU/wiring then return to PID.
• Keep brown-out detection enabled; it protects your board. Update firmware only after electrical fixes.
• Route bed leads away from endstop/thermistor lines; add ferrite on LCD/USB if you suspect EMI.
Customer
Any way to monitor VIN without a meter? I’m on Marlin; OctoPrint connected.
Technician avatar Technician
Many boards expose VIN to firmware. Options:
• Marlin LCD often shows V in the info screen; some builds report it via host.
• On RepRapFirmware, M122 reports VIN min during the session (useful to catch sags).
• OctoPrint plugins and some Klipper setups can log supply voltage. Still, a DMM at the board is the gold standard for fast dips.
Customer
I tightened the terminals and swapped to a 350W PSU. Now the sag is 23.8 V→23.3 V and no reboots while heating.
Technician avatar Technician
Perfect. For long-term stability:
• Run a bed PID tune after hardware fixes. Keep bed bang-bang hysteresis modest if you prefer less ripple.
• Re-crimp with ferrules, zip-tie strain relief, and re-check torque after a few heat cycles.
• If using an AC bed + SSR, choose a zero-cross triac SSR rated for the load, on a heatsink, in an enclosed, grounded box.
Customer
Rock solid now. I can heat to 100 °C without a hiccup. Thanks for the help!
Technician avatar Technician
Summary: If the printer restarts when the heater kicks, verify with a PSU sag test. Fix VIN drops by tightening/rewiring bed power, using proper gauge and ferrules, offloading current via MOSFET/SSR, and upgrading to a quality PSU with 20–30% headroom. Keep BOD enabled. Need help reading your meter or choosing a PSU? Open the chat.

Quick answer: the accepted fix

Accepted fix when the printer reboots as the bed/hotend turns on: measure VIN during heat to confirm a PSU sag, tighten/replace bed power connectors, use correct wire gauge/ferrules, consider an external MOSFET/SSR, and upgrade to a quality PSU with 20–30% headroom. Re-tune PID and keep brown-out detection enabled.


Why heaters trigger resets (current draw, brown-out & connectors)

Heated beds draw large inrush/current. Weak or counterfeit PSUs, loose screw terminals, undersized wiring, or burnt XT60/screw blocks cause voltage drop. The controller’s brown-out detection resets the MCU to avoid corruption. PID pulsing can accentuate ripple; PSU regulation and solid wiring eliminate the dip.

How to verify the fix: repeat the sag test

With the new PSU/wiring, repeat the sag test at the board VIN. Expect <3% drop from idle to full bed load and stable voltage during PID. Log VIN if your firmware/host supports it; on RepRapFirmware, check VIN min after a heat cycle.

Power stability — quick troubleshooting tips

Most sudden reboots are undervoltage from PSU headroom or connectors.

Quick path: run a PSU sag test → re-terminate bed power with ferrules → check terminal heat/browning → upgrade PSU (quality 24V/350W typical) or move to AC bed + SSR with proper mains safety → re-tune PID and keep BOD enabled.

Disclaimer: Mains wiring can be hazardous. If you’re not comfortable with AC work, consult a qualified technician. For device-specific guidance, chat with a verified expert on Whizz-Tech.


Getting Started Is Easy

Ask for help, 24/7

Members enjoy round-the-clock access to 12,000+ verified Experts, including tech support, mechanics, vets, and more.

Expert will respond in minutes

We match you with an Expert who specializes in your situation. Talk, text, or chat — whichever you prefer.

Save time & money

No scheduling hassles or expensive consults. Membership can save you significant time and money each month.

FAQ

Why does the printer restart when the heated bed turns on?

The bed draws high current; weak PSUs, loose or undersized wiring, or burnt connectors cause a VIN drop. The controller hits its brown-out threshold and resets.

Will a bigger PSU always fix it?

Not if the connectors are bad. First fix wiring/terminals; then choose a quality PSU with 20–30% headroom. AC beds with SSR are another option (mains safety required).

PID vs bang-bang — which is safer for marginal power?

Bang-bang has longer on/off cycles and can hide minor sags, but proper fix is stable PSU + sound wiring. After hardware fixes, PID gives steadier temps and better prints.

How can I log undervoltage events?

Some firmware reports VIN on the LCD/host; RepRapFirmware shows VIN min via diagnostics. External logging with a DMM or meter module captures fast drops best.

Sources

Matt Verified Icon
Matt ● replies in < 1 min
Technician Expert
⭐⭐⭐⭐⭐ Trusted by 27,000+ customers
Avatar
Alex · Expert's Assistant