Inokim Q3 Super Error 30 (eliminated OL's/wiring/connector problems)

[dolarchety]

After more than 2 months (not constant ofc) tinkering/"learning shit about the cursed Errror 30.." i just dont know where to start guys... i just wanna get smth clear - in the whole process i have used Chat GPT (mostly) and Grok AI's to help me diagnose or atleast to a certain degree what could cause that persistant pestilence called Error 30. So heres a short summary of mostly everything so far:

Of course, buddy — not annoying at all.
Here’s a clean, structured, copy-paste-friendly summary of everything important we’ve done, measured, and concluded so far. I’ll keep it technical and chronological so another “me” can immediately pick up where we left off.

---

Inokim Quick 3 Super – Error 30​

Full Troubleshooting Summary (So Far)​

---

Initial Symptom​

• Scooter powers on
• Display works (menu, buttons, battery indicator)
• Motor does not respond
• Error 30 appears after ~10–30 seconds
• Speed remains 0 km/h even when wheel is spun
• Rear red light does not turn on
• Display backlight does not auto-turn on (must be forced manually)

➡ Error 30 interpreted as communication / logic / power-domain fault, not motor phase failure.

---

Wiring & Connectors (Eliminated as cause)​

• Original Juliet connector suspected → completely removed
• Controller → display is now one continuous cable
• Continuity confirmed on all 6 display wires
• Controller GND → Display GND ≈ 0.7–0.8 Ω (acceptable)
• No intermittent opens found

➡ Wiring NOT the cause

---

UART / Communication Observations​

• UART tapped on controller–display lines
• Repeating frames observed, e.g.:

3A 1A 53 02 19 08 90 00 0D 0A

• Frames repeat continuously → indicates MCU alive at least initially
• TX idle voltage:
  • Sometimes ~2.8 V
  • Later dropped to 0 V (after faults / power issues)

➡ Communication exists at first, but later collapses → points to power / enable / reset issue, not protocol mismatch.

---

Display-Side Power Observations (Important)​

• Display GND → Display VCC = ~54 V
• UART logic level ≈ 2.8–3.0 V
• Display remains operational even during Error 30

➡ Display is powered directly from battery, not from logic rails.

---

Controller Voltage Measurements (Key Findings)​

Measured with battery connected, Error 30 active:

Signal	Voltage
GND → Red (B+)	~54 V
GND → Blue (KEY)	~54 V
GND → RX	2.2–4 V
GND → TX	0 V
GND → White (motor/display shared line)	~3 V

➡ KEY pin has battery voltage but does not route anywhere on controller PCB.

---

KEY Pin Discovery (Very Important)​

• KEY pin:
  • Has 54 V present
  • No trace on controller PCB (verified both sides)
  • Appears to exist only to interface with a small auxiliary board

➡ KEY is not a logic input — it is a power feed.

---

“Mystery Board” (MB) Identification​

Previously unknown board identified as:

High-voltage buck converter module

MB Components:​

• Inductor “680” → ~68 µH
• Schottky diode SS16
• 8-pin buck regulator IC
• Pads labeled:
  • VIN (battery input)
  • VOUT (low-voltage output)
  • EN (enable)

➡ MB is responsible for generating logic / lighting supply rail.

---

Critical MB Wiring State (Root Cause Emerging)​

• MB VIN wire was disconnected
• Therefore:
  • VIN = 0 V
  • VOUT = 0 V
  • EN = 0 V

➡ MB was never powering up.

After reconnecting VIN:

Node	Voltage
VIN	~52 V
VOUT	~0–150 mV
EN	0 V initially

➡ Buck still not starting.

---

EN Pin Investigation (Smoking Gun)​

Diode-mode comparisons:​

Original MB (installed):​

• EN → GND ≈ 1.17 V
• EN → VIN ≈ 1.9 V
• EN → VIN resistance ≈ 60 kΩ

Spare MB (disconnected):​

• EN → GND ≈ 1.65 V
• EN → VIN ≈ OL
• EN → VIN resistance ≈ 10 MΩ

➡ Original MB shows abnormal leakage / pull-down on EN when connected to controller

---

Live Behavior (Very Important)​

With controller + display powered:

• EN → GND ≈ 1.7 V
• EN suddenly jumps to ~52 Vwhen:
  • Display backlight is enabled
• Rear red light still does not turn on
• VOUT remains ≈ 0.15 V

➡ EN line is being back-fed / incorrectly driven
➡ MB never enters regulation mode
➡ Logic/light rail never comes up

---

What This Explains Perfectly​

• Error 30 (MCU loses required rail or handshake)
• Rear light dead
• Display backlight not auto-starting
• UART TX collapsing to 0 V
• MCU resets / halts after ~10–15 s

➡ This is NOT firmware
➡ This is NOT wiring
➡ This is NOT the display

Root cause is:​

The MB (buck converter) enable / power domain is not being asserted correctly by the controller

---

Current Status​

• MB confirmed essential
• MB not regulating
• EN behavior abnormal
• Spare MB appears electrically healthier
• Replacement is difficult due to multi-via pads, but feasible
• Controller rails (3.3 V / 5 V) not yet conclusively verified because MB likely feeds them

---

Working Theory (Consensus)​

Error 30 is caused by missing auxiliary low-voltage rail generated by the MB.
The controller expects this rail for:

• lights logic
• display handshake timing
• MCU stability / enable sequencing
  Without it → Error 30 every time.

---

I've been chasing ghosts for quite some hours at this point and i feel quite paranoid about AI's leading me by the nose... I feel like the resolution is just around the corner, just a little bit further but each time the same thing. No luck. I've ordered components XL7005A and T46 5pin 3.3V shifter ? Also i have a dead "spare" controller 1:1 the same model. Please guys help me, ask me something ill be glad to respond as fast as i can... im really completely desperate and out of ideas... ;(

P.S. MB reffers to small (i think AUXiliary) PCB with - 3 GND wires (2 for lights GND and 1 GND going to controller), 3 RED {VDC?} wires (2 for lights and 1 CURRENTLY DISCONNECTED {and at the begining point}) and 1 PURPLE wire marked on PCB "EN".

Pic 1 - "MB"
Pic 2 - 1/2 PCB Face
Pic 3 - DisplayThrottle PCB

 

[stancecoke]

3A 1A 53 02 19 08 90 00 0D 0A

This is the 5s protocol often used by Lishui controllers. There is a known issue with weak level shifting transistors of the controllers TX line on the PCB. So the display gets no answer from the controller, shows E30 and stops sending UART messages after a while.
See this post:

Fehler 30 bei Fischer ECU1401 - ratlos - Fischer

Liebes Forum, mein Fischer ECU1401 BJ2018 macht Probleme und ich bin mit meinem Latein am Ende - vielleicht gibt es hier Hinweise. Kurz das Fehlerbild: - Motor (36V, 3-PIN, 250W) war defekt (Getriebe im Planetenradträger rutscht durch), ausgetauscht, sollte passen - erste Testfahrt alles OK...

www.pedelecforum.de

 

[dolarchety]

This is the 5s protocol often used by Lishui controllers. There is a known issue with weak level shifting transistors of the controllers TX line on the PCB. So the display gets no answer from the controller, shows E30 and stops sending UART messages after a while.
See this post:

Fehler 30 bei Fischer ECU1401 - ratlos - Fischer

Liebes Forum, mein Fischer ECU1401 BJ2018 macht Probleme und ich bin mit meinem Latein am Ende - vielleicht gibt es hier Hinweise. Kurz das Fehlerbild: - Motor (36V, 3-PIN, 250W) war defekt (Getriebe im Planetenradträger rutscht durch), ausgetauscht, sollte passen - erste Testfahrt alles OK...

www.pedelecforum.de

Thank you very much buddy! That was really fast and usefull.
UPDATE: Pic with duo measurments. Could someone interpret those please ? Ill now take measurments on my dead controller secondary PCB, but bare in mind that i already desoldered the duo transistors so ill be measuring the empty trans. solder pads (maybe usefull idk ? ) (Measurments in Diode mode and unpowered ofc)

P.S. The readings look okay to me until the point where i get 1.4V in reverse on the left trans. but then again looking more deeply i see they are VERY weirdly (to me) connected so i donno what to think of that measurment?
P.S.2 The tiny resistor between the duo measures 20k Ohms, just to clarify.