E-Bike Tester Tear Down

by TheRadMan in Circuits > Electronics

4038 Views, 6 Favorites, 0 Comments

E-Bike Tester Tear Down

TearDown_BoAiZhi_Ebike-Tester_002.png
TEST_LEADS_and_CLIPS.png
Cable_Functions_Case_side.png
removerearscrews.png
desolder_power_switch.png
TearDown_BoAiZhi_Ebike-Tester_001.png

I obtained an Electric Bicycle tester (image1) from an Alibaba merchant site, ordering it 11 weeks ago for $28 USd. I thought I ordered a ROMAI brand, and received Bo Ai Zhi (image2).

Removing four 5mm screws from rear, and opening case revealed image3 and required desoldering the low quality power switch (image4) to free the board for inspection, per image5.

The Tester indicates using 15 LEDs in five distinct stages.

The ABC/CBA LEDs are actual Wye Loaded bridge to interpret the 3 phase drive _from_ a Controllers output, perhaps the most important test. However, to 'fool' the Controller into thinking the Hub Motor is 'rolling', a second stage using the common 5pin HALL EFFECT DETECTOR connection, an internal unidentified IC simulates the three Hall Phase signals back to the controller. Its a very slow phase 3 step sequence Hb+Hc, Ha+Hc, and Ha+Hb, repeating at 1-2 Hertz rate.

There are three Hall Efect Device output LEDs and the Tester provides +5V from an internal 78L05 regulator. There are another three LEDs to test HUB windings, taking advantage of 'generation' (ala Nikola Tesla) of back EMF from a HUB to verify Xa, Xb and Xc windings.

And finally, a simple continuity circuit to test THROTTLE potentiometer (normally 0.7 to 4.3V) using LED brightness approximation. As a bonus, two of the wires can clip into BRAKE switch to test the safety cutoff as well. The design adds a LED to indicate +5V low voltage feed from the controller under test as well as a low 9V NEDA1604 battery level (threshold about 8.3V).

So on second task with box, the Hall connector crimp for Blue 'opened', and during the teardown I found numerous cold soldering and low quality board issues which all had to be fixed.

Supplies

BO AI ZHI or ROMAI E-Bike tester device

5mm Cross Screwdriver

silicon worksheet

soldering pencil and braid of vacpump to desolder Power Switch SPST 2 blade

Disassembly

IMG_20210220_222628021.jpg
rear_of_PCB_002.png
BOAIZHITESTER_FACE_LEDS.jpg
Component_Side_001.png

Removed 4 screws 5mm cross blade Philips

removed rear cover case

unsolder two blade power switch with minimum heat (switch is the inexpensive type)

use magnifier and camera to document

EBike_TESTER_Schematic.png

Follow all circuit paths and document using Camera and light table

Capture circuit to Schematic using KiCAD (free at www.kicad.org)

Notes:

There is an 8pin IC which has no house marking, on this board. Its purpose is to accept external +5V power _from_ controller under test, and create a johnson counter like overlapping strobe pattern such as the kind that 3 Hall detectors create when the E-Bike Hub Motor 'spins'. Pin 1 accepts +5, pin 8 is the return. Pin2 is the "Yellow (W)" , pin3 is "Green(V)" and Pin4 is "Blue(U)" signal output, which looks like a slow squared pulse at 1-2 times per second (Hertz). You can "loop" the output back to the other connect by joining them and you can see the sequence on the testers LEDs. Leave a comment if you can guess what IC this is, and I think the manufacturer created these custom 3phase generation ICs (so they may not be available commercially). It is very probable that IC can be functionally replicated using ATtiny85 in self clock RCmode in 8pinDIP.

Power +5Volt is regulated from 78L05 in TO92 package so very common, but there are no protections if a Controller low voltage regulation is defective and outputs 12V then this 9V component will burn inside this low quality tester.

PCB and and soldering is very low in quality so expect to 'fix' items here and there.

High efficiency RED LEDs should be used with low Vfwd and water-clear lens in 5mm T1-3/4 size. The resistors were all 1% and only the 3phase Controller output bridge is 1/2watt; all the rest of the resistors are 1/4W. Load impedance of 2K Ohms on Controller FET output bank is surprisingly light. I have used turn signal 12V incandescent bulbs right up to 2.4Amp Halogen 12V bulbs in Wye-Delta configuration, as loads for E-Bike controller testing.

I publish my captured artwork here.

I think this is the only schematic of an E-Bike tester anywhere on the NET although many sites have all of these individual stages, and theory well covered. The circuitry is not patentable, and certainly not unique.

The tool is remarkably simple, and surprisingly useful to Tech's or amateur's that service Electric Bicycles.

A pending Instructible will use the concept of all 5 stages described here, adding a 6th context, using Arduino NANO and LCD for display.