Shimano E8000 - E010 / E01020 / E01021 madness

[th3c0d3r]

First of all this might sound like a vent (and maybe it is) but this is so damn frustrating.
We own 2 Merida eONE-TWENTY bikes:
- 1st one bought new in 2017 so the missus could keep up with me on the climbs
- 2nd one bought last year since I got fat and lazy , it's was almost new, had around 200km

This spring, around March, after it sat the entire winter, I've serviced the wife's bike fork, shock, brakes, etc, we're all equipped and ready to go, put the battery in, start the bike, boom! E010 on display, I didn't know but this would be the first step in the tumble down the rabbit's hole called "Shimano is kicking their own customers in the nuts". The back was out of warranty, of course, after I tried everything I found online I've sent the motor to the LBS I bought it from (not so "local", it is 250km away) and managed by a small miracle so find a 2 second hand motors and we rode all summer.
The motor came back from the LBS and the end of August, they've sent it to Shimano and their verdict was "it's dead" without any kind of other explanation and they offered me a new one for 800eur, no thanks.

About 6 weeks ago I was pushing my bike (not riding it) on some uncleared forest paths and it also gave the E010 error, it gave me the chills, but it cleared after restart, it appeared again, then it magically cleared and I've put it out of my mind.
Two weeks ago I was riding, luckily I've just climbed a hill and the bike gave the E010 error and stopped, couldn't resurrect it so I went home.
As I still had a spare motor, I quickly plugged in it to see if the motor was the issue, once that was confirmed I postponed the replacement.
Later that day I want to see what the error was and I plugged in my laptop (I have the PCE02 interface) and it had a bunch of "E01021" errors with no date & time in the log, I've forcibly upgraded the firmware (it was already 4.9.0) and the error was gone, hooray!
Fast-forward this Saturday, went for a test run, everything was fine, after about 1 hour of riding I was resting on top of a hill & taking pictures, the bike was 3 meters away from me when I heard the dreaded "beep,beep,beep", E010 again, went home, hooked it to the laptop, a bunch of dateless "E01021" and 1 "E01020" with the time and date, nothing I did erased the error.
Sunday morning I've quickly swapped the motor with the last spare as some friends came to go on a longer ride, everything was OK, rode about 40 minutes on a paved downhill I heard the "beep,beep,beep", stopped and started swearing, restarted the bike, the error vanished, after about 15 more minutes the bike stopped and it wouldn't start.
Again, home, laptop, it has a bunch of "E01021" errors with no time & date, upgraded the firmware several times, no luck, tested with the working motor from my wife's bike, it works. So I've just plugged everything back in and let it there for about 2 hrs, after that time, I wanted to remove the battery and put the bike in the garage and decided to start it one last time, and it starts ... and it works without any logical explanation.

Now I'm stuck with 2 dead motors (both with E01020 + E01021 errors), 1 motor (E01021) is working but who knows for how long, this is what I've done over the course of 6 months:

- replacement motors had NO ERRORS when I tested and upgraded the firmware to 4.9.0
- cleaned and cleaned and cleaned all the connections - NO CHANGE
- checked ALL the cables with a multi-meter and rattled them - NO ERROR
- swapped the batteries between the 2 bikes - NO CHANGE

I'm out of ideas, I don't want to spend 800euros on a new motor, I can't sell my bike to get a new one since I don't know for how long this motor will keep working or why it wasn't working, I can't use my bike for longer rides since I don't know where will it stop.
After seeing that the new EP8 motor failed when it was reviewed with the error E01020 and it had to be replaced, it's safe to say that I will not spend my money on a Shimano powered ebike anytime soon.

Thanks for reading the saga

 

[R120]

Have you checked the alignment of the speed sensor with the magnet? If the wheel is slightly out of true it can take the magnet out of tolerance with the sensor, and cause errors. I had a similar issue on one of mine and it turned out to be the speed sensor on the chianstay needed a washer behind it to decrease the distance to the magnet, and it solved the problem.

I would do a cable by cable check over the whole bike, sometimes simple things like a loose connection, the battery mount being slightly loose et etc can cause the error.

This thread has the relevant manuals

Official Shimano Service Manual, and User Manuals for all Steps EMTB motors

Here is a link to the official service manual for all the Steps System's, a useful resource for bodging and fixing https://si.shimano.com/pdfs/sm/SM-SHIMANO_STEPS_US-000.pdf This is the E7000 user manual: https://si.shimano.com/pdfs/um/UM-79E0A-002-00-ENG.pdf And the E8000...

www.emtbforums.com

 

[th3c0d3r]

Yep, checked the speed sensor and replaced it since the bike I bought had a cheap derestrict which I removed and soldered the wires and thought that maybe it's that so replaced the speed sensor with a brand new one.
Also checked all the wiring with a multi-meter while rattling them to simulate riding, no luck.

 

[Spiff]

Check for pinched DI2 cables in the frame, unplug the DI2 connectors from display and motor and clean plug and socket with compressed air and contact cleaner spray. Connect them in place using the Shimano tool TL-EW02 and make sure you hear the "CLICK" when it engages.

In case of doubt, replace the cable from display to motor.

Good Luck!!!

 

[th3c0d3r]

@Spiff thank you, did that too, buying a new display to motor cable was one of the 1st things I did with no luck.

 

[th3c0d3r]

So, the motor had the E01021 but it was still running, nevertheless:

- I changed the cable between the motor and the display
- reseated the speed sensor
- changed the ports in the display, used other ports
- cleaned all connectors again and again

Took the bike for a spin, after 35 minutes when I was pushing it uphill it errored again, now the error is E01020 which, in my experience, means that the motor is dead. I'm the proud owner of THREE dead Shimano E8000 motors and an unusable bike.
I have a feeling that if I'd buy a brand new motor it will also die as this doesn't seem to be something related to the motors, besides buying new motor, display and shifter I don't know what to do, maybe throw the bike in a lake.

 

[Spiff]

I'd try to replace the cable battery to motor, and check it for corrosión.

If this connector is bad it could damage any motor plugged to it.

And may be try another battery. May be the failure is in it and you have three good motors.

 

[th3c0d3r]

This time I was with the battery from the other bike, the cable between the battery and motor was tested, all connectors cleaned.

 

[Spiff]

This time I was with the battery from the other bike, the cable between the battery and motor was tested, all connectors cleaned.

This cable could be the issue, even if it looks fine. I'd try it before throwing the bike into a lake.

 

[th3c0d3r]

So, I've resurrected the motor (again), it's still spitting E01020 errors BUT if I restart the bike the error is gone which means I'm getting somewhere .
It appears that downgrading the firmware, first to 4.8.0, then to 4.7.1 makes it a bit more "error" friendly, now the bike works, question is for how long.

@Spiff I see no other way of checking/testing/whatever then cable besides what I have already done, so the option would be to buy a new one and see if this still happens?

The latest E01020 errors I've got were with the bike motionless on the bike stand behind me which makes me think it's either the battery to the motor cable or the display or the shifter.

 

[Spiff]

My bet is for the battery to motor cable. It has two thin wires for communication that if fail might cause this error.

The cables to display and shifter are the standard DI2 cables which are more robust.

 

[th3c0d3r]

@Spiff thx for the help, I'll order a new cable.

 

[Spiff]

@Spiff thx for the help, I'll order a new cable.

Good luck!!!

 

[th3c0d3r]

So, after a 20 mins ride (the dog ran away and I had to chase him uphill) Friday the motor died today after 1h ride, at least it had the decency to die almost on top of the last hill.
The error today was E010, then W013, then E010 which I cleared with ST Unlocker by changing the chairing size but will most probably fail again.
So far I have changed:
- battery to motor cabling
- speed sensor
- motor to display cable
Also tried the battery from the other bike with the same results.
All the connections are checked and re-checked and cleaned.
Only thing I haven't changed is the display, I'm not happy to throw another 150euros to this bike but I'll do it if it could be the culprit.

 

[Spiff]

Hi, I am sorry that you are still having issues with your motor. First I have to say that all my assumptions are deducted from the reading in forums, I have not hands on experience repairing motors, so I might be wrong.

Initially you had Error Code E10, which means:

• E010 – System error detected

And this looks like a generic error that could be caused by different problems like cabling issue, battery not making good contact. So you did the right thing by replacing display cable and battery cable. But your errors E01020 + E01021 are not in the Shimano documentation so I do not know if they are the same as E10.


But yesterday you had Error Code W013, which means:

• W013 – Initialization of torque sensor not completed successfully

And it could be caused by moving pedals at power on or defective torque sensor inside motor (this means replacing the motor). But being optimistic, if there was a power interruption that caused E10 it might have caused the W013, although it is not normal, I suppose it may happen.

Another try could be to use conductive grease (But not WD40 or similar Dielectric lubricantes) in the battery and motor connections:

In the thread below, some people fixed the issue by replacing the motor, but looks like others fixed it with conductive grease:

E010 followed by W013

E8000 motor is now 6 months old and done approximately 1200 miles and recently I have experienced a series of E010 errors and finally to a complete shutdown and constant W013 warning. First time was about 2 weeks ago, after riding all morning with Kernow, without any faults as i restarted after...

www.emtbforums.com

 

[Madonas]

In my experience Internal issues ( torque sensors - electronics ) with Shimano motors will always progress to an initial W013 and e010. These errors just progressively get worse and will not clear easily
.These will be intermittent and seemingly can initially be reset by turning off and on . Eventually the error will not clear
When the error code will not clear , the motor is now unserviceable and needs replacement .
I have almost 10000 km on my bike and two motors have been replaced . All other electric parts are original.
The last motor that was replaced booted up just fine ( no error codes ) but the minute you put pressure on the pedals -----E010 error code. The system works great but I expect to buy another motor ( $1100 Canadian plus installation ) and receive another two year warranty.

 

[th3c0d3r]

So, now I have 2 motors irrecoverably dead with E01020 error and one about to die with W013 error, my idea is to move the electronics, the main board OR the torque sensor between one with the E01020 and the one with W013.
I didn't open any of them until now, which do you think it would be easier to move? The main board or the torque sensor?
Is this even possible? Will it work or are they "paired" in some weird Shimano-hidden-closed-buy-a-new-motor way?

 

[Spiff]

Torque sensor and its electronics are not paired to the Main Electronics PCB. Asuming that the W013 is only a defective Torque Sensor, then the easiest way would be to move a good Torque sensor from a donor motor with E10 Error to the motor that is having the W013 Torque sensor error.

EDIT 28 Nov 2021: I said the above in September, and at that time it looked like a good assumption, but since that time, other forum members have shown new opinions and assumptions, so let's be in the safe side and assume for now that the torque sensor calibration values are stored in Main Electronics Board, so they are paired and parts from different motors should not be mixed. But we will know for sure when somebody brave enough has two broken motors and tries to make a good one by mixing the torque sensor and the Main Electronic Board from different motors, then depending if it works or not we will know the answer.

It is assumed that the Torque sensor is in the Central shaft and is a "black" putty glued to the shaft, with a flex cable that is plugged into a round electronics PCB, and also a small PCB with a White cable that is plugged to the Main PCB.

All the information is in thread:

STEPS E8000 Motor Service

I've just serviced my E8000 motor and replaced some bearings, it's now working quietly and smoothly like new. If anyone needs any advice, I've got the bearing numbers and some pictures to help see what goes where. Gimme a shout if you are stuck.

www.emtbforums.com

.

If you post your questions in it, you may get the answer from people that has opened and fixed thir motors.

 

[joxelitor]

Torque sensor and its electronics are not paired to the Main Electronics PCB.

Have you tested this point?
Here it is said that calibration parameters are saved on the main board uC o memory and that calibration probably is dependant on each particular sensor.

 

[th3c0d3r]

Hey, I haven't got the time for this, I "solved" my issues by replacing speed sensor, battery harness, cables, display and finally the motor itself. And, out of frustration, I ordered the 2022 Merida eOne-Sixty with the EP8 Shimano motor, talk about Stockholm syndrome!

 

[Spiff]

Have you tested this point?
Here it is said that calibration parameters are saved on the main board uC o memory and that calibration probably is dependant on each particular sensor.

I also have not tested this. I said the above in September, and at that time it looked like a good assumption, but since that time, other forum members have shown new opinions and assumptions, but as far as I know nobody has tested it. So let's be in the safe side and assume for now that the torque sensor calibration values are stored in Main Electronics Board, so they are paired and parts from different motors should not be mixed.

But we will know for sure when somebody brave enough has two broken motors and tries to make a good one by mixing the torque sensor and the Main Electronic Board from different motors, then depending if it works or not we will know the answer.

BTW, new assumptions about the type of sensor are that the torque sensor is of the type "Strain gauge" and not magnetostrictive as you said.

 

[joxelitor]

Yes could be an strain gauge one, but for me it´s strange because the ring magnet is too far from the fixed PCB. There is where the encoder should be for detecting the pedal stroke.

For example the encoder Allegro A1333 used on makerbase servo stepper motor for 3D printer in mounted on top of the magnet tied to the shaft. or this one from M15 Modular Magnetic Encoder | Speed & Position Sensor has the magnetic element close to te sensors

If it´s an strain gauge things could be easyer for us as i believe, calibration would deppend on the black sleeve material and the glued sensors, and not on the pedal shaft

Ebike torque sensors use both type of sensors among others
Anyway I edited my assumptions on the sensor type leaving room for doubt until it´s confirmed

 

[Spiff]

The grey ring magnet looks like a magnet for a Hall sensor for cadence and crank position.

The 2 sensors buried in black putty look like the strain gauge sensors that are connected to the rotating PCB, that communicates wirelessly (NFC?) with small fixed PCB

 

[JimmyBoy]

Be careful with conductive grease on terminals, it is conductive and could short out the terminal.
I use contact grease which has been designed to remove or prevent the build-up of oxides and other contaminants on the surface. I apply it sparingly with a cheap glue brush.

 

[Spiff]

Be careful with conductive grease on terminals, it is conductive and could short out the terminal.
I use contact grease which has been designed to remove or prevent the build-up of oxides and other contaminants on the surface. I apply it sparingly with a cheap glue brush.

I use an ear swap to deposit a thin layer of grease in the metal pin in the bike battery dock.

Which grease do you use?

 

[JimmyBoy]

Which grease do you use?

Contact Grease