Analog Motion e-bike troubleshooting

[Eruss17]

Hi

I’m new to e-bikes, but have 30 years of electronics experience in work.

I have acquired two Anolog Motion AMX Step e-bikes.
I have one functional battery (36v 7Ah).
Both bikes have the Topology SW102 controller and a bespoke controller/battery holder (Analog Motion Model LSW1188-34-3F, possibly a Lishui) I believe a Bafang rear hub motor labelled bbc36700a250w (250W 36V), but bike manufacturer says it is made by Aikema.

With a fully charged battery, battery ON (100% capacity in the 4 green LED’s), I press and hold the orange power button in the SW102, but it does not boot.
I have checked the connection to the SW102, and the wiring loom has 41v on the Power pin to the SW102 controller (on the JULET connector).

This is the same for both bikes. I guess it is possible for both controllers to be faulty or both displays, but I don’t know the history of either bike so now I’m stuck.
This bike doesn’t have a throttle, just a bottom bracket rotation sensor, which is plugged in.
I disconnected the 9 pin motor cable but no luck.

Is it possible to disassemble the SW102? Does the screen unclip of does it need to be gently heated for removal (like an iPad)?

Should the display boot with just power and no communication with the controller? If it does boot with power only, I would expect a communication error.
I attempted to disassemble the controller, but it appears that there are no connectors inside the controller housing, all wires are hard wired to the PCB.

Without a working display, I don’t know what is going on.

I understand that not all SW102 displays are the same (different communication protocols UART & CANBUS and potentially bespoke firmware) so how can I find a replacement controller. It doesn’t help that the manufacturer Anolog Motion, are not in business anymore.

Any advice appreciated

Regards

Russell

 

[Apartunis]

First, closely inspect the connectors that electrically join the battery to the bike. If they are loose, bent, burned, discolored, or corroded, that might be the point of failure. It's a fairly common failure that has become somewhat less common with a trend towards round pins rather than flat blades.

The state of charge indicator on your battery only checks total battery voltage. If the battery has gone out of balance through long disuse or mistreatment, its BMS may have cut off discharge even if its total voltage is relatively full. Because the BMS uses FETs to cut off an out-of-spec battery, you many still be able to measure a voltage that is only leakage across the FETs. Typically this voltage will drop as you measure it.

Check voltage at the battery discharge terminals. If that voltage is high and steady, the controller *should* be able to turn on, even if the battery is too weak to sustain a motor load. By extension, a low or unsteady voltage indicates a battery problem, and a high steady voltage suggests a lack of continuity or problem with the controller.

If the bikes are electrically identical, then any endemic design or manufacturing defect could easily affect them both.

 

[E-HP]

Hi

I’m new to e-bikes, but have 30 years of electronics experience in work.

I have acquired two Anolog Motion AMX Step e-bikes.
I have one functional battery (36v 7Ah).
Both bikes have the Topology SW102 controller and a bespoke controller/battery holder (Analog Motion Model LSW1188-34-3F, possibly a Lishui) I believe a Bafang rear hub motor labelled bbc36700a250w (250W 36V), but bike manufacturer says it is made by Aikema.

With a fully charged battery, battery ON (100% capacity in the 4 green LED’s), I press and hold the orange power button in the SW102, but it does not boot.
I have checked the connection to the SW102, and the wiring loom has 41v on the Power pin to the SW102 controller (on the JULET connector).

This is the same for both bikes. I guess it is possible for both controllers to be faulty or both displays, but I don’t know the history of either bike so now I’m stuck.
This bike doesn’t have a throttle, just a bottom bracket rotation sensor, which is plugged in.
I disconnected the 9 pin motor cable but no luck.

Is it possible to disassemble the SW102? Does the screen unclip of does it need to be gently heated for removal (like an iPad)?

Should the display boot with just power and no communication with the controller? If it does boot with power only, I would expect a communication error.
I attempted to disassemble the controller, but it appears that there are no connectors inside the controller housing, all wires are hard wired to the PCB.

Without a working display, I don’t know what is going on.

I understand that not all SW102 displays are the same (different communication protocols UART & CANBUS and potentially bespoke firmware) so how can I find a replacement controller. It doesn’t help that the manufacturer Anolog Motion, are not in business anymore.

Any advice appreciated

Regards

Russell

If you suspect a display problem, you can bypass the display to turn on the controller as a troubleshooting step. If you bypass and measure 5v on the controller’s PAS connector, then chance are the controller is functional. If you have the display manual, it may have the connector pin out. In the example below, shorting pin 1 with pin 5 should turn on the controller.

https://www.cycloboost.com/media/guide/CYC%20%C3%A9cran%20SW102%20guide%20d'utilisation.pdf

 

[Eruss17]

First, closely inspect the connectors that electrically join the battery to the bike. If they are loose, bent, burned, discolored, or corroded, that might be the point of failure. It's a fairly common failure that has become somewhat less common with a trend towards round pins rather than flat blades.

The state of charge indicator on your battery only checks total battery voltage. If the battery has gone out of balance through long disuse or mistreatment, its BMS may have cut off discharge even if its total voltage is relatively full. Because the BMS uses FETs to cut off an out-of-spec battery, you many still be able to measure a voltage that is only leakage across the FETs. Typically this voltage will drop as you measure it.

Check voltage at the battery discharge terminals. If that voltage is high and steady, the controller *should* be able to turn on, even if the battery is too weak to sustain a motor load. By extension, a low or unsteady voltage indicates a battery problem, and a high steady voltage suggests a lack of continuity or problem with the controller.

If the bikes are electrically identical, then any endemic design or manufacturing defect could easily affect them both.

Hi,

Thank you for your reply.

Battery terminals appear good (blade type).

Regarding the “out of spec battery”, I would expect a significantly lower voltage reaching the display, if the battery was being disconnected due to an out of spec battery. I can measure 41v dc at the connector (with display disconnected, so battery not under load). Difficult to test the voltage with display disconnected.

Regards

Russell

 

[Eruss17]

If you suspect a display problem, you can bypass the display to turn on the controller as a troubleshooting step. If you bypass and measure 5v on the controller’s PAS connector, then chance are the controller is functional. If you have the display manual, it may have the connector pin out. In the example below, shorting pin 1 with pin 5 should turn on the controller.

https://www.cycloboost.com/media/guide/CYC%20%C3%A9cran%20SW102%20guide%20d'utilisation.pdf

Hi,

So if I connect pin 1 (Power from battery) to Pin 5 (Key in - I assume power to controller), controller should be activated and the bike will function as expected, but with no display to show speed etc., is that correct?

Regards

Russell.

 

[docw009]

One Battery. Two bikes. Neither display lights up, Odd..

Yes. You can measure battery voltage at the display connector. Pin 1 and Ground, The display needs nothing except power and ground to light up. When it powers up, it switches battery power to pin 5. marked Key-in in the diagram.,

Inside the controllerm if it follows standard design, the battery power is always connected to the Mosfets that power the motor, but the low voltage circuits that do the control logic, sensors amd drive logic to the transistors is powered by that Key-in line,

In summary, tying pins 1 and 5 should power up this controller, Still, check pin 1 and 2 for power before you do it. Must be something common to your battery not powering up two bikes,

 

[E-HP]

Hi,

So if I connect pin 1 (Power from battery) to Pin 5 (Key in - I assume power to controller), controller should be activated and the bike will function as expected, but with no display to show speed etc., is that correct?

Regards

Russell.

It’s a troubleshooting step, not a solution. Once you test it, the results will dictate next steps. Im using a tablet, so one step at a time avoids a lot of unnecessary typing for me. As docw009 mentioned, there seems to be something wrong common to both bikes, which is unusual. The battery is the common element, but you said you measured battery voltage at the display connector.

 

[Eruss17]

Hi

Thank you all for the replies.

At this point, all I have done is established that battery voltage is present through the wiring, to the connector that the display plugs into. Neither display lights up in any way.
I will try to bypass the display to see if the controller becomes active and report back.

I do have a second battery, but this is dead (about 10v on the output terminal) and doesn’t charge.
I was given the bikes as a pair “for spares or repairs “ so I am trying to make one good one out of two bad ones. I have no further information on the bikes’ history so I can only assume that neither work. I was advised that a local e-bike repair shop was unable to help the original owner, but that may just be because the manufacturer is not operating any more.
Judging by their condition, it may be that both displays are bad, so if bypassing them works, at least I can move forward.

Can the SW102 be opened up to access the internals?

 

[A-DamW]

I would check the battery with some kind of load test before any other troubleshooting.

The battery is the foundation of the system, if it can't supply at least enough current to fill the DC link capacitors without the BMS cutting out, MCU brown out, etc, there's no point moving forward without solving the battery.

You are going to have to get creative here, do you have any 5 watt resistors? 12 volt automotive lamps?

Most multimeters can measure up to 10 amps for a few seconds, using an inline resistance with your multimeter and measuring a few amps from the battery through the resistance, should be enough to verify the battery can at least supply some current.

 

[E-HP]

Hi

Thank you all for the replies.

At this point, all I have done is established that battery voltage is present through the wiring, to the connector that the display plugs into. Neither display lights up in any way.
I will try to bypass the display to see if the controller becomes active and report back.

I do have a second battery, but this is dead (about 10v on the output terminal) and doesn’t charge.
I was given the bikes as a pair “for spares or repairs “ so I am trying to make one good one out of two bad ones. I have no further information on the bikes’ history so I can only assume that neither work. I was advised that a local e-bike repair shop was unable to help the original owner, but that may just be because the manufacturer is not operating any more.
Judging by their condition, it may be that both displays are bad, so if bypassing them works, at least I can move forward.

Can the SW102 be opened up to access the internals?

Without knowing the history of the bikes, one might assume that the previous owner may have gone through a similar process trying to get one bike to work, and maybe even gone through a random swapping of parts, rather than a systematic approach. It's unusual that both displays show no life when you have verified that they are receiving battery level voltage. If the previous owner swapped parts, that could just propagate the problem, killing the "good" bike/controller, etc.
When you bypass the display to turn on the controller, check the battery output at the same time as the load test. If you still get battery voltage, but the controller doesn't output the 5V, then the controller, or the controller's 5V supply, is dead.
You could also test the displays separately. It would be trickier, since on the display side connector, they are male pins, so you'd want to take precautions not to short anything. If you apply battery voltage and press the on button, it should light up. If not, it's probably dead, and if both displays are dead, then it could be from the previous owner's efforts.

 

[Eruss17]

Dear All

Thank you for your support!

I will first test the load capabilities of the working battery by using three 12v 21W bulbs in series (36v 63w =1.75 Amps but with 41v terminal voltage will be approximately 1.54 amps. Which would be sufficient to establish battery capability. I can then short pins 1 & 5 on the controller side of the wiring loom which will hopefully enable the controller for test only. My understanding on this particular bike is that as soon as you pedal, the assistance is engaged, judged by a bottom bracket integrated hall sensor. There are no “brake applied” switches and no throttle.
I also have 3 12v lead acid batteries (each 15Ah), so I could experiment powering the bike with these (in series). I may also try powering the SW102 display from my bench PSU.

Thanks for the ideas!

Refsrds

Russell

 

[E-HP]

Dear All

Thank you for your support!

I will first test the load capabilities of the working battery by using three 12v 21W bulbs in series (36v 63w =1.75 Amps but with 41v terminal voltage will be approximately 1.54 amps. Which would be sufficient to establish battery capability. I can then short pins 1 & 5 on the controller side of the wiring loom which will hopefully enable the controller for test only. My understanding on this particular bike is that as soon as you pedal, the assistance is engaged, judged by a bottom bracket integrated hall sensor. There are no “brake applied” switches and no throttle.
I also have 3 12v lead acid batteries (each 15Ah), so I could experiment powering the bike with these (in series). I may also try powering the SW102 display from my bench PSU.

Thanks for the ideas!

Refsrds

Russell

Their documentation is incredibly detailed, and extensive, as are their photos. Did you already go through their troubleshooting guides (theres one for the display)? All models use cadence sensors rather than torque, for PAS. But even with such detail, I still can’t figure out where the PAS sensor actually is. I didn’t go through all 40 documents though.

Analog Motion Support | Notion

Hosted by Notion Sites — The easiest way to get a website up and running.

motion-pistachio-32e.notion.site

 

[r3volt97]

You either swap your battery's BMS or buy a refurbished one , it's the battery and after many many uses the orange button of these SW102 just sink in , a design to make u struggle . You can get rid of it though, Eggrider V2 is a good replacement for now .

Analog motion Cellpack 36v | eBay UK

Analog motion Cellpack 36v

www.ebay.co.uk

 

[Eruss17]

You either swap your battery's BMS or buy a refurbished one , it's the battery and after many many uses the orange button of these SW102 just sink in , a design to make u struggle . You can get rid of it though, Eggrider V2 is a good replacement for now .

Analog motion Cellpack 36v | eBay UK

Analog motion Cellpack 36v

www.ebay.co.uk

Update:
I bypassed the SW102, in both bikes, and both bikes work as expected.
I then pressed the power button on the SW102, and found that on both bikes, the system works as expected, just the screen doesn’t work. The lights can be turned on/off using the display button, so conclusion is that the OLED display has failed on both bikes. I decided to carefully open one display, to see if it can be resurrected, but it seems not, so I’ll just get a cheap one from Aliexpress.
Can anyone advise the likely protocol that will be required for this controller?

Thanks in advance

Regards

Russell

 

[docw009]

Well done!

It will be a hit/miss job to replace the displays. The SW102 is used in a lot of different applications and often uses different firmware to match the associated controllers.

Instead, you could continue to ride the bikes w/o a display. Buy a $10 bike computer and you have your speed/odometer. Add an external voltmeter if you want to track voltage. Given that there is no throttle, you're going to click the up/down buttons til the bike gets into an assist level that hopefully meets your needs.

If it were me, I'd buy this dreaded S866 display, associated 17A Brainpower controller. and 9 pin motor connector. Hack into the battery mount and find the power pins. The pic says it will support the pedal assist and it adds a throttle. For only 50 bucks, it's a good deal. 17A is plenty for that bike, but you could get bigger controllers.

Three Mode Ebike Controller 17A 24V/36V/48V with S866 LCD Color Display Waterproof Connector 350W Motor Conversion Parts - Walmart.com

Buy Three Mode Ebike Controller 17A 24V/36V/48V with S866 LCD Color Display Waterproof Connector 350W Motor Conversion Parts at Walmart.com

www.walmart.com

 

[E-HP]

Can anyone advise the likely protocol that will be required for this controller?

Here’s the search results for ‘Lishui protocol’, including an ES member that is pretty knowledgeable on Lishui controllers. If you don’t find the info in the posts, you could try editing your thread subject to include ‘Lishui controller and display’

Google search results

endless-sphere.com

 

[r3volt97]

Update:
I bypassed the SW102, in both bikes, and both bikes work as expected.
I then pressed the power button on the SW102, and found that on both bikes, the system works as expected, just the screen doesn’t work. The lights can be turned on/off using the display button, so conclusion is that the OLED display has failed on both bikes. I decided to carefully open one display, to see if it can be resurrected, but it seems not, so I’ll just get a cheap one from Aliexpress.
Can anyone advise the likely protocol that will be required for this controller?

Thanks in advance

Regards

Russell

good luck finding SW102 with a male end and in this configuration

 

[r3volt97]

you can retain the aspect of the SW102 if u buy an Eggrider once again , 100 quid or go for a King Meter KM529 which has the above wiring and lack bluetooth connectivity , locking the bike .

​