BionX: how to access bad strain gauge sensor?

I can't help you with your issue but I would love it if you could help me with mine. I have a bad strain gauge sensor and was wondering if I could access it without delacing the wheel and taking the motor apart.

Does it look like the freehub body can be removed from the side of the motor and would the strain gauge sensor be hiding on the axle under that location?

EDIT: This topic was originally a reply to the following post but the moderators moved it:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=90037&start=75#p1455083

I've just taken a BionX hub motor (P series) apart, and the single strain gauge mounted to the axle is well hidden under some black potting compound well inside the motor. You can't access the inside of the motor without unlacing the wheel, removing all axle nuts from both sides of the motor, and prying apart the motor halves. The freehub stays mounted to the drive-side half of the motor.

Just out of curiosity, how did you determine that the strain gauge is "bad"?

I've never posted a photo, but this is probably as good a time as any to try. The original is over 5 meg, which I know is too big, but this version is ~ 71K.

It shows -- I believe, but am not positive since I haven't scraped off the potting compound-- the location of the BionX hub strain gauge. The fact that this is where the two small white wires marked "Gage" on the controller board go is supportive of this guess. I measured the resistance of the gauge at 350 ohms, and am guessing that the controller board likely has three more equal resistance items to make a full bridge.

Thanks. My original message was in response to someone who had taken his D500 motor apart but I guess the mods didn't think it was an appropriate message for that thread.

I determined the sensor was bad because it says so on the console (strain gauge maxed out) and also gives full power most of the time.

I think I found a site that shows the sensor on the D500 being accessible after removing the freehub. I hope that is the case because I don't want to take the wheel apart.

Any chance you can estimate the approx size of the sensor?

I appreciate your wanting to avoid opening up the motor, but unless you are a surgeon with laparoscopy-type skills, I think you'll find this to be very challenging.

My experience with strain gauges suggests that this BionX gauge will be no larger the size of an average human's thumbnail, and likely much smaller than that. It's thickness will be similar to one or two sheets of normal paper.

While it's certainly possible for a strain gauge to fail, if your gauge is as well positioned and protected as the one in the photo above, it's unlikely. It would have taken a very powerful stress event on the axle to twist the gauge beyond its limits to cause a physical failure, and even the ingress of water or corrosive elements would need to get past that potting.

If removal of the freehub gives you sufficient access to the area containing the gauge, you might be able to probe the gauge's two wires with an ohmmeter. I measured a very typical 350 ohms (actually 349.6) across my motor's gauge, but your model may be different. Strain gauges rarely have a resistance outside the 150 to 1,000 ohm range. If you have access to a meter than can measure milliohm resistance, you can also flex the axle and see if the resistance reading changes slightly, which would suggest the physical gauge is still ok (which I suspect).

Remember that the gauge itself is just the beginning of the chain that makes the assistance system work. If your controller's gain settings are too high, or other adjustments are way out of range, you might see the same errors.

I was finally able to get the freehub off my motor and it provides enough room to replace the strain gauge without opening the motor.





I have ordered new strain gauge sensors off aliexpress and I will attempt to replace the faulty one when the shipment arrives.

In the mean time I may replace the existing sensor with a 360 ohm resistor to trick the system into thinking that the max force is being applied at all times. Then hopefully I can get a steady/consistent level of assist by toning down the assist percentage of the 4 assist modes.

I like your approach to toolmaking -- kudos to you for great ingenuity! May I assume that the freehub threads are standard/left hand threads?

It looks like you have the newer, big 'pancake' hub motor, in which the part of the axle with the strain gauge appears to be even *outside* the motor's casing. This seems to be a 'lucky break' as far as access to the strain gauge is concerned.

I have no idea how BionX implemented the amplifier circuit necessary to obtain useful information from this gauge, so I've no clue as to whether your strategy with a slightly higher value resistor will work or not. When I design/use strain gauges in ebike applications, I always use four strain gauges at a time, arranged in what's called a "Wheatstone Bridge" configuration. I capture the *differential* voltage across this bridge and amplify/use that as input to the controller. In my approach, simply 'unbalancing' the bridge with a strain gauge (resistor) of a different value wouldn't give me anything other than what a 'stuck throttle' would, which is generally *not* desireable behavior.

Let us know what you find as you go forward, since this is of interest to a number of people here.

The threads are standard (right hand, counterclockwise to loosen).

I removed the strain gauge sensor and tested it (console code 3772). The bike showed that the sensor was maxed out(which is what I wanted) so I didn't need to solder a resistor between the two terminals of the wire. I just insulated the end of of the wire and put it back in place.

Putting the freehub back on turned out to be a pain with the axle there. The axle made it difficult because it didn't allow the freehub to line up quite right and it also made it difficult to apply pressure while tightening. I may have cross-threaded it. :?

After getting it back on I set the assist modes to 5%, 10%, 15% and 25% which gives me a nice range of steady assist. It's much better than trying to hold the tiny throttle at a set position.

When the sensors arrive from China(in 1-2 months) I will attempt to put one on. Hopefully I haven't messed up the threads too much.

Thanks for the update. Let's hope the freehub threads remain functional.

I agree with your take on the BionX throttle -- it's very difficult to hold it in any particular position. I'm not sure this applies on your system, but on mine, just holding it WOT (wide open throttle) resulted in the bike going to max speed, and then having the throttle/assist just cut out until the speed dropped below the set limit again.

Keep us in the loop on your strain gage repairs, and best of luck going forward.

Hello. This is my problem too. Should I change the gauge sensor.How many sensors is in bionx motor? Can you give me a link where you ordered it? Thank you very much.

There is one sensor on the axle. Do you have a D500 motor? If not, you'll probably have to take the whole motor apart which would be a major pain.

I ordered 10 sensors because the procedure is apparently a little difficult. Here is what I ordered:

https://www.aliexpress.com/item/Free-shipping-10pcs-lot-BF350-3AA-BF350-Precision-resistive-strain-gauge-strain-gauge-for-the-pressure/32820683582.html

If that seems too difficult You could also just remove the old sensor and insulate the terminals on the wire. I have been running my bike without a sensor and it has been quite nice. With no sensor the motor thinks I'm pedaling with maximum force at all times so I changed the 4 assist levels from 25%,50%,100%,200% to %5,%10,%15,%25. It's kinda like having cruise control.

bikemikem said:

Do you have a D500 motor?

Click to expand...

I have 250 W motor.Thanks for replay.

bikemikem said:

I was finally able to get the freehub off my motor and it provides enough room to replace the strain gauge without opening the motor.

IMG_20190407_215424[2].jpg

IMG_20190407_220445[1].jpg

I have ordered new strain gauge sensors off aliexpress and I will attempt to replace the faulty one when the shipment arrives.

In the mean time I may replace the existing sensor with a 360 ohm resistor to trick the system into thinking that the max force is being applied at all times. Then hopefully I can get a steady/consistent level of assist by toning down the assist percentage of the 4 assist modes.

Click to expand...

Hello, what is the diameter for the pipe you get to open that side ? Thanks, i suspect bearing to make so much noise now...

bikemikem said:

There is one sensor on the axle. Do you have a D500 motor? If not, you'll probably have to take the whole motor apart which would be a major pain.

I ordered 10 sensors because the procedure is apparently a little difficult. Here is what I ordered:

https://www.aliexpress.com/item/Free-shipping-10pcs-lot-BF350-3AA-BF350-Precision-resistive-strain-gauge-strain-gauge-for-the-pressure/32820683582.html

If that seems too difficult You could also just remove the old sensor and insulate the terminals on the wire. I have been running my bike without a sensor and it has been quite nice. With no sensor the motor thinks I'm pedaling with maximum force at all times so I changed the 4 assist levels from 25%,50%,100%,200% to %5,%10,%15,%25. It's kinda like having cruise control.

Click to expand...

Hi bikemikem,

I've seen your 2019 post about running without a sensor, "cruise control" like. I have tried the same thing, the BBI sees maximum gauge signal, as you mention but unfortunately I get zero assistance. What am I doing wrong? Any help is highly appreciated. Thank you! Dan

danj said:

Hi bikemikem,

I've seen your 2019 post about running without a sensor, "cruise control" like. I have tried the same thing, the BBI sees maximum gauge signal, as you mention but unfortunately I get zero assistance. What am I doing wrong? Any help is highly appreciated. Thank you! Dan

Click to expand...

What I also notice in the BBI is that the motor parameter v12v is only 5.6V, while on a good motor it is rather close to 12V. I have a second motor with a train gauge issue (probably detached?) and v12v is 9.6V. Any ideas?