What I didn’t know about electric bikes

After a lot of research this weekend, I learned quite a bit more about electric bikes and how they work.  I particularly concentrated on how the motor in the rear hub operates and how it can provide power without any moving parts.  My first goal from this point will be to open the hub shell of an Ultra Motor brand hub motor in the rear wheel of a Stromer.  I’ll detail all of it in a well-covered post.

Most of the information I found was on just a couple websites which I somehow had not come across yet. Electric Bike is definitely well organized and provided the most information about a variety of hub motors and their corresponding controllers and diagnostic procedures.  Ypedal has an amazing and vast knowledge of the systems and has created tons of custom setups as well as details repair on several Youtube videos that is a complete procedure and well explained.

Most of the motors have a similar layout:  An outer shell to protect and encase the motor, then a ring of magnets that are positioned between the shell and coils of copper wire.  Other than a couple sensors and wires that exit through the center of the wheel near the axle, that’s about it.  Very simple and once familiar with the parts, moderately easy to work on.  I chose to first learn about the motor because it seems like this is the only component that I haven’t been able to service other than to simply install a replacement wheel.  To have the ability to fix the motors once the manufacturer’s warranty expires will be essential — especially as the number of electric bikes is increasing and they are becoming more common.

Back to the sensors and wires in the motor.  Usually there are three small square shaped sensors that are positioned between two of the outer magnets in line with each other.  These are called Hall sensors and they measure the electrical current coming out of the motor as the wheel spins.  A tapered side of the sensor body is always positioned outward and usually is set with a tiny amount of JB Weld or epoxy. Out of the top of each sensor are three uncovered wire leads that are soldered to colored wires carefully running to the center of the hub and exit from the wheel to the controller and potentially a torque sensor.  A torque sensor is generally on higher end electric bikes while cheaper versions simply use a cadence magnet to calculate the input of added power.  The torque sensor is nicer because it adds power based on a direct measurement of the deflection of the wheel backward against the rear dropout when you put pressure on the chain and pedals.  Most are robust and work accurately, giving the rider a better feel of added power when accelerating.

In some situations, I read that one or more of the Hall sensors can go bad and with the right tools, it is a small project for an afternoon that will avoid a replacement ($500-750) or service elsewhere ($100/hr).  Even if you aren’t up to doing the task yourself, you will be better able to diagnose and familiarize issues that occur.  For the avid mechanic, it is possible to upgrade your hub motor for minimal cost.  Larger wires (if they can fit through the frame and center of the wheel) will boost available power and a better controller or throttle can more accurately distribute the power.  Unless you have an extreme desire to mod your electric bike, the magnets and copper wire are difficult and costly to replace.  Some motors are spun low with larger gauge copper wire which can provide more power, but less torque on hilly terrain while high spin copper will have better power at a high cadence and will last longer.

In order to remove the motor from its hub shell, a car bearing puller is necessary.  It is an extractor style tool with three arms that pull on the shell while a center bolt pushes the axle and opposite hub shell side.  It is helpful and recommended that you tap the shell with a mallet during this process to aid in freeing the two halves.  Otherwise, direct extraction with the bearing puller could break the hub flange and then you’ll be left with a useless motor and no shell.

Considering the Stromer that was detailed in the review and diagnosis, I believe that the torque sensor was overloaded in the frame without that necessary bolt and the motor overheated and fried one or more of the sensors.  This would explain the NO_COMM lack of communication between the LCD controller and the motor.

It makes a lot of sense once the elements are broken down like that.  I also think that one of the wires exiting the rear hub is a faulty wire with a weak spot somewhere near the center of the wheel.  The torque sensor itself seems okay.  I am going to check current through each of the wires to determine the faulty one and then solder in a new one along with the replacement of the sensors.

I was very happy to find this information and come to realization that these bikes can be fixed from the ground up if necessary.  Why wait a week or weeks for new parts to install when the local hardware store and Radio Shack have all the necessary parts for a solution?  I am going to continue to do more research into the motor repairs as well as connecting different controllers to the bike for operation.  It is exciting to realize that with the right research, I can create incredibly fun bikes that are useful for both fun and transportation. Now, once I can locate a car bearing puller, I’ll open the motor and take photos of everything so you can see it.  Stay tuned!



Stromer Update Photos and Review Part 2

I have an updated gallery of photos for the Stromer ST1 and will be continuing to review it.  After few more issues and discoveries, I will be posting an article later to detail diagnosing the differences between issues with the display, issues with the motor, and issues with the battery.  Some of the information I have gathered will work to fix the problem and some is for correctly diagnosing what needs a replacement or solution.  In my recent article, I talked about the display having problems and have refined the diagnosis a bit further.  As I compile the information and gather the remaining photos today, browse the gallery and compare the components to what is on your Stromer or other pedelec bicycle.  Once again, I would like to mention that the support from both Specialized and Stromer have been top-notch and most of the models that roll through my shop are trouble free and don’t experience any interruption in use.  It can be overwhelming to read an article that mainly details problems, so I will also try to highlight the good points of design and features that I see.  If you own a Stromer, you probably have spent a bit of time on Google to read forums about issues you may be experiencing and found that the information is loosely organized at best and offers very little in the way of a compiled source other than the company’s website.

The article should be finished up by tomorrow morning.  Enjoy the photos!

Stromer Troubleshooting and Diagnosis Part 1

So, I have done a bit of research in my off time to get more acquainted with the Stromer electric-assist bicycle.  It is the other brand of ebike that we carry in our shop and is notable for its price point and performance.  Not only will it go quite a bit faster than the Turbo (about 50kph or 31mph), but it also comes in at a price point of about $3,500 USD versus the $5,999 tag on the Turbo.  Both equate to an amazing experience on a bicycle, but this option seems more affordable than the Turbo without sacrificing quality.  I like to think of the Turbo as an ebike with style where the Stromer is a workhorse that won’t quit and won’t deteriorate under pressure.

That being said, I wanted to go over a few parts of the Stromer that I have worked with and what expect for a long term review in the future once I ride them a little more.  The Stromer is designed by BMC (the company chose to call itself the initials of its UCI code), a Swiss manufacturer that’s founder was the owner of the famous 711 racing team.  They make two models: the ST1 and the ST2.  Both can be configured with various options from a suspension fork or carbon fiber fork to three levels of motor power and two battery levels.

Stromer ST1 Specs

Stromer ST2 Specs

The ST2 is an upgraded version that has a possible range of 150km on one charge.  That is on of the longest lasting battery specs on the market with 814 Wh.  The newest model has a smartphone integration for real time stats and data.  With built in lights, fenders, and a rear rack, it is a notable competitor for the Turbo S and when the newest model hits our shop, I’ll be posting a review on it.  Until then, I will be dealing with troubleshooting the Stromer ST1 and ST2 from prior years since they are more common.  Having all the information I have gathered in one place will help both you and myself as it expands.  From past research, I have not found very useful information on the Stromer on the Internet and have relied mostly in swapping components from a new build to the repair and having replacements sent from the representative for the company in the USA.  That being said, there is still a great deal of information that can speed up repairs and diagnosis and keep your customers happy and riding.

Much of the issues I have come across seem to deal with replacing the display unit or cleaning connections.  One issue dealt with a bad charger (It should be noted that the charging process of the battery off the bike is very specific).  First, plug the adapter cable to the battery.  Then plug the opposite end of the adapter cable to the charger plug.  Next, plug the power cord into the opposite side of the charger from the adapter cable.  Lastly, plug the power cord into the wall.  You should see a red light appear in the LED bubble on the side of the box charger.  After a few seconds (up to about 5 sec), it should change to either an amber color (signifying it is charging) or green (Stromer recommends leaving the charger on while the LED is green for about an hour for maximum charge level).  If the charging connectors are not plugged together in the correct sequence, damage can occur to the charger and the battery and result in a solid red LED on the charger box.  If this is the case, consult your dealer for a replacement charger.  I have had confusion come from customers on the charging process and it has led me to have to deal with recharging a supposedly malfunctioning battery with a new charger.  In these instances, I also try charging it through the bicycle, which helps to eliminate the battery being the issue.  Here are the photos in order of the process.

Cleaning the connectors should be done carefully and with the battery out of the bike.  While an electrical discharge is not likely, be safe and take the battery out first.  This also provides an opportunity to check the connections for the battery inside the frame and on the battery itself.

Here are a few photos of the various electrical connection on the bike.

The two main issues of display replacement have been related to the information messages shown on the display.  One issue of NO_BATT appeared on the display after only a couple of weeks of use.  I checked the battery both in and out of the frame and the connections leading from the display to the battery.  All of them seemed clean of debris or liquid and were securely connected.  The next step was to reset the display to see if the system would correct itself upon start up.  I removed the display from the handlebars by disconnecting the wires and taking the two anchor screws out of the band mount.  Once removed, I used a quarter to turn the battery cover to the open position underneath the circular gray foam pad protecting it.  Taking a scribe, I popped the cover open and removed the CR2032 battery and inspected the two terminals inside the battery compartment.  Both also seemed clean and correctly positioned for contact.  Taking a new CR2032 battery, I installed it carefully and then put the battery cover and foam pad back.  Mounting it back onto the handlebars, I proceeded to carefully connect all of the wires appropriately and turned the system on.  The message of NO_BATT remained and I contacted the rep about a replacement.  In the short-term, I went ahead and connected a brand new display from a model not yet built to this Stromer and everything seemed to work great afterward.  The replacement unit came in and I reinstalled it onto the new build without issue as well.

Another issue I came across (though I believe it to be from either improper assembly or repair in its past and not from a defect) was a mechanical one.  The spacer nut on the rear axle drive side that correctly positions the freewheel away from the frame was completely missing.  I was able to find a suitable spacer and nut at a local hardware store and the wheel spins freely once again.  As the wheel was initially mounted, the high end of the freewheel was locked against the inside of the rear dropout on the frame and scored a fine line into the metal.  If this had continued, the frame would have likely had to been replaced.  You can see it in the photos here.

I am still not finished with the diagnosis of this particular model.  It has seen quite a few miles of use and also has an issue with the display showing an information message of NO_COMM.  This error means that there is a disruption in the communication between one or more of the components.  I scanned through the connectors, but could find little evidence of a bad wire or connector.  The only connection that seemed to have suffered from the weather was the connection to the rear dropout shown above in the right photo and above on the inside of the chainstay.  I question whether attempted use of the bicycle with the freewheel in its original condition as it came to the shop would have overloaded or shorted the sensor in the frame.  This information has been forwarded to the rep for possible solutions and extra wires to swap and test.  Otherwise, I will attempt to receive a replacement display to test (the models in the shop utilize a 2 wire display, where this model uses a display with 4 wires).  My general feeling is that correction of the rear wheel axle assembly and replacement of the display and rear dropout wire will fix all of the issues.  The bike has had a replacement rear wheel already, so the motor should be in good working order.

I will add to this article as the project progresses!  Feel free to message or comment any questions or suggestions.


Specialized Turbo S Troubleshooting and Diagnosis Pt.1

Welcome to Part 2 of 3 in my Specialized Turbo S Long Term Review.  I feel like this is some of the most essential information that I can apply in one place that will keep Turbo pedalelecs running beautifully.  However, I do feel that there is a “low-ceiling” limit to the depth that the Turbo can be maintained — even by an experienced and well-equipped shop.  In certain instances, the bike may have to be sent to the factory.  I am of the belief that this is being addressed and the bike will become even more modular and accessible.  That being said, there is a ton that will be covered here and includes several years of work and knowledge.  While I will be mostly highlighting troubled issues, the Turbo overall has been very trouble free and generally without issue.  This just compiles issues from many Turbos I have serviced since their release.



So, I turned on my Turbo and started pedaling and the motor did not kick in as expected in Full Active and Eco modes.  What’s up with that?

I am starting with this problem because I think it has occurred more than any other issue.  There are a number of symptoms that cause it to happen and a number of solutions.  First, the battery is nearly dead.  Charge it up and try again.  The next thing I would try is to simply turn it off and turn it on again to try it.  This occurrence has spanned several Turbo S bikes on a rare occasion and has been remedied every time in the shop.  The next reason it might not turn on is that the battery may not have been installed properly or the pressure plate at the bottom of the battery “hanger” is not correctly adjusted.

With the battery pressure plate, I have found it to be different on each bike and definitely has a “correct” pressure for the Turbo to function right.  On the bottom of the bike behind the bottom bracket is an access plate that is secured by small 2mm bolts.  Remove these and set aside in a magnetic parts tray so they don’t end up lost on the ground.  Once the anchor plate has been removed, look inside and you’ll see wires, a couple hex bolts, and the shell of the bottom bracket.  Locate the 5mm hex bolt right in the center anchored into a flat plate.  This is an adjustment bolt for moving the battery up and down within the battery dock.  It provides pressure against the battery so vibrations do not disrupt the connection at the opposite end of the down tube.  To be noted, one would likely expect the phrase, “Why not just crank that thing down and really lock it in there?”  Well, here’s why.  If the motor does not seem to be engaging as aforementioned, this same issue can occur if the battery is adjusted to tightly in the dock.  Just as the battery has a little “wiggle” room, so does the connection to the battery inside the frame at the top tube and head tube junction.  If you remove the battery and look at the top of the battery dock, you’ll see where it plugs in.  With your fingers or a scribe, you can definitely see gaskets on the side of the connection and its ability to move slightly up, down, left, or right with a small amount of push.  The general rule is to begin tightening the pressure plate while using your free hand to try and move the battery up and down in the frame.  If you hear any sharp noises when it contacts the top or bottom, continue tightening.  Once there is little movement (0.25-0.5mm), stop tightening and use the key to remove the battery.  If it is too tight, it will be difficult to get the battery out.  I have found that each Turbo is slightly different, which is a great reason to have the plate for adjustment.

Another item to check is the connections between the brake and the control unit and the connection to the wire coming out of the frame.  Also, as a measure of being thorough, there is an additional connection inside the frame at the head tube where the light plugs into a parallel connector with the control unit (Several times, fault signals on the battery will indicate the light and control unit both having an issue when it is mostly one or the other).  By checking these connections, I imply:  Are they tight and correctly aligned for the individual connector pins? Have you applied a tiny amount of dielectric grease to the threads of the connectors? Are the connections and wires constantly rattling or loose?  Try checking each of these regularly when servicing as a quick way of eliminating them in the event of an issue.

So, I turned on my Turbo and everything started up, but a flashing code reading “short circuit” is displaying on the control unit?  Is that bad?

Well, it’s not necessarily good, but can be fixed easily.  Sometimes, constant vibration, continued exposure to the elements, and leaving the rubber charging cap off of the Turbo for extended amounts of time can cause an interruption in the electrical system and send a red flag to the control unit.  It’s a feature designed to override the system much like a circuit breaker so no damage occurs to vital components.  By using the Turbo diagnostic tool plugged into the frame (with battery in) and into a laptop, you can run the diagnostic software and release the hold on the circuit with a convenient button on the screen.  This software is a very important tool in confirming that the motor, battery, control unit, and lights are working properly and displays a wealth of information and reports for keeping Turbos happy and healthy.

My Turbo has gone insane and decided to run at 28mph like a throttle as soon as I hit the pedal.  I don’t even need to pedal. Glad it’s spec’d with great disc brakes.  Does it need an exorcism?

Not quite.  In the early days, Specialized might have taken this approach, but with the “Innovate or Die” theme currently in trend, we now know it’s quite real. This issue has only occurred once, yet is important enough to catalog and solve for any future happenings.  In my prior post, I detailed the symptoms of the bike while it was secured in the bike stand.  I took the opportunity to  remove and clean some of the components on the bike and the frame (After slowly and carefully removing the kickstand, I have realized that unless it is absolutely necessary, don’t ever remove it.  It is tedious because of the frame angles to loosen and tighten the 8mm hex bolt and I found that it didn’t improve access to the electronics and pressure plate under the bottom bracket.  It is a very sturdy dual kickstand that stays tight and balances great.). That being said, I removed the bottom bracket (EVO386) with a socket and outboard cup adaptor and found plenty of access points into the frame to see the wires running to the motor and seatpost LED and the hex bolt adjustment for the pressure plate (5mm) and the lock mechanism with it’s integration into the battery release.  I wiped the battery dock and the battery with isopropyl alcohol.  The terminal connections were also cleaned with isopropyl and a swab.  With some time open in the service schedule, I also decided to remove the external frame charger / battery dock connection unit from the top end of the down tube to make sure all of the connections looked secure and free of debris and moisture.  Thanks to a well-sealed compartment, everything looked great.  I took notes as well on the direction, colors, and pairing of wires that interfaced with the unit.  Red and black paired wires run directly from this to the bottom bracket shell, under the EVO386 BB, and through the non-drive chainstay to the external port near the rear wheel.  A shorter set of these two wires also connects the battery dock connection to the external charge port.  Two sets of blue and brown wires run from the unit to the headtube and into the brake / control interface and also to the bottom bracket shell and into the seat tube to connect into the wires running from the saddle’s integrated LED through the seatpost. One triple set of wires (blue/brown/green) run from the external charge port in the frame to the top of the battery dock where the battery connects to the system.  Lastly, an orange wire runs from the front light to the connection with the control interface (wired in parallel).  As a side note, this is why the diagnostic on the battery will show both the light and control interface having an issue when it could potentially only be one of them.  Good things lights are pretty easy to check.

Once I had satisfied my curiosity of understanding the Turbo S better, I moved on to reinstalling the parts and activating the wheel like before.  I let it run continually for over five minutes with no discernible change.  Continuous operation, only to be paused by giving the brake a solid squeeze.  If the wheel was touched, the throttle would pick up again.  This was also true for pressure on the pedals.  In these situations, I was lucky to have another in the store to help eliminate components.  Taking a brand new rear wheel from the second Turbo (after testing to make sure it was operational), I installed it into the first and powered the system.  No activation of Full Active or Eco modes, but No Power worked as expected and Regen mode seemed more difficult than normal as mentioned with the original wheel.  This is the only area that I am not quite sure what to think of yet.  That’s what the techs at Specialized are for. Continuing with the diagnosis, I also swapped the battery with no effect.  Reinstalling the original wheels and batteries to each Turbo, I then swapped the control interfaces.  This also did not yield any effect.  With the original wheel, the throttle accelerated as before (One thing I might try today is to try both battery, control interface, and new rear wheel from the second Turbo into the first.).  The last thing I did was to put the wheel from the service Turbo into the new Turbo.  It accelerated to 28mph.  Thus, my conclusion (though I will wait for confirmation from Specialized) is that the wheel needs to be recalibrated–specifically the motor.  If this is the case, it will go back to the factory for a quick turnaround and then I’ll have an update on a future post about the issue’s resolution.  This was a great opportunity to fully inspect the bike from top to bottom, inside and out.

I am going to continue this post through the weekend with some other minor issues and fixes that will keep the Turbo S running smooth and fast, but figured that it’d be good to start posting the article for any feedback or questions.

Here are photos of the above parts:



More to follow!  – SNC

Turbo Diagnosis Update

Hey everyone! The Turbo S Diagnosis and Troubleshooting article is almost complete. There has been a slight delay because of a new issue that arose within the past few days that is still in progress of being solved and fixed.

One of the 2014 black Turbo S models came in with a particular issue with the motor engaging once pressure to the pedals occurred in Full Active Mode and Eco (30%) mode. You might think, “Isn’t that what the bike is supposed to do?” Well, yes. However, with this particular one, the motor engaged to the Turbo max speed of 28mph without disengaging when the brake was pulled and acted as a throttle (no pedaling required) rather than an electric assist. This could be compared to the gas pedal on your car sticking on the floor and your car traveling at top speed until you shut it off or change gears (on the Turbo, going to No Assist mode or Regeneration mode would disengage the motor). As a side note, one the Turbo was in Regeneration mode, it was much more difficult to pedal than normal while in this mode and the wheel would take a small 1/8 to 1/4 turn backwards when the pedals stopped turning.

As you may have guessed with continual throttle at 28mph, it totally cooked the rear brake pads and most of the front.

So, I will be problem solving it for a day or two more and running several tests while collaborating with Specialized technicians to diagnosis this and of course, finish up the results in Part 2 of the Turbo Review. Thanks for being patient. I hope this gets you curious about the bike and working on it more. I still think the bike is fantastic and doesn’t normally reflect any or all of these problems. In large, it is the most trouble-free electric assist bike I have worked on out of over twenty different brands. These issues are published to conglomerate the most information possible in order to have the largest audience possible understand how to work on the them and maintain them.