Nova Electric Racing: European champion 2017 in MotoE competition
During the weekend of 7-8 October, the TU Delft student team achieved the first place of the European Motor Championship of the MotoE. The self-designed and built electric race motorcycle has been specially developed for the races of this unique competition. After the three race weekends, the first place can be celebrated!
The Team
During the weekend of 7-8 October, the TU Delft student team achieved the first place of the European Motor Championship of the MotoE. The self-designed and built electric race motorcycle has been specially developed for the races of this unique competition. After the three race weekends, the first place can be celebrated!
The Technology
Although the electric motor is a lot lighter than the classic combustion engine, the biggest challenge is the weight and placement of the battery pack. The total 200kg weighted motorcycle has a continuous power of 140kW (190hp) and a torque of more than 500Nm on the rear wheel. All in all, the motor is in a race for about 40 kilometers. By designing the frame and the backbone itself, it was possible to efficiently create space for the battery pack and electronics. In addition, the self-produced carbon fiber seat and body parts also provide a lot of weight savings.
The MotoE
The MotoE competition has been specially set up to offer innovation in electric motorcycles a stage. Both student teams and companies compete with the competition to be these developments from the start. This year, the races were held on the Anglesey, Pembrey, and the late MotoGP circuit Donington Park. In order to show the potential of the electric motors, it has been possible to join a combustion class during the weekends. Inside the MotoE had her own classification. On the first two circuits, the team participated with 400cc motorcycles, and joined the headgroup. At Donington Park it was possible for the competition to ride the 600cc class. Starting from the pit street, the team of the last place has achieved a promising finish in the midfield.
Connecting the system
A connector…. or maybe not?
The first question to ask was where we actually wanted connectors in the electrical system. It is of course possible to answer this question with: “nowhere!”, and then take a bunch of wires and solder everything together. This, however, is not a smart option, because if everything is soldered together you can’t take the system apart easily to test an individual subsystem or replace a part. Also, it will become a big mess of wires, and if something does not work you won’t have the slightest idea where to start looking for a bad connection. This is why, in the design of the electrical subsystems, we already had to pay attention to the grouping of signals and putting these groups on separate connectors. This way it became clear how many connectors we would need, and how many pins these connectors were supposed to have.Different connectors with different specs
Next up, we actually had to choose connectors for our system. These connectors come with a lot of characteristics which may or may not be important to us. For example, they can be resistant to loosening when subject to vibrations, they can be waterproof to a certain degree and they can be protected for penetration by sharp objects to name but a few. The latter two are included in the IP-rating, short for Ingress Protection rating. This rating is commonly specified for electrical plugs and connectors, which was useful when selecting connectors for the battery pack. In order to protect the battery pack, its sensitive electrical components and LiPo cells, it was designed to be as watertight and dust tight as possible, and put the focus for connectors on those with an IP68 specification. In this case, the 6 stands for complete dust tight and the 8 for total immersion in water. For most equipment on the bike, IP55 is sufficient (protection against dust and splashing water), but the battery pack deserves a little more care.The LEMO connector
A commonly heard brand in the automotive- and racing branch when it comes to connectors is LEMO, which is known for its innovative push-pull system. This plug/connector design prevents plugs to be pulled from the cable or to easily vibrate loose. To disconnect the plug, it must be pulled by the grip of the plug, which releases the clamps that holds the plug in the connector. Not only is this a great feature to have on our motorcycle, it is also a very satisfying feeling when you click together your electrical systems.
A fully assembled LEMO plug, including push-pull grip and bend relief. This assembly will deliver power to the Safety Control Unit.
LEMO THT connectors visible on the right side of the Safety Control Unit. The connectors have a locking washer and nut to clamp them tightly onto the wall of a protecting casing.
Designing your cable tree
Of course there are situations where it’s more convenient to use other types of connectors. Usually because a component is manufactured with a specific type of connector, or because it is much easier to assemble or replace. Soldered connections are also prone to break with vibrations, this can be improved by applying a bit of glue. Crimping contacts are then a better alternative. Another thing that greatly improves a cable tree is consistency. Sometimes it is not totally clear, but if you can make some agreements on common connection types it helps a lot. For example on our motorcycle, all power delivering connections use female plugs, while power consuming connections use male plugs (similar to regular power plugs). Also always using pin 1 for 12V and pin 2 for GND prevents mistakes. In some cases where multiple similar connections need to be made, it is wise to use different types of plugs to make sure you never connect something in such a way that it breaks the equipment. For example, the fire extinguisher on our battery pack should never be able to be connected directly to its 9V battery, and the battery should never be able to be short circuited. Choosing these connections will prevent critical mistakes when connecting the 3 components:- Fire extinguisher: 1 male blade connector, 1 female blade connector
- 9V battery: 2 female blade connectors
- Thermal switch: 2 male blade connectors
Three phases of separation
When it comes to building an electric motorcycle, there are components you immediately know you’re going to need, but also a lot of systems of which you didn’t have the slightest idea why it would ever be needed on an electric racing bike. Obviously a pair of wheels might come in handy for a motorcycle, and some kind of electrical energy storage on the bike eliminates the need for a very long cable to power all the onboard systems. However stuff like a safety control circuit to ensure a smooth cooperation between the subsystems, and a BMS for monitoring and protecting the battery pack might not immediately come to mind. That having said, there is one component you know for sure you’re going to need. Something to convert the available electrical energy to actual movement. Correct, I am talking about an electric motor.
The motor and motor controller
The motor that NovaBike is using is an EMRAX 268 axial flux synchronous motor, which is shown in the picture below on the left. This is an AC motor being supplied with three-phase, sinusoidal voltages. The stator is the shiny metal part in the middle of the motor, and is connected to the frame. The rotor is the outer, black part and this is the part that actually rotates during operation. This motor is controlled using a BAMOCAR D3 digital three-phase servo amplifier, which is shown in the picture on the right. This motor controller can be supplied with DC power from the batteries, and will output three voltages, each 120 degrees out of phase with each other. These three phases control three groups of windings, which are placed at different angles (120 degrees apart) in the stator, giving rise to a rotating magnetic field. The rotor in this specific motor contains permanent magnets, creating a magnetic field in the rotor with a fixed north- and southpole. This magnetic field will try to align its field with the rotating magnetic field in the stator, thus creating a torque which rotates the motor in the same direction as the rotation of the stator field. In steady state the rotational speed will be the same as the speed of the rotating magnetic field, which in turn depends on the frequency of the three-phase voltages.
Left: the EMRAX 268 synchronous motor; Right: the BAMOCAR D3 motor controller
Regenerative braking
A special feature that we are currently working on is regenerative braking. In general, a synchronous machine can work both as a motor and as a generator. If three-phase voltage is applied the motor will output mechanical energy in the form of rotation, as explained above. If no electrical power is supplied however, mechanically rotating the rotor will induce three-phase voltages, and the machine is working as a generator. Of course in the practical case both processes occur at the same time. While using the machine as a motor there will be a load torque present, which is trying to slow the motor down. Vice versa, if the machine is used as a generator, there will be a load current flowing which will also generate a torque that slows the motor down.
In a normal braking operation the supply current is removed from the motor, and a mechanical brake applies a torque in the direction opposite to the rotation of the motor to slow it down quickly. Removing the supply current will prevent the motor from accelerating, but because there is no current flowing anymore, there is hardly any torque slowing the motor down. That’s why the additional torque from the mechanical brake is needed to bring it to a quick stop.
Instead of only removing the supply current during braking, a load can be attached to the electrical input of the synchronous machine. Because the motor is still rotating, it will work as a generator and a current flows into the load. This current will generate a torque that is slowing the motor down and therefore acts as a brake. By varying the current that flows into the load, it’s possible to control how hard the motor brakes. If this load is a device that can store electrical energy, for example a capacitor bank, then the current used to slow the motor down can later be used to accelerate it again. This is a great feature for a racing motor since it can make the difference between finishing the race or having to stop early because of a depleted battery pack.
To regenerate or not to regenerate
The electric motorcycle that is being built this year will have a regenerative braking option, but the system is very limited. This is because the idea only came up when the design was finished and the building had already begun. Designing an elaborate system would take a lot of time, and more space in the motorcycle would be needed. The current system can regenerate a bit during braking, because the battery pack can also act as a load. However, it is not possible to brake fast while regenerative braking is allowed, because the voltage gets higher when the braking is more abrupt and battery cells don’t like to be charged above their maximum rated voltage. Furthermore the cells also have a maximum charging current which may not be surpassed, so the braking torque is limited.
Still, the first steps are taken towards regenerative braking in electric motorcycles, and we keep searching for solutions to improve our regenerative capacity. Now I’ve got to go, because supper is ready and I have to regenerate my own energy storage.
The CAN opener
- The data acquisition program that runs on the myRIO that collects the data through measurements and the CAN bus.
- The readout program that runs on a computer that reads the files generated by the data acquisition and visualizes the data for the user.
The black box of Variants
By default LabVIEW has a very nice function, called Write to Binary File. This function is extremely user friendly because it accepts any type of data to be written in binary form to a file, from floating point numbers to arrays to the most complex clusters (similar to C structs). This is all made possible by the “Variant” data type, a set of data that contains both the type information and the data itself. The first approach of the data architecture uses these Variants in the following manner. A cluster called DAQ Frame was declared as follows:
It contains the type of data, the timestamp when data was collected and the “Variant”. This variant could then be filled with anything, for example the voltage data, which is a cluster that contains 4 floating point numbers of BMS voltage measurements.
The DAQ Frame cluster is a nice single object that can be placed in the “write queue”. The items in this queue are passed to a different thread that writes the data to a file, asynchronously. Writing a frame to the file boils down to the following Block Diagram:
This (very simplified) block diagram seems to work very programmer friendly, as it is able to write any type of data to a file.
The main idea behind this implementation was that every measurement type that needs to be stored gets its own DAQ Type (which is an enum) and also its own cluster to be placed inside the variant of the DAQ Frame. This turned out to be actually very annoying to work with, mainly because the measurement cluster needs to be known and implemented both at the writing stage and when reading out the stored data file. This resulted in a lot of work to be done for each measurement type that was added. Also, when the measurement type was not yet implemented on the readout-side of the program, the amount of bytes to be read could not be determined. Well, LabVIEW probably could, but because variants and the Write to Binary File functions are really just black boxes, it was difficult to know what bytes were written exactly.
An additional disadvantage of this implementation was that the CAN messages (containing 64 bits at most) were converted on-board to floating point data (32 or 64 bits per number), for example when measuring the BMS voltages. This increased the file size without increasing the information carried by the file.
Know your bytes
More control over the exact bytes written was necessary. In the first place to be sure the bytes could always be read (but maybe not yet interpreted) by the readout software, because the software knows it needs to read bytes instead of a specific type of cluster. In the second place to compress the file size. A quick realization of this architecture was to change the DAQ Frame cluster to the following:
Here, the variant was replaced by a simple byte array. The byte array can be of any size. The DAQ Type property lets the readout software know how to interpret the bytes. Here the bytes will be inflated to, for example, floating point data that is more easily understandable by the user.
An advantage to this system is that it is actually easier to process CAN messages. All that needs to be done is to match a specific CAN identifier with the right DAQ Type and just store the bytes of the CAN frame into the DAQ Frame.
For other measurements that are not communicated over the CAN bus, it is a bit less straightforward to store them. Fortunately LabVIEW has a neat Type Cast function, that does a direct reinterpretation of data. For example the Acceleration measurement that simply uses the accelerometer built into the myRIO, as shown below. I have encircled the Type Cast function. In this case, the 64-bit floating point array is first converted into a 32-bit floating point array, whose binary data is then reinterpreted as a byte array.
Concluding
What I learned from this is that I should not depend too much on already existing solutions. It is generally considered bad to reinvent the wheel, but if that gives you more insight in how the wheel actually spins, it really is an advantage. Apparently the use of LabVIEW’s Variants was not ideal in this case due to its opaque implementation. Controlling the exact bytes written by the program makes debugging easy and the code more efficient. Not only at runtime, but especially at “program-time”.The stack
Two months ago I wrote about the “stacks” in the battery pack. “Stack” has become often-used jargon within Nova Electric Racing for the building block of the battery pack. It contains a certain amount of battery cells connected in series, the BMS in order to monitor the health of the cells and structural support for the cells to keep them tightly in place. The latter is often done by using spot-welding or laser-welding, but last year it was done by soldering the cells onto a PCB. This provides a good electrical conductivity, but also puts a lot of heat from the soldering iron into the cells, which may damage them severely. It is also impossible to replace cells when doing this.
To solve both issues of soldering, we looked into the method of clamping, which is gaining popularity amongst pouch cell based designs, just like ours. In contrary to cylindrical cells, which have to be spot welded, pouch cells contain two thin copper tabs, a positive and negative terminal, which can be clamped between two materials. This method usually required more space, but by using the latest technology PCB components from Würth Elektronik it was possible to invent a clamping method that only requires some extra space in the length-direction of the cells. This extra space is occupied by an additional PCB to clamp the cell tabs beneath.
The latest stack prototype. Thanks to Eurocircuits and their short delivery time we were able to do some quick prototyping with this innovative design.
The major concern of clamping solutions is the possibly increased internal resistance of the stack and in the end also of the battery pack. Before assembling, some measures have to be taken to make sure the contact resistance of the cell tabs with the copper busbars is as low as possible. These measures are sanding of the cell tabs and the copper to remove any oxidation, which has a high resistance, and degreasing the cell tabs and the copper to remove any grease, which is not very conductive either.
The next step in the design process is to calculate the power losses of the battery pack due to the internal resistance. Measuring this resistance is not just a matter of using the resistance measurement function of a multimeter. This is because the resistance is very small, in the order of a few milliohms, and in the second place because a multimeter does not work when there already is a voltage applied on the device under test, which is the case for the battery cells.
However, the solution is not very difficult either. If we can flow a current through the stack, a voltage drop will be seen, just as with a resistor. Now drawing a current is exactly what the stack is designed for, so this is not very difficult. By using a programmable DC load it is possible to control the drawn current very accurately and indeed, when the load is enabled, it is observed that the total voltage on the terminals of the stack drops slightly. Using the well-known Ohm’s law, the internal resistance of the stack is the voltage drop divided by the drawn current. The linearity of this equation even allows to extrapolate the obtained resistance to the complete battery pack.
After we carried out these measurements and calculations, in total the estimated efficiency of our battery pack is 97% when full to 96% when the battery is low! With this positive prospect in mind, it feels great to continue with this design and implement it in our motorcycle.
A part timer's experience at nova
My eagerness to use and enhance my knowledge in power electronics has led me to join Nova last November as a part-timer in Power Train Department. I find power electronics to be a versatile technology to contribute to the fight of the energy crisis and climate change. One of its application is the electric vehicle which Nova is working on with its Electric Racing Motorbike. Although the courses I took at my master program in TU Delft are very extensive and has a lot of hands-on experience, working at Nova still gives me a lot of learning point which makes my ten hours per week worthwhile. In this Nova’s second electric bike team, I was involved in fault analysis of motor controller breakdown, battery cell redesign and currently working on Safety Control Unit (SCU).
In November, I was directly put into a team to investigate the failure of the motor controller due to unknown faults in the last experimental test. At first, as someone who never works intensively with an electric vehicle and barely saw Nova electric motor design, it was very hard and abstract to see what went wrong. In the limited amount of time which I have between my master project, I tried to absorbed what was available in a well-documented design folder of Nova. I also proposed to the team to use fault analysis tree and tried to work what went wrong from there. The team was quite open to suggestion and the discussion resulted in better understanding of possible failures that could happen in our system albeit the root cause of the fault was hard to conclude due to an insufficient amount of data from the measurement system. The results of fault analysis are being used as an input for redesign improvement in this year electric motorbike.
Due to the fault, the battery was no longer in good condition as well. We use this opportunity also to redesign the cell to have more sufficient capacity and power to win the race. The process of redesign involved many tricky trade-offs from selecting the right cells, putting it into limited space, and many other aspects. My role in this redesign was to assist the full-time engineer by processing the cell information from a different perspective, exploring the possibility and implementation of regenerative braking and participating in the decision-making meeting.
The current responsibility given to me is to design the Safety Control Unit (SCU) of our motorbike. The power train of our bike is mainly composed of three building blocks; battery, motor controller, and electric machines. Battery Management System (BMS) is responsible for monitoring the battery cell by cell. The motor controller is a pair of electric machines which maintain its own and the electric machine safety. There are, however, several reasons why SCU is required in our system. First, BMS cannot handle the battery protection on its own because it requires external action to influence critical parameter in its systems such as discharge current and other electrical parameters. Thus, it needs to communicate to the motor controller, relays, and other protection devices through SCU. Second, the interaction between building blocks results in an unsafe region which does not exist if each component works alone. An example of it is a special procedure required to shut down the power train in case of an emergency which needs to be regulated by SCU. The relation between SCU and other components are presented in its I/O block diagram below.
Schematic of the Safety Control Unit
In four months, I was working in fault analysis, battery cell redesign and safety system design. Along the way, there was challenges and learning point such as communicating with a different culture and educational background, adapting with the highly iterative design process and time management between my master program. If there were a regret, however, it is that I did not join Nova sooner.