Trip to Concord!

This weekend we took the eBike out to Concord MA.

We woke up early and loaded up the truck then headed off to Concord!

Nick and Joey loading up the truck.

Victober communicating with the Kelly Controller in the cold.

We ran a roughly 10 mile track over around 30 min (lots of starting and stopping). Here is the track view:

After the track our small battery pack was almost out so we headed off to a pizza place to fulfill another EVT tradition...

Charging and Eating Pizza!

Charging took longer than expected and we were almost out of time so sadly we didn't have any time to do a second run, but we were pretty happy with the handling of the vehicle under the first loop.

One step closer to NYC trip!

New Bike, New Brakes

After the testing in the hallway and in the alley outside N52, we realized we needed a more reliable bike. The aluminum Geoby bike was a nice bike but it was definitely not designed to handle the loads we were putting on it. We had no issues with it but before anything happened we decided to switch to a steel frame, disc brake bicycle. Luckily Lennon had a Surly Disc Trucker he was willing to let us use for the project.

Joey working on the new frame.

We moved all the electronics onto this new frame and we also replaced the stock brakes with very nice high end Shimano disc brakes with integrated heat sinks. Rango, Lauren and I also decided to revamp the electrical wiring system with better connectors and a better more logical layout.

new controller-motor wiring system

New Suspension Design

During February we underwent a big suspension system redesign. Our previous design used 6 springs, and 12 brackets. It was a very machining intensive design, furthermore it was unconstrained and hysteretic (when the suspension was pushed on, it wouldn't spring back up because of jamming issues due to the high number of joints).

Johnathan and Nick came up with a neat redesign using only two shocks and a pair of hinges. The idea is to emulate what you see on a motorcycle or bicycle rear suspension where you have a combination of a pivot and a spring/shock that absorbs the loads. Nick even designed it so that it would be easily waterjettable with only some slight machining needed.

New suspension design with a fancy FSAE shock.
Notice the heavy duty hinges towards the front.

This new suspension gets us a little step closer to a full system test. We are currently planning a trip to Concord sometime during March to see how our vehicle handles in the real roads of Massachusetts.

TechFair 2013!

EVT participated in MIT TechFair 2013!

We took our eBike and the pocket bikes! We even teamed up with MITERS and did an EV parade around the building.

Lauren on the eBike!

Connector Meltdown

We were out testing the eBike today and while starting up we saw some smoke! Immediately we cut power to the bike and stopped. This is what happened:

TOTAL CONNECTOR MELTDOWN

The connectors melted and shorted each other terribly (thus the smoking), interestingly when the power was off and we were pushing the bike forward we saw some crazy vibrations from the front wheel, this was due to the back EMF of the shorted motor.

The connector failed terribly, these are supposed to withstand 50A but they obviously do not live to the challenge. I phoned the company and they said they had seen this before and blamed it on the connectors that come from the manufacturer, they suggested we switch them with better connectors (which we did). The motor seemed that it undergone no severe damage, we spun it back up and it ran smoothly as before.

Wooden Electronics Prototype

After working on the suspension for the last couple of months during IAP we decided to get the system working with our 72V batteries.

We put together all the power electronics on a big 3/4 inch piece of plywood and wired one of our Geoby bikes to it.

The electrical system has a 12 V system powered by the Vicor DC/DC converter, this system powers the contactors. It also has a pre-charge switch and a killswitch that closes the contactors from the handlebars. The wooden prototype was nice because it allowed us to see the different parts of the electrical system clearly. We also got really good practice at de-bugging LED error codes from the Kelly Controller as well as playing the hall effect sensor game numerous times!

We used a big plastic container filled with three 80lb sandbags to simulate the future battery weight. 

Here is us working on the prototype:

We were very excited when we got the motor spinning (finally!)

Notice the big black container (where we stored the cement bags.

After it was "done" we took it out into the hallway for a classic EVT tradition, the maiden crawl in the N52 hall! While doing this we found out we needed to rethink our wiring strategy and add many quick disconnects to the system to make it much easier to assemble. Rango and I ordered some fancy MOLEX automotive connectors and made the wiring much more reliable.

Making sure everything was wired up correctly!

After ample testing indoors it was time to battle the Boston winter and take eBike out for a spin in the road.

Nick suffering in the cold!
No electrical system yet, we were getting used to the suspension and the weight.

Here is a video of Rango out in the cold on the eBike!

IAP 2013 was extremely productive in getting things rolling (literally!). Our next step is to continue testing and hopefully go out and do some long range testing outside the city!

Suspension Update and Range Modeling Intro

Last week Roberto outlined EVT's new project and how each week a different a different team member will overview the weeks progress and their personal contributions. I'm Rango and this week I'll talk about range modeling but first, the progress of the entire group this week!

Bike trailer with suspension clamp

The trailer that is going to haul our batteries is an unsprung mass so if we run into a bump or a pothole, we are sending a large jolt through the batteries as well as the rider. To prevent this, we are designing a suspension system to dampen these sudden impulses. All of the detailes of this will be better explained by someone on the mechanical engineering team, but for now I just want to share saturday's work in the shop.



The picture shows our fancy trailer and our shock mounting system. Because we don't want to drill into our trailer, the mechanical engineering team designed a neat clamping device using two 90 degree aluminum extrudes. The point is we have a working prototype of part of our suspension made from scrap material and we will soon buy material to machine the final suspension! (Again, the details of the suspension will be described in the near future)

But now, back to range modeling:

New York is a long way away from boston. A very long way. We have 230 miles of road ahead of us so it is important to know our expected battery usage, and not just in a back of the envelope sense. I'm developing a model to determine the state of charge of our system over the course of our journey. Of course, it isn't possible to simulate the exact conditions on the bike or the rider's driving style but we these simulations will lead to insight about average behavior of the bike as wel as optimal driving strategy.

So then the obvious question comes up: how does one actually model this bike system?

The model really breaks down into three coupled systems that can be coded and verified independently.

First, we have the physical forces acting on the bike. The torque applied by our motor through the wheels, aerodynamic drag, rolling resistance, change in elevation, and acceleration are part of this sub-model.

Then, there is the electrical model including our batteries, motor controller and hub motor. This accounts for the inefficiencies involved in electric motors and their control systems.

Finally, we have changes in physical conditions as we travel our route. Changes in elevation over the course of our route as well as traffic conditions fit in here.

Over the next few days I'll detail the model for each of these subsystems