The journey has continued. Yesterday we visited Copenhagen and Gothenburg, and today we stood by Sergels square in central Stockholm. Maybe some of you saw us there.
As usual, there’s not much time to write about our ventures, but there are images to show. So stay put and we will get back with some more info at a later stage.
.
It is late and we really have no time for anything right now, so we’ll just present the last couple of days in photos. There will be more coming, soon, whenever there is time.
.
Let’s start in the bad end. Just a couple of hours ago the whole thing seemed to be a disappointment.
We went off early to get a good position in the gigantic queue of vehicles waiting to participate in the race. Since there are so many breakdowns and failures among the participants there’s no idea to have a fixed starting order. Instead everybody just waits in line for their turn.
The first race for the Skilži showed some improvements: 650 km/l. This is a more than 20% increase, but still nothing extraordinary in the race as a whole. Immediately as the vehicle came back to the paddock we made some minor improvements and put it back in the line again.
Baldos II had an early start as well. We had been testing it all night – running the ICE to charge the battery, unloading the battery, running it again and recharging etc. This to make sure that the vehicle could handle all the repeated starts and stops that signifies a real race, and where we by some reason always suffer wear-outs. Because of the testing, the ICE did fine during the whole measurement. Instead something else failed.
We haven’t really understood why, but after 6,5 out of 7 laps the electrical motor broke down. Erika tried desperately to start it, but failed, and Baldos II had to be carried back to the paddock. A great disappointment, especially since we didn’t know if there would be time for another attempt.
At this stage everything looked bad. Then it got better.
While waiting in line we did some small adjustments to Skilži. We improved the wheel suspension by aligning the wheels (improving the toe-in, so to speak) and preheated the engine just before takeoff. Although we know that a common engine drains a lot more during a cold start than during a warm, the results were extraordinary: 1201 km/l. Almost double the previous result after just a few adjustments. Imagine the amount of ideas for improvements that suddenly popped into the minds of the team members.
And what’s even better – Baldos II finally, after five unsuccessful attempts (three in 2009 and two in the last few days), made it to the finish line! Once we started the vehicle after the second attempt, it ran like a Dalecarlian grandfather’s clock. So we just put it in line again and rolled out to the track, and it just kept going. The look seen in our faces at that moment was sheer joy. The result of 137 km/l wasn’t that good at all, but we had broken the curse of the red pearl, and we are still the #1 vehicle allowed for street use that we claim to be.
We’re right now waiting for the award ceremony. We will put up some photos in a while.
.
Baldos II on the starting line of SEM2010
So, it happened again. Just before the fifth artificial stop during yesterday’s measurement, Baldos II broke down. Luckily the energy in the double-layer capacitor was enough for Erika to drive it to the paddock on her own.
As described, the ICE on the vehicle is merely used to recharge a battery (a double-layer capacitor to be specific). Whenever the energy level in the battery drops beneath a certain value, the ICE kicks in and recharges it. The ICE in Baldos II typically runs for 5-7 seconds, then turns off and the vehicle is driven silently on the electrical motor for 30-60 seconds, then the ICE ignites again etc.
Baldos II during the first measurement
During the race Erika discovered how the ICE intervals increased from 5-7 seconds to 10, then 13, 16. This is a sign that the engine is running with low efficiency – it takes longer and longer for it to produce the same amount of energy in the battery. Erika also noticed how the engine behaved unusually irregular. This escalated until just before the artificial stop (a short stop in the middle of the track to resemble e.g. a red light) on the fifth lap, as the ICE kicked in and just kept running. At this point Erika realized it would be impossible to finish the race, so she turned the automatic engine control off and carefully rolled over the finish line.
Once again the letters DNF – did not finish – appeared in our result sheet. Something we’d hoped never to experience this year.
A rain covered Eurospeedway Lausitz
It is a part of human nature to immediately start generating theories for what happened. Jonas thinks it was all because of deformations on the crankshaft, which in turn drove the pistons towards the cylinder walls. This would not be the first time that happens – the crankshaft for this engine type is quite weak (despite being made of titanium). Fredrik blames it all on mistakes in the ignition module; especially on higher rpms. Erika thinks it has to do with the temperament of Baldos II; he’s just not up for races. Trying to satisfy as many of these theories as possible, we continued working on the vehicles during the evening and the night.
Older model mechanical drivetrain in Baldos II
During the night the winds increased and the expected rain reached us. There was kind of a commotion on the campsite as people from other teams tried to improve their multi-colored plastic shelters to keep everything from getting soaked. Team Baldos, on the other hand, slept comfortable in their rain-proof Tentipi tents.
This morning the weather gone worse and the safety responsible decided to postpone today’s races. There was thought to be a chance for some clear weather in the afternoon, but later on it was announced that there would be no races today. Tomorrow, on the other hand, we would start off early.
Dr. Matthias Bichsel inspecting Baldos II
This has given us some extra time to fix everything up. Apart from continuing the improvements from last night we’ve also cleaned up the whole paddock and allowed for those with the worst sleeping statistics to get some rest.
We’ve also been practically invaded by visitors who’ve heard about Baldos II and wished to see it with their own eyes. Some of them were journalists from e.g. France (TV: TF1) , Germany (TV: N24), Bulgaria, Ukraine, Holland, and Sweden. We’ve participated in a live TV webcast, and tomorrow we’re one of four teams chosen to give our views in a formal debate about future energy efficiency. We’re also glad to announce that Dr. Matthias Bichsel, member of the executive committee of Royal Dutch Shell, joined the prominent group of people who’ve tried out the ergonomics in Baldos II.
.
Skilži during one of the test runs
All previous posts have been in past tense, but considering the race and how fast everything is developing, we think it’s about time to move into present.
One of the first things we were told of when we arrived yesterday was that the races had been put forward to this day, Wednesday, since Thursday was expected to come with bad weather. This morning and early afternoon was used for test runs, as was planned, but at 15:00 they stopped all testing, and at 15:45 the first actual competition rounds were made.
During the night we worked with the rpm and the oxygen sensors (lambda) since they were giving faulty values, and made some other shape-ups and last minute configurations. We did a few tests and tryouts and everything actually seemed to work out.
Skilži on its very first official measurement
In the morning we were among the first to visit the safety inspection. This is done to ensure that the vehicles are safe to use in the competition, but also to make sure there are no irregularities with our equipment. The vehicles are put on a 20 degree slope to test the brakes, the engines are started and immediately shut down to test the emergency stop device, and the driver has to prove s/he’s able to escape the vehicle within 10 seconds, among other things to ensure that there’s no risk in participating. Both Baldos II and Skilži passed flawlessly, the inspectors even smiled a little when Stina managed to come out of the car in less than five seconds.
The coming test-runs were also flawless. Both Baldos II and Skilži completed several laps without any issues or breakdowns. The systems were not perfect but worked alright, and compared to last year we felt confident in that we’ll at least get a result.
Which we did!
Baldos II just after start
Skilži just completed its lap and made it to 575 km on the equivalent of one liter of gasoline – the first result officially measured with that vehicle. That is a really good start-off point for coming measurements.
Baldos II initially had some trouble. During refueling just before race start, someone broke the fuel tank. We ran over to Shell shop to order a new one, but were informed it could only be done online. So we ran back to our computers, ordered another one, ran back to the Shell shop and got it, back to Baldos II and immediately installed it. Luckily the fuel tank is easily detachable and in a few minutes we were back on the queue for the refueling. Right now Baldos II with Erika behind the wheels is on its second lap on the very first measurement in the contest.
.
Monday morning: Packing the cars
Apart from many other projects – ours has a distinctive climax where months of work is evaluated in the lapse of minutes – kind of like an athlete participating in the Olympics. This is what Shell Eco-Marathon means to us. In the morning two days ago we began the journey from Luleå to Eurospeedway Lausitz in Germany, where the competition is held.
The last two weeks of work in Luleå has been filled equally with excitement and worries for the coming journey and the race. Looking at the statistics it all looks kind of promising though. In 2008 and 2009 the car was put together for the first time in the paddock at the race. This year both cars are not just running but also measured for fuel efficiency (twice) which of course is a giant leap forward when it comes to our chances of achieving a good results.
Monday: Cameron taking photos of the Höga Kusten Bridge
To enable this we’ve had great help from the Airforce base in Luleå (F21), and the Student’s sports association (StiL), who’ve borrowed us wide spaces where we’ve been able to try out our vehicles. We also have to mention Brand & Fordonsskydd from Älvsbyn who’ve supplied us with small fire extinguishers. Since we went from manufacturing vehicle parts into driving the resulting vehicles, we’ve highlighted driver and observer safety in our overall safety planning. Safety has always been our top priority, and whatever results we get they would be worth nothing if it would have compromised the wellbeing of a driver, project participant or anyone else.
Two days ago all of our preparations shifted into actions. Monday morning we gathered already at 06:30 to have an early start of our long journey. That didn’t happen. Instead we spent the next few hours battling bulky packs (we now know that it’s really difficult to pack something two meter long and drop-shaped), trying to get a hold of roll-ups, searching the workshop for parts needed for the rims, discussing what and who to place where and why, and finally going over the pack list and the list of things that had to be done before take-off, over and over again. At 08:45 we finally left campus and headed south.
Tuesday: A quick stop on the way
We travelled in three cars and one small truck (containing both Baldos and Skilži). Since most of us had different agendas of what to fix on the way down – people to see, places to go, things to buy – we decided to go separately.
Let’s follow one car as an example. I (Kristian) ended up in the BMW together with our exchange students – Cameron and Christophe – and one of our locals – Joakim. Since Cameron is Australian and used to left-hand traffic he was a bit uncomfortable driving in the beginning, but after doing his first takeover he eased up and everything went smooth.
So, what might four 20-25 year old guys, sitting together in a car on a long journey, talk about with one another? Here are a few examples:
Tuesday: Passing by the woods north of Berlin
The trip consisted mostly of socializing in the car. We made our very first stop in Piteå to pick up sandwiches made by Joakim’s mother (thank you). At 14:00 we reached Örnsköldsvik and took a quick break for lunch. After that we more or less travelled constantly. At 20:30 we had an appointment with Mahsa from Shell (thank you for the fika) and had dinner on a bench outside a Shell station in Stockholm. The night we spent in the car travelling through Västergötland, Småland and Scania – and there’s really not that much to say about it. One of us were driving, another handled the delicate task of entertaining the driver, and the two others slept waiting for their turn.
We arrived in Trelleborg, the harbor for our ferry to Germany, at around 04:00 in the morning. To our surprise all of the other vehicles arrived within an hour. Considering previous setbacks in the project, the ease of the transportation through Sweden was almost miraculous.
We had a few quiet hours on the ferry, where most of us tried to catch up on lost sleep during the previous night (for most of us; nights).
Tuesday: Arrived at Eurospeedway Lausitz
As we arrived in Sassnitz – well – we just continued driving. Cameron once again sat behind the wheels and led us out on the Autobahn and further into the continent. Going at between 140 and 200 kilometers per hour, we moved a lot faster than the maximum 110 km/h allowed in Sweden.
The landscape was beautiful. On our way south we had seen how the snow on the ground in Luleå transformed into green grass, then coltsfoot (Tussilago farfara), windflower (Anemone nemorosa), then budding leaf trees. When we arrived in Germany the trees were all green, the fields had received their first crops of oilseed rapes (Brassica napus), and the woods north of Berlin reminded us of the season to come in Sweden.
At 16:00, after a 2030 km car ride, we arrived at Eurospeedway Lausitz. We split up – half of the group setting up the tents and the other half setting up the paddock and working with the cars. The latter continued during the whole night.
.
Fredrik soldering the circuit board for the ignition module
Not everyone in Team Baldos is a final-year master’s student. Fredrik is one of these exceptions; he already holds a bachelor’s degree in automotive engineering. Done with his thesis in electrical valve controls, he is back in school again to reach master level.
While doing this he was assigned a project by Johan to construct an ignition module for a small engine. This happened to be our engine, and middle of February Max introduced him in a project group meeting, and afterwards he became another member of the Team. Since then he’s been involved in several subprojects – he’s helped Kristofer and Erika with the computer nodes, Max and Sebastian with the ECU, and Jonas with ICU testing – but his primary concern has remained the ignition module for both ICE:s.
An ignition module might seem like overcomplicating a standard feature in cars. Most often the sparks appear whenever the cylinder has reached a specific position; creating that kind of system is fairly simple, and Fredrik’s ignition module does that as well. But to build a truly efficient vehicle requires a little more effort.
Janne, Daniel, Harald and Sebastian testing Skilži on our improvised dynamometer
Now, apart from creating sparks, our ignition module repeatedly measures the ionization level in the combustion compartment. This is done to determine when, during the cycle, the cylinder pressure reaches its maximum. The trick is to get the maximum cylinder pressure to coincide with the exact instant for when the piston is most responsive to impact (around 14 degrees after TDC, for those interested). What makes the whole thing so tricky is that there’s a delay between ignition and maximum pressure, and to compensate for the delay the ignition module continuously shifts the spark time back and forth.
Talking about time – this all happens extremely fast. Doing 6500 revolutions per minute, the interval within which the module can adjust the spark is about 4 microseconds. To get the whole thing right Fredrik’s system does about 120 thousand measurements per second. As seen, the project requires high-quality electrical components, like those supplied to us by McXpress.
The finished printed circuit board in Baldos II
Another thing to consider is a phenomenon called pre-ignition. This happens whenever the fuel-air mix ignites before the spark has fired. This often occurs because of a too high cylinder pressure or hot spots in the combustion chamber. Pre-ignition is harmful to the engine, and therefore the ignition module is designed to detect the phenomenon and automatically compensate for it.
When working with such extreme tolerance levels there’s a need to keep things in order and under control. Fredrik has developed a habit for checking, double checking and triple checking so that everything fits together and works as it should. There are other reasons for that than maintaining engine performance. A casual spark is at around 40 000 Volts, and although the amperage is rather small there’s no telling what will happen if it reaches a human finger. Fredrik has so far done a good job in avoiding this.
.
Cameron and Mattias having a look at the old alternator prototype
Cameron, Nina and Mattias have spent most of their time working on the alternator. Cameron and Nina are both exchange students: Cameron comes from Monash University, Australia, and Nina from University of Ljubljana, Slovenia. Our last exchange student – Christophe from the University of Technology of Troyes, France – has been focusing on mechanical systems. Mattias comes from Råneå.
The alternator is the thing in Baldos II that transforms the mechanical energy from ICE into electrical energy by charging the electric double-layer capacitor. It is a type of generator that only produces Alternating current (AC). We already have one of these in the car, but that one has a maximum efficiency of around 80% and was designed to be an electrical motor (i.e. doing the exact opposite of an alternator). The aim for this year has been to surpass those 80% – with some efforts it could reach up to 90%. This 10 % difference would increase the efficiency of the whole vehicle with 12,5% – which is a significant raise for an already high performing vehicle like Baldos II.
The construction of the alternator is probably the single process that has suffered most from massive design changes at late stages. This is a stepwise description of what has happened.
Equipment to do copper coils by hand
First attempt: The start off point for the whole project was to create a sound magnetic field. An alternator typically consists of two things. First, magnets rotating together on a “rotor”, creating a magnetic field that spins around an axis. Second, copper wire coiled and attached on a “stator”, which responds to the changes in the magnetic field by producing current (i.e. electricity). The initial idea was to modify a prototype alternator (which was built but never completed during last year’s project) to create such a superior magnetic field.
First setback: While doing calculations on the efficiency, the trio discovered that the alternator they were about to build would generate 50 V already at 700 rpm. Since the ICE would run at somewhere around 5-7000 rpm, the voltage would be far too high. Also, the old prototype was a rather fragile construction, and the round shape of its magnets was far from ideal.
Second attempt: So, Cameron, Nina and Mattias abandoned the alternator from last year and started out with their own prototype. They also found some bar magnets and decided to use those as a guide for the whole design.
Initial (left) and final (right) rotor design
Second setback: But after a while working with the bar magnets they got cold feet. By putting bars in a circle (see the image on the left) they received open spaces on the outer rim of the rotor. These spaces would have a relatively low magnetic field strength and thereby produce very little current. The only way to make these gaps negligible would be to run the alternator at extreme rpm:s.
Third attempt: There was no way out but to change the shape of the magnets, and thereby the whole design of the alternator. A big issue was to get a hold of such permanent magnets – it was already quite late in the production process. Luckily Vacuumschmelze supported us with new ones (a big thank you to Stork Drives as well, who supplied us with an Incremental Rotary Encoder). These magnets were shaped like circular sectors, and the efficiency calculations suddenly looked a lot more promising than before.
Third setback: So far the trio had only concentrated on optimizing the magnetic field, but after a while they realized that they had neglected the second part: the copper windings. When they compared their design to their efficiency calculations it was clear that there simply wasn’t enough space to fit in all the necessary copper.
Fourth attempt: Fortunately there was a way to solve this. All of their previous work focused on a single rotor (magnets)/double stator (copper) approach. If they instead switched to double rotor/triple stator it would allow more room for copper. The fuel efficiency calculations gave thumbs up, the computer model stated that everything would fit in physically, and even producing the parts appeared manageable. So that’s where we are today.
The final rotor design in the CNC
Although it was painful, Mattias, Nina and Cameron are happy to have passed through all of these stages. As they see it there’s no way they would’ve reached any results with their initial approach, but now they’ve gone through their solution so many times that they feel quite confident in it.
This is what we call the “Alternator paradox” – to balance the need to investigate and improve a design solution theoretically with the time needed to produce it. Among engineers there’s often an inherent aspiration to achieve excellence; but it has to be put in relation to when it’s time to stop developing and start doing. This issue accompanies everyone in the project, and is vital for the final output of the team.
According to the latest calculations, the alternator will reach a stunning efficiency of 94,5%. But will it be completed? The coming days, evenings and nights will tell us.
.
As you may know, we had our unveiling ceremony last Thursday. We invited our sponsors, some press and everyone at the university to see our brand new vehicle for the first time.
Elisabet unveils Skilži
Unfortunately, none of us has been set on “update homepage”-mode since then. Instead we switched right over to “final testing”-, “summing up”- and “packing”-mode; which is understandable since today is just four days away from takeoff. And, not very surprising, we might also have had a fling of the old “work as a nut to get the last pieces together”-mode, again.
Anyway – about the unveiling.
Above all, the ceremony meant the death of our mysterious “Baldos ‘X’” and the proud birth of our new prototype vehicle, Skilži. Skilži is pronounced something like “skilDJCHÍ” and means “icicle” in Sami. Baldos means “ice floe”, which is meant to associate the gentle movement of the vehicle with the almost frictionless motion of ice travelling on the surface of cold water. Skilži follows the same winter theme as Baldos but refers to the shape of the vehicle rather than the motion. The basic idea for the designers, Martin and Micke, was to make it like an extended drop of water – the shape with the least wind resistance of all. And, as you know, an extended and solidified drop of water equals an icicle.
Skilži, our new prototype-class vehicle
It was a good feeling to present our vehicle to everyone after so much time keeping it a secret. You can see a video from the unveiling here. Hopefully it’s going to serve us as an experimental vehicle for trying out new engine concepts and even more advanced solutions for fuel efficiency, which then can be applied in our standard car – Baldos II.
Another thing we did for the first time was to show Baldos II in city traffic. After all it is a street-legal car, so it was just a decision concerning “when” rather than “if” the citizens of Luleå would see it on the streets. The university office of communications made a nice video of the test drive which can be found here.
Baldos II in Luleå city center
We will list some of press that we received around the event in the “press” page soon. Among others we had the Swedish national television’s Rapport here. We will also send out a press invitation for the Baldos on Tour – read more about that here. Also, there’s a few more lessons learned from production to be told in the coming few days – next time about Mattias, Cameron and Nina’s work on the alternator. Stay tuned.
.
Niklas and Daniel during assembly
Niklas, Stefan and Jonas have all been working on the drive train for both vehicles, which basically means everything between the gas tank and the propelling wheels. Jonas has concentrated on ICE improvements, Stefan on Baldos II’s drive train and Niklas on Baldos ‘X’s drive train. This is the pumping heart of the cars and of course vital for Team Baldos ability to compete as a whole.
The drive train for Baldos II has changed a great deal since the test-run in January. In fact, everything but the alternator (the alternator generates electricity and charges the supercapacitor) has been exchanged, and it won’t be long until that one is upgraded as well. The drive train has a more intuitive distribution of parts, a new position in the vehicle and is a lot more stable than before. As with Baldos ‘X’ it is mounted on a plate to make it easy to detach and configure outside the vehicles. All the brand new components – the ABB Torductor, the new type of alternator and the Tesla exhaust turbine – have been given space in the engine compartment. The former system for the starter engine, which was a typical cause for breakdowns, has been redesigned (the new starter engines were supplied by Asia Motors). Combined with Jonas complete revision of the ICE (helped by Oerlikon with a coating to reduce friction) the drive train both looks and works smoother than ever before.
The Baldos 'X' drivetrain. NuVinci down left in the photo.
Since Baldos ‘X’ is new for this year, its drive train had to be designed and manufactured from scratch. The biggest issue while doing this was to squeeze everything in to such a limited space – there’s only about 350 x 400 x 300 mm for everything needed to move the vehicle forward. Stefan and Niklas worked for a long time with the CAD-drawings trying to puzzle everything in. This was partly solved by a clever positioning of the bearing supports (bearings have been sponsored by SKF) to require minimum space. During the design, functionality and low weight has been prioritized, but the results also look kind of good. Overall the Baldos and Prototype has been given similar characteristics in the drive train to ease production and to use knowledge from one system in the other.
The typical work process while doing this has been to start off with a CAD-drawing, sometimes backed up with FEM calculations. Thereafter the finished blueprints were transformed into physical products in the workshop, or handed over to Janne to do the CAM preparations and actual CNC operations. Stefan and Niklas later assembled all produced parts into the finished system. Most parts have been made out of ALUMEC supplied by Uddeholm.
Stefan in the workshop. Mascot in the background.
A feature worth mentioning in the drivetrain is the NuVinci hub. It’s a hub with variable shifting levels – enabling good efficiency values and decreased tear of the engine while shifting (more information can be found here). Valvoline has sponsored us with traction fluid for the NuVinci. Apart from that Niklas and Stefan have also helped with various driver’s controls – shifting, steering and brakes (springs for the brakes supplied by Sodemann). They’ve also designed the fuel supply system – Bosch Rexroth helped is with various hydraulic components for that.
Although Jonas, Niklas and Stefan set out with the intention to start with production really early, it turned out to be too late either way. The reason is that everything linked to detail design and production has taken about double the time expected. If they’d have the chance to do it again they would have started even earlier – both of them have been working day and night to get everything ready. This has meant putting their focus on parts that work rather than parts optimized for the race. Next year’s Team Baldos may receive the privilege of optimizing the engines rather than designing and building it, which could be a real pleasure considering the massive groundwork already done.
.
