We have a new batch of PCBs, issue fixed. The Greenpower teams who ordered the last one have already had the new boards, and we can now send out new kits. To help with the build process I have just posted a video showing how to put together your eChook Nano v1.2 board.

Since the first prototype we’ve answered lots of questions through email and messages, and always wanted these to be more public so they could help others with the same or similar questions – something like a forum! Since the GP forum is unfortunately closing, it seems even more relevant to have somewhere to discuss all thing eChook and help each other out with anything from building the boards to analysing that spreadsheet full of data. So for further discussions, we’ve set up http://echook.boards.net. It’s a little empty right now but I’m sure we can start filling it up.

We will also be writing articles for Greenpowers new ‘How To’ section, both eChook specific, and more general ways of using an arduino with your car, so hopefully the eChook forum won’t detract anything from the greenpower site.

 

 

Disappointingly we are having some issues with some of the boards from our first stack of eChook Nano v1.1’s. So far the quality of boards (v1 and other projects I have worked on) from dirtypcbs.com has been great, their cheap and fast service always delivering good quality boards without issue for my needs.

I was initially a little displeased when I received this latest batch, 12 eChook Nano boards, the first kit board we would send out for schools to test. I had hoped the schools would mostly be testing our documentation not our board quality!

The registration of the silk screen is quite poor. This means that the silk (black lettering/numbering) was not nicely centred on around the components. While this is annoying and leads to a less professional looking product I didn’t think of this as a major concern as it would not impact functionality and the board that was put together showed no signs of any issues and passed all of our tests.

Looking in more detail at some of the other boards received (work have recently purchased a microscope for my lab) I noticed that it wasn’t just the silkscreen that wasn’t on centre. The pictures below show the extent of the problem quite well, with the plated pad area (shiny metal) clearly not centred in the area where the solder resist should be removed (the black area). The solder resist on these boards is the white stuff on top, this resists solder and makes a PCB much easier to solder together as it will stop the solder flowing, in our case down the tracks and potentially in to other components. Specifically here, because the board has a large ground plane pour, it will prevent a solder bridge of the PCB pad to the ground plane as the ground plane is hidden below this protective white layer of resist.  This is important as (for all track apart from the ground pads), a short to ground will result in either incorrect readings of the sensor data, or worse, a damaged PCB.

Unfortunately…this hasn’t happened in this case. You can see in this lower image the tip of my multimeter probe, just skirting the edge of the solder resist. In this case I am checking around the 12V battery pin. Around the very edge of the white region I was finding my multimeter continuity test was  showing this region as connected to ground, not good. This means that the ground plane is slightly showing through the solder resist. If I were to solder this pin it is likely that this would be bridged and cause a short between the 12V input and ground.

Can this all be blamed on the manufacturer of the PCB? Simple answer, no! Although the offset here has resulted in an issue I (as the PCB designer) could have taken steps to reduce the likelihood of this occurring. What I now realised I should have done is reduce the size of the area that is removed for the solder mask, effectively making the solder mask (white stuff) bigger and closer to the plated pad (a smaller black area). Doing this without changing the clearance of the pad area to the ground plan, this would have the effect of covering up a larger amount of the ground plane so even when the registration is poor between these parts of the PCB it is less likely to leave exposed ground plane showing.

An annoying blog post to have to write but a valuable lesson either way and certainly something that has added to my learning of the finer details of PCB design. The team will be modifying the gerbers and getting some new boards in to rectify this issue as soon as possible so we can get the kits out there!

We’ve finally got our (hopefully) final prototype boards together! The eChook Nano v1.1

These boards have a different layout, different DC-DC Voltage regulator with short circuit protection and reverse polarity protection, a lower component count and design changes to make them easier to solder. We have also differentiated the power connector to all the rest so that it is far harder to accidentally fry the Arduino! The functionality remains the same as the original prototype boards.

Thank you very much to the teams who gave us feedback on the initial boards 🙂

We are also re-organising and re-writing a lot of the documentation to make it easier to follow. The build instructions for the eChook board are complete, I’m currently writing up the section on connecting the board to the car and instrumenting the car.

We are providing these eChook boards in kit form, for students to build and program. All instructions are provided in the documentation. We’re still selling them for £25 posted for the kit shown below. You will need to purchase the LEM HAIS 50-p current sensor(~£18) separately. This is cost price – we’re not aiming to make a profit.

*Kit Photo missing a diode and length or ribbon cable that are also included.

Alternatively if you are lucky enough to have PCB making facilities, all the files are hosted at https://github.com/eChook and you can make your own 🙂

A fantastic day at Rockingham saw the weChook racing team manage to run an echook nano for 2 hours on our car Electric 2Galoo. Thanks to all who came to speak to us on the day about the car, as always it’s a pleasure to talk to like minded people who just love to build race cars!

We had great success with out 2 hour test using the echook board both as a motor controller driver board as well as a telemetry data logger. The phone screen was used to inform me (the driver) how much current we were drawing and was perfect for testing our race strategy plans. We also managed to collect 2 hours of good data from our sensors, result!

David @cullimoreracing has kindly offered to run an eChook Nano on Jet2 but we will be looking for more test cars to get these boards on to for the start of the season. Before we do do this we need to tidy up our documentation and work out how we are going to fund the project at this point (the boards are cheap but with components the cost is looking to be ~£30 before you add on the £17 current sensor…we aren’t trying to make any money from this project but we are also not trying to lose too much, we have racecars to build!). We have 9 more boards from this initial batch, watch this space for what we plan to do with them….

Here’s a video describing the main board features

Now that all our panels were ready we are ready to start gluing the driver cell together. The important part here appears to be ensuring that you get as large a contact area as possible. The tabs on our design increase our contact surface for a start. We can further improve the bond of the chassis by ensuring that the panels pull together as tightly as possible to minimise the gaps between them and ensure the epoxy we are using can do it’s job. To achieve this we have found that ordinary wood screws actually work really well in to the dense Divinycell H80, even better than the Rohacell we had previously used in 2galoo’s design. We put washers (and when I ran out of washers all number of metal spreading plates we could find!) on the heads of the screws which helps spread the load and not crush the foam.

We have used a fast curing epoxy for this so that hopefully we can apply our fibreglass to the outside of the driver cell later in the day.

Matt uses a pound shop paint bursh to apply epoxy to the panels before slotting the jig saw together. While he does this I screw the panels together to squeeze out any gaps. Where I can I use proper clamps to pull the car in to shape.

We had a little spare epoxy left over at the end of the session so rather than wasting it we decided to add some strengthening corner pieces. These were quickly cut on the bandsaw and have been placed in areas of the driver cell where we feel a little more strength may be beneficial but won’t get in the way of the driver. These pieces will increase our bond surface area and for the additional weight of them (tiny!) we think they are worth using.

Next up, glassing the outside of the driver cell.

Success, a second blog post about actually building the car, perhaps we are on a roll….

Last night Matt and I got home from work and cracked on prepping our Divinycell panels for gluing. Matt cracked on with a file and some sand paper to neaten some of the edges that we would be bonding to. While he was doing this I got to work building a simple router planer machine.

We needed to be trim the floor pieces down from 20mm thickness to 15mm, and the front panel would need to be 10mm. We have tried in the past using a home made hot wire to cut the foam in large sections however, doing such wide panels results in the wire deflecting a lot inside the foam which means we often end up with a wave like pattern in our foam, not ideal for keeping a nice flat floor! The router planer would remove/destroy more material than the hot wire method, but this wouldn’t matter as we are only removing a small amount.

Hopefully it is obvious how this machine works from the pictures above, it was very simple to make and would be a fantastic DT project in a school. The router sits in a rail which gives us our X axis slide. The Y slide is purely done by manually pushing the panels through the tool. Z axis is set by plunging the router down and setting it’s height stop in order to remove the desired amount of material. The important thing is to not push down on the router too much as it may flex the MDF frame and cause a deeper cut than you want. This unit didn’t take long to make and worked perfectly for our purpose.

Pushing the panel in from one end I then move the router across, removing 1 router bits worth of material per stroke. 10mins and ALOT of foam mess later (have a shop vac and dust mask to hand here!) and the panel is trimmed. A quick pass with some sandpaper and we have a panel that is thinner and ready for bonding to the rest of the chassis.

Now that the panels are the correct size and thickness Matt is able to throw them together as seen here. Next up, gluing the chassis together.

Welcome to the ‘build log’ of Electric 3Galoo, Matt and I have always thought about showing the build process of one of our cars but a rushed build has always taken priority over documenting it. Perhaps that will happen this time as well….who knows, we will try to keep this updated as best we can!

One of the things we really liked about 2Galoo’s design was the use of a fibreglass reinforced foam driver cell. Some of the reasons we like this build method are:

• Strong enough to bolt front and rear subframes to without additional framework
• Immediately meets the requirements for keeping the driver safe in the event of a crash, without the need for bodywork providing any additional crash protection, this leaves our options open regarding a how we make our bodyshell
• Lightweight
• Cheap (the rohacell panels that made up the last car was free from a skip!), we got an incredible deal on the Divinycell we are using for this build
• Easy! It really is, as long as you don’t mind the possible mess of epoxy resin the foam is easy to cut and shape to your desired form. You can then test and add to it as required to ensure things like your batteries fit in and your driver is comfy before then sealing the structure solid with fibreglass.

We are going to approach this build in a slightly different way. Build the entire drive cell first and fibreglass it then add our subframes and other components afterwards. It was very messy in the last build trying to glass in the steel subframes and we think we ended up with an overall weaker structure. This new method should allow us to add extra glass where needed to strengthen the areas where the most stress will be put in to the driver cell (harness mounts, subframes, etc). It will also mean we can unbolt things if we want to move them in later versions of the car. Realistically this design cannot get any smaller and still fit me in it so the rest of the car (bodywork primarily) will have to be built to this footprint.

Enough babble, on with some build!

Matt uses the CAD drawings (yes we have CAD this time, thanks Solid Edge) to draw out the shapes we need on to some Divinycell H80 foam. In an ideal world we would get this cut by a laser/water jet cutter for absolute accuracy, but we don’t have one of those so we are making do! Once we have the shapes drawn I got to work with a jigsaw to cut out the shapes roughly to size. It doesn’t take long and the foam is great to cut by hand as well, perfect for schools I would think.

While I cut the main shape out my girlfriend Jodie gets to work cutting the detailed lines using the bandsaw. I then finish off some of the more awkward cuts that my small bandsaw can’t quite get in to do. Once we were finished we had a pile of car shaped jig saw pieces. Overall this took around 2 hours with a few people working. Next job is to try and reduce the thickness (currently 20mm) or some of the panels, then glue them all together!

We have also really enjoyed reading the build blog found here http://greenpowereasybuild.blogspot.co.uk, some great ideas for construction and a valuable read to any teams thinking of scratch building a car.

Every race we took the previous 2 cars to we had plenty of people crowding round our rear end to get a look at the gearing system we were using on the car, a system that was successful enough to make it between Boogaloo and 2Galoo almost completely unchanged. Due to slightly different wheel sizing offering up a different set of options for our next car it may not make the cut (design stages still in progress here!) but I thought I best take the opportunity to document the setup here for others to use hopefully with the same success.

Note, this post is by no means a plug for gearing your car, let’s not forget that plenty of fixed gear cars are in the top 10 in F24+! Clearly fixed gear cars can be done very very well (Rotary Racer and Jet to name just a few of our competitors).

Another thing to remember is that geared systems add inefficiency in your drivetrain. This is primarily down to the chain having to curve round gears, each link having to rotate slightly causes wear on the chain and generates a small amount of heat. This is pretty much unavoidable in a green power car as you (almost) always have a chain drive, but imagine in a derailleur system how many bends the chain has to go round vs a single gear ratio system! More bends = more chain link rotations required = more heat generation = less efficiency. There is also the issue that the chain is not always running in a perfectly straight line, this not only wears the chain faster but also causes inefficiency. However, as with all good engineering, this is a compromise that the team has to consider, are these inefficiencies in your drivetrain compensated for by the more efficient running of your motor/batteries/vehicle system?

Why do we like having multiple gears on our cars? The main advantage to having driver selectable gearing the car (over having a single gear ratio that the team determines at the start of the session and has to stick with for the whole race) is the flexibility it offers. You use the gears the same way you would in a combustion engine car, to get the torque or RPM output at the wheels you require for the situation that you are in, while keeping the motor/engine in a sweet spot for torque/RPM out.

Now there is plenty of debate regarding race tactics (constant speed vs constant current draw that I am sure Matt will go in to in another post) but we still feel that gearing offers the flexibility to the driver to get the best out of the car in all situations and, gives the driver something else to do to keep them occupied. With our telemetry system feedback (eChook Nano boards coming soon!) we feel that the gearing offers us a chance to learn a lot about the different strategies a team can use, how you then choose the systems ‘sweet spot’ is then up to the team on the day.

The Gearing

When we built Boogaloo and 2Galoo we only had basic tools and no precision engineering equipment to speak of (lathes, pillar drills, milling machines). The result of this was that I wanted to use as many off the shelf bike parts as I could as it meant that we could guarantee that these would be made concentrically and we would end up with a gearing system that would hopefully be as efficient as possible.  (Unfortunately this does also mean that Matt hasn’t done any of his excellent Solid Edge CAD that I can post up in this blog post, but there will be more of that to come later)

Hopefully this blog post will help other teams who are wanting to build something similar to this design.

The good bits about this design:

• Easy to make with basic tools
• Mostly off the shelf bike components, mostly cheap!
• Very small gear ratio increments between gears (in our eyes this is perfect for a GP car), this is mostly thanks to gearing on one side of the lay shaft before then gearing again (maths bit to come!)

The bad bits:

• 2 chains = more loses
• More weight including the dreaded extra rotational mass
• Lay shaft adds frictional loses
• High chain speeds (~4x that seen on a push bike) on the derailleur side chain, though we have not had any issues with chain dropping for the whole 2015 season!

The use of a lay shaft in this system allows all of our gear ratios to end up very close together (see the ‘maths’ bit below), far more so than having put the gearing directly on the wheel end. It also allows us to use a derailleur the ‘correct’ way up while getting the advantages of a freewheeling hub (you can push the car backwards without worrying the chain will fall off!). If you look closely realistically you can see the rear end of a bike has been used, though the pedals/crankset have been replaced with the motor and the spokes/wheel have been replaced with the second chain to send drive to the driven wheel.

Parts of the system and tips on how we built them:

1. The system starts at the motor output shaft. Here we have used a 10T gear that we managed to salvage from an old bicycle cassette I had lying around. It is important to remember that the chain on this side needs to be the correct ‘speed’ to match your cassette, in our case 9sp. By using an old cassette gear I know that it will be around the correct tooth width for 9sp. An old bicycle freehub is cut down to get the neccessary spline fitting to mesh with this gear which allows us to slide the gear off and put a different one on however, I am sure you can get away without this if you expect to change this gear size by simply welding the gear on to a coupler. There are 2 other larger gears shown which were a temporary (still going 1 season later!) measure to act as chain guides should the chain try and jump.  This spline fitting is then fitted to the motor shaft via a steel coupler that we already had in stock, the two parts are welded together. Once I had access to a small lathe this part was later machined in to what you see in these pictures, this was simply to improve our accuracy.
2. A 9speed chain carrys power from the motor to the rear cassette, an off the shelf part that is mounted on to a disc braked mountain bike hub. The hub is the clever bit of this design. It gives the system a great way of achieving an idler shaft without the need to make your own bearing mounts etc. The freehub built in means that the ‘geared’ side of this setup doesn’t rotate when your car is freewheeling.  You can see the slotted piece of steel we used to mount the hub in as well as the conventional bike ‘skewer’ that is still used to clamp the hub in place. This allows you to easily tension the ‘fixed gear’ side of the gearing, putting the correct amount of tension in to the chain on the left side of the car, the derailleur takes up the slack on the right had side chain.
3. Over the other side of the idler shaft the disc brake mount is used to fix a 20tooth ‘single speed’ gear to. These are cheap but are made of very hard steel, very hard to drill! It can be done at home but I had a friend with a decent drill setup drill 6 holes in this gear so it would mount to ths standard disc brake bolt pattern. http://www.chainreactioncycles.com/gusset-1-er-single-speed-sprocket/rp-prod17778
4. A single speed chain carries the power to the driven wheel. Our wheels have disc brake compatible hubs which means we have another opportunity to use this to pass torque in using this simple 6 bolt pattern. We wanted to be able to use normal bicycle chainrings which bolt up to a 5 bolt pattern PCD, this would allow us to change chainrings later if we wanted to change our ratio set.  This part was a little hard for us to sort  but would be easy with the correct access to laser cutting/decent fabrication. We ended up using a spider made from aluminium (link below), this was not the neatest solution as we had to open out the centre diameter to fit our axel as well as drill our own 6bolt pattern on to it. http://www.ebay.co.uk/itm/REDLINE-110-mm-BCD-SPIDER-for-one-peice-cranks-BMX-old-school-/371147285617?hash=item566a1a2471
5. Gears are selected with an indexed shimano shifter (9speed). Currently we preference having ‘thumb’ shifters MTB style but may try some other options in the new season. With this setup I highly recommend getting the best gear cables you can afford (Jagwire seems quite good) as the length of inner/outer cable you need to run is quite long. Keep kinks to a minimum in order to reduce the friction, otherwise you may have trouble selecting gear.

The maths bit

This is a great exercise to go through in schools and excitingly when you do have a gearing system with telemetry (eChook Nano boards coming soon!) you will actually be able to check your calculations using the wheel speed and motor speed sensors to measure a ratio of gearing between the two!

The facts:

• Motor gear: 13tooth
• Lay shaft gear cassette: 14,15, 16, 17, 18, 19, 21, 23, 25
• Lay shaft fixed gear: 20tooth
• Wheel gear: 34tooth

The formula for working out your gears (with a lay shaft) in this setup works like this:

Overall Ratio = (Current Selected Lay shaft Gear / Motor Gear) * (Wheel Gear / Lay shaft Fixed Gear)

For our setup above this gives us a great set of ratios for us in F24+ on our car. The ratios we end up with start at 1.83 (very high wheel speed per motor RPM, a good high speed gear)) and end at 3.27 (very  low wheel speed per motor RPM, a good pullaway and hill climbing gear), the difference between ratios is 0.13 for the single tooth steps on the chain ring and 0.26 for the 2 tooth steps. For our car this set of ratios offers a good range as well as resolution between gears.

We would suggest you have your team make a nice excel document with these calculations in (hopefully using our numbers you will get the same set of results we did!). Although these ratios work well with our car with our wheel sizes in F24+ they will not be optimum for other cars. An excel sheet will allow you to tweak things like the cassette you buy and the other gears in the setup in order to get the best result! It’s an exciting piece of maths to play around with and understand the impact that each gear has on your range (difference in ratio between top and bottom gear) and your resolution (differences between each gear you can select). Every gear in the system plays an important role in the result! You team probably has a good idea of the ratios that your car runs best at already from being on a fixed gear system, make sure these are included in your range of ratios available.

Whilst fixing some bike wheels for the Driven project Rowan, a member of the team, mentioned that he had recently been having some problems with his nearly new (well at least in terms of my car history!) Mazda 3. Specifically he had a drivers seat mounting that had failed causing the whole seat to rock around with over and inch of play.

We decided that this should be a nice quick fix and a good excuse to whip out the welding gear. I have written this up purely becuase the terrible quality of these seats makes me feel that oithers will find theirs fail and may want a quick guide on what to do when this happens!

The seat came out quickly with 4 bolts in the corners, the most time consuming part was actually trying to disconnect the airbag/heated seat connector found underneath the squab (not so hard once you work out how the connector works, just play around a little).

Once we had the seat out we could assess the damage. Rowan had told me already that he had already had to use a jubilee clip to hold the rails on to the pins, this appears to be where the sprung washers have failed and gotten lost. They don’t seem to be easy to find in the correct sizes and a jubilee clip appears to function well enough for now. (Example photos are above of a ‘good’ rod connection to the rail and a slightly more custom one)

The pictures above show where the bracket has failed. It’s difficult to see from but this section connects the main seat base the rails. It is hard to believe this has failed. It is eaily 4mm thick steel and it hasn’t failed around a weld or joint. With a bucket of water on hand and using some leathers to protect the seat cushion I used my Black & Decker Powerfile to clean the metal (excellent tool) and then used the mig to set it back in place. Once the part had cooled we put the seat back together and back in to the car. In true haynes style this process was reverse of removal and just like that the seat was solid again.

For those of you who have been paying attention I have been working on a project called ‘Driven’ for the last 18months. ‘Driven’ is a single seat electric car project where myself and other Jaguar Land Rover graduates design and build a car to race in the Greenpower race series. On this project I had some help from Zoe and Will, graduates in the year below me on the scheme who will be taking over the project when I leave, thanks for the help guys!

The lead acid batteries we use in our electric car are heavy and the team needs an easy way of transporting them to races and storing in the workshop. Currently we use insulated battery boxes to store our batteries in, transport them to the race however, the boxes are too large and have no way of actively heating their contents. Since batteries are allowed in the rules to be heated to 25degC it makes sense for the team to always be running the batteries at this temperature: with lead acid the warmer the battery the more capacity we can get from it (within reason).

Solution: New battery boxes that incorporate a 12V heated seat pads from old car seats. These will allow us to both transport the batteries and warm them prior to a race simply by plugging in a spare car battery. By making the box just big enough there isn’t too much excess air in box that has to be heated up.

For this project I actually constructed a list of requirements before starting the build:

• 2 boxes should be constructed to hold 4 batteries each (4 batteries are used per car for race days and the team has 2 cars).
• Boxes should be easily moveable
• Boxes should be easy to strap in to the van for transport
• Boxes should use a Red SB50 anderson (12V) for power connection to the heated pad, this is the teams standard connector
• Boxes should be made from a wooden frame base with plywood cladding
• Boxes should look presentable, logos on side prefered.
• Lids should be removable and allow charging of batteries inside the box + battery access.

The bottom frame of the box is made from 2×4. This effectively transmits the weight of the batteries to the castors and gives space to insert a piece of insulation foam. A heated seat pad is re-purposed and simply pressed in to place using the piece of foam. Holes are drilled in the bottom of the plywood base for temperature sensors and to give an easy way to push the foam and pad out in case of maintenance. Castors are mounted so that the box can easily be pushed around the teams workshop.

With 2 bases completed Zoe cut up the plywood sides while Will and I glued and screwed them to the base. Extra timber is used inside the box to stiffen up the corners and give the insulation something to press up against. The 50mm insulation is easy to cut with a saw and simply presses in to place, it seems to hold any heat in very effectively.

A quick coat of paint and some securing handles later and we are finished. Another nice quick project that should help the team for many years to come. Now we just need some nice ‘Driven’ branded stickers to add the final touch.

It’s that time of the year again, the annual Leamington raft race, organised by the Rotary Club of RLS. My team and I had so much fun last year that we decided to go for it again, after making a few upgrades to last years set up.

Van-essa, the van roof come boat has been sitting in my garden since last year but all she really needed was a good clean and a little fibreglass to repair some of the larger holes we picked up at the last race. We prefer to think of Van-essa as a boat that sinks constantly…but predictably. The main issue we were having last year was that the boat did not want to go in a straight line. Primarily I think this was due to the ribs running horizontally along the roof, acting like chinings but in the wrong direction. The boat isn’t much longer than it is wide so this seemed to give it a tendency to veer off course at every given opportunity. To fix this I attached 2 pieces of hardboard (not ideal but cheap) to the bottom of the hull to flatten off the underside, then added some 2×4’s to both stiffen the hull slightly and force the water to flow in the correct direction. This was all screwed up through the hull and in to the inner frame, small screw holes would leak slightly but it was a nice quick way to attached the sheet besides, going quicker would mean less time in the water which would mean less water would make it in to the boat?

This solution clearly worked pretty well, 17mins shaved off of our time for last year an a very respectable 3rd place. The boat seemed a lot easier to steer but perhaps that was just me being optimistic, the sped up footage below makes us look like we are playing a game of Pong to get along the river!

The smaller amount of work on the boat meant that I could focus more time on the trailer! Last years design worked fine but was pretty dangerous and a lot of hard work. I wanted to address both of these issues and fortunately a quick call to Action 21 in Leamington soon saw me with a few steel bike frames to play with.

I am calling this a ‘bicycle pick-up truck’. It has full suspension and 2 front V brakes (plenty of stopping power for the speeds we were achieving). Both bikes are welded together and extended (3 and a bit lengths of chain required per side!), with the steering linked with some rudimentary Ackerman set up. A bed made from 2×4’s and OSB was bolted to the back to give me something to rest the raft on. Not only does this solution allow for 2x the power of my previous design, it is also much more stable and comfortable. We certainly got a lot of great looks cycling it down the parade to the race. Both bikes are set up with a single gear ratio and despite having a lot of chain droop we didn’t drop a chain once!

This year we managed to get some video of the race/eventful journey to it! So here is is in full, my first proper attempt at video editing, thanks to Matt for getting the shots on the way down. If you were a big fan of Van-essa the boat I suggest not watching the final 30 seconds, where she meets her end, sorry all I just can’t keep it in the back garden for another year!

Thanks again to the great team who crewed with me and helped prepare the boat for race, we will be back next year with a much faster design to hopefully take home gold.

It has been some time now since I purchased my 1973 French Tandem. This great find on ebay was supposed to be a quick restoration project however, the project quickly got out of hand when I realised the bikes origin. Old French parts are now hard to come by and pretty much the entire bike required some form of TLC. So began a full rebuild/conversion to more modern components.

First problem I found was the wheels. Though I loved the idea of using the original wheels complete with hub brakes (often used on tandem bicycles during long descents to avoid heating the rims and exploding tyres!) these were old steel walled jobbies, no way near as good a braking surface as more modern rims. The hubs also needed a serious service and some spokes needed replacing. On top of all of this they were an odd size wheel which makes buying modern tyres more difficult. £45 picked me up a used but good Rigida Sputnik 700C wheelset, nice strong rims with shimano disc hubs, these fit the frame despite the slight change in diameter.

The new wheels meant I had to change the front fork. This was also convenient as the headset from the original bike was beyond repair, and finding one that would accept the old fork steering tube seemed impossible. A 700C steel fork was found with cantilever brake mounts that would fit a modern threadless headset. Conveniently a modern 1-1/8th threadless headset fits perfectly into this frame, finally a part of this build that seems to be on my side! I found a set of cantilever brakes taken from my touring bike and fitted these as the primary stopping power. A quick bit of welding on the rear of the frame has resulted in a disc brake mount for the rear wheel, I will use this at some point if I find that the rim brakes aren’t giving enough stopping power for 2 riders descending.

The main issue I had with the frame was the bottom brackets. Both of the existing units were totally unusable and finding new units to fit the existing frame threads was proving difficult. Some research uncovered some options such as re-tapping and using an Italian threaded design. Other options were to use special brackets that don’t utilise the threads in the frame at all and instead just fasten in 2 parts to each other (like a nut and bolt but with a bottom bracket inside). All of these options were prohibitively expensive, especially when you consider that the rear bottom bracket on a tandem is known to take a bit of a beating and requires more regular replacement.

To get around my bottom bracket issue I decided to cut the old shell out and weld in a new shell that would accept a threaded English bottom bracket. Perhaps not the neatest solution but certainly cheap at just over £6 for each pre threaded steel shell. The front bottom bracket was fairly easy to accommodate, due to it being an elliptical type set up that is pinch bolted by the frame. This allows the sync chain to be tensioned by rotating the elliptic bottom bracket and bolting the gap closed. Some 5mm sheet steel was cut and welded in place to fashion my own version of what the frame came with, only this time with a nice modern thread.

The rear bottom bracket was slightly more of an issue. Care was taken in cutting out the existing shell and then perfectly lining up the new shell in place. This is a highly stressed area on the frame and I wanted to make a neat job of the welding. Some chooking later and the frame is back in one piece, only this time with a modern thread.

The frame and new fork were then sprayed with some Rustoleum using my HVLP spray gun in a rudimentary booth outside, the finish isn’t fantastic but it is certainly a tough coating, ideal for this build. I built the bike up using new cables and some spare parts I had from various other builds. I am pleased with the end result but I am still not used to riding on the back, the lack of control you feel as a stoker is something I will have to learn to love.

We then tried to ride the tandem solo, from the back. here is a video of my house mate Ben attempting the fairly uncomfortable ride.

Wow it’s been a long time since I posted here apologies, I’ve just been doing a lot of engineering…..which I will now attempt to write up from scraps of paper and numerous logbooks.

After a successful set up running for a few journeys, counting gears and displaying them correctly, Adam has been having some issues with the gear indicator that I made for him. A quick diagnosis showed that some of the power supply debouncing caps had been rattled loose (rally cars are a harsh environment to engineer for!) which was causing some spurious switch presses to be seen by the msp430. This problem quickly manifested itself by counting gears erroneously.

While we were fixing the issue I decided it would be best to ‘re-invent’ the gear indicator system now that we understood Adams requirements a little better. Firstly we decided to go for a much larger 7 segment display, choosing a 25.4mm model that required a 5V forward voltage (2 series LEDs per segment). This is both bigger and brighter which will help Adam see the display in future events. This added some complication as the msp430 on the board runs a 3V3. To get around this I dug around my electronics collection and found a uln2004a (transistor array from TI), this allowed me to switch the 7 segment easily using the msp430 GPIO. As can be seen in the pictures 2 LDOs are now required but the main design hasn’t changed significantly.

The extra ICs (plus the fact we wanted a more robust solution!) has meant that for V2 I have switched to using breadboard. I should have done this in the first place really but hadn’t appreciated how much space I had to play with behind the dash. Note to self: fully understand requirements before inventing a solution. Hopefully this will mean the indicator will be less susceptible to vibration.

I also found some bugs in the code where I was setting up ports for the LED. These have now been fixed and the latest code has been uploaded below. Hopefully this final design will suit Adam’s needs, I will keep you up to date with his progress.

GearIndicator_source_V2

Merry Christmas all. Over the Christmas break I have found some time to put together some of the telemtry and control system I have been designing for the driven electric car. I have written this post primarily for my own benefit as it has taken me a couple of hours to get to a stage of programming my board which would have been avoided had it not been for a simple misunderstanding.

The main ‘brain’ of the board is an STM32F103RB, fantastic bits of kit and my first push away from the AVR family I had become so comfortable with at university. At the time of writing I have designed and soldered up a test PCB and have been plating around with getting a program on to the STM.

I am using my STM32VLDiscovery board’s built in STLink-V1 as a SWD programmer. A simple procedure to wire up between the boards (discovery and my pcb) and get comm’s…AS LONG AS YOU DON’T NEGLECT THE ANALOGUE VDD PIN. I had not forseen the importance of connecting this pin initially and had left it disconnected (for the simple reason I couldn’t find the 0603 inductors I had ordered to nicely smooth the analogue supply line going IN to this pin. After a few hours headscratching, checking and rechecking for shorts, I finally found this disconnected pin to be the source of all my problems. Christmas truely has come early.

So watch out for this in the future Ian, and hopefully this note may help someone else in the future.

During one of our many days off work (must be taken before the end of the year) Adam and I finally got around to installing the Gear Indicator in his rally car. Since Adam put in the request for this project back in March he has crashed and subsequently repaired the entire front end of the car. Not great news for him but every cloud has a silver lining, and the great news for me is that Adam has put some weChook graphics on the newly painted wings.

The final step in getting the indicator working in the vehicle was designing a bracket to hold the microswitches in position around the gear lever. Some experimentation and we settled on a design made from a single aluminium plate that utilised the bolt holes already holding the selector bracket.

Once the microscwithes were mounted on the plate and the wiring tidyied we powered the indicator on (from a spare 12V fused feed) and went for a test drive. The indicator worked perfectly when the engine wasn’t running however, alternator noise on the power lines sometimes led to miscounted gear shifts when we drove the car. Note to self: cars are an incredibly noisey environment so make sure to add smoothing capacitors before you pot any projects. Fortunately with this modification the device seems to work perfectly and Adam hopes to test it out on a rally in the near future.

I have included a couple of videos of us testing the indicator:

It’s been a while since I have posted on the blog. The Driven Telemetry and Control project I am working on is taking up a lot of my time (now I have taken the role as Electrical team lead) and as such I have spent almost no time on the eCumbent or the Tandem.

Fortunately I found some time this evening to finally complete this little project so thought I best update weChook before my time is further absorbed. A friend of mine, Adam, has recently purchased a rally prep’d MG ZR complete with a sequential gearbox. Though the box seems fantastic and allows him to shift quickly through the gears he currently has no way of knowing which gear he is in, since the lever is a simple push pull type.This quick project will hopefully help him with the confusion. A simple up down counter implemented using an MSP430G2211 increments and decrements through the use of 2 micro switches. These will be fixed at the base of the gear lever. The output is handled by a 7 segment LED display (blue, of course) which shows the current gear (Gears: n, 1, 2, 3, 4, 5, 6). Since joining Jaguar Land Rover and discovering the beauty of ‘surprise and delight’ features I have incorporated a gimmick in to this project: it spells out the drivers name at initialization.I haven’t bothered building a PCB for this design as it has so few components. The MSP430 is just soldered to the back of the LED and other components glued around. Adam plans to pot the indicator when we are fully happy with it’s operation anyway.

I have made the following available for this project: Source, Schematic

Like all good engineers I like to have too many projects on at any one time. In the true spirit of this I have picked myself up another one! After this years L2B bike ride a friend and I decided we HAD to do the event next year on a tandem bicycle. The idea has lay dormant for a few weeks until I saw a stunning tandem bicycle come up on ebay for a reasonable price and close to my house. A quick bid later and I am now the proud owner of a Globetrotter Tandem bicycle.

The bike is in need of some work but hey, I wouldn’t enjoy riding it as much unless I had built some bits of it myself. So the project begins.

I have started by completely stripping the bike down. It will be sandblasted then I can give it a quick lick of paint. In doing this I have noticed that I am going to start having some problems dragging this tandem in to the 21st century. Both bottom brackets are in poor condition as is the headset. The worry now is that everything on the bike seems to be strange old French standards, unlike the classic 1″ threaded headsets and cartridge bottom brackets I have become used to.

So this is as far as I have got so far and I am already sensing the project will spiral out of control (instead of being a simple resto). Nevermind! I will keep you updated as things come together.

It has been some time since I have stayed up til 1am working away to get a project done. It seems once you leave University and start working for a big company times like this are rare and I find myself missing the feeling of satisfaction when, at the end of a long session of chooking, you get to go to sleep.

Fortunately this void has recently been filled by Driven…..from my projects page, ‘since starting my graduate scheme for Jaguar Land Rover I have joined the ‘Driven’ engineering project. A team of graduates who design and build a new car every year for an electric vehicle race hosted by Greenpower.’

So far the Driven project has helped me meet many talented people all with the common goal of making an electric vehicle. This has also presented some unique challenges as the team is far bigger than those I have worked with in the past, with the idea behind this being that we function as a microcosm of the overarching company. In addition to building a car every year the team hopes to (and hopefully succeeds in) inspiring the younger engineers at the race events. I plan to use this blog to document some more of my work for the car in the hope that some principles can be applied by other teams to their cars.

Covering the wheels of our cars we can reduce the loses we have due to aerodynamic drag and hopefully this will increase our top speed and efficiency. Although I am no aerodynamics expert I am aware that a spinning wheel made of spokes presents a drag factor when those spokes are moving through the air. My housemate Shawn would be able to describe the effects of this in far more detail than I can but essentially forcing more air to move in and around the car is BAD.

As well as the spokes going through the air creating turbulence we also want to make the side of our car as flat as possible so that the air is passed down the side, rejoining neatly at the back. Since both our Driven cars have open front wheels covering these can further improve (reduce) the drag coefficient of the body. As an added bonus covered wheels can look pretty cool.

This is a quick and easy way to cover wheels we have been experimenting with. It is cheap and can be done easily at home (this was done in my kitchen). So far I haven’t done any quantitative testing on these wheels vs uncovered but as far as I know the aero benefit should outweigh the slight weight penalty.

Tools used:

• Stanley Knife
• Hole Saw + Drill
• String
• Pencil
• Electrical Tape
• Cutting board
• 1 screw / nail
• 1x beautiful assistant, thanks Ben!

We started our build with a big sheet of thermoform plastic, around 1 – 2mm thick should be perfect. Ours is thermoform but only because it is what we could get our hands on, any plastic should work here. We did try some basic thermoforming but found that the plastic would just sag on to the spokes and create a very wavy surface.

Start by drilling a hole where the centre of the wheel will go. We then insert a screw in to this to act as a centre reference. A piece of string and a pencil can then be used as a giant compass to draw a circle of approximately right size to suit your wheel. Cut big to start with, it’s always easier to make the circle smaller than bigger if you go wrong! Once you have your circle drawn you should be able to use a stanley knife to cut around the line. Be careful here.

I apologise for the poor picture quality, it was late and my kitchen isn’t the most photogenic place anyway (though much improved by a photogenic assistant!).

Next up select a hole saw that is a bit bigger than a convenient flange on your hub. Use the hole we drilled earlier as a guide and the hole saw create a bigger hole.

Now the ‘overlap’. Our wheels (like all bike wheels) are slightly convex in that the hub sits out wider than the rim. This means that a flat piece of plastic won’t simply fit on top. To overcome this we must create a very slight cone to fit the wheel. This ‘overlap’ method is far from perfect as it does create a slight bump on the wheel which makes for a not perfectly flat surface, however, it is easy to achieve at home.

Start by cutting from the inside of the cover to the outside. Then place on your wheel. It should be noted that as the outside of the cover is pulled down to the rim a natural overlap can occur at the new cutline, simple. At this point you may have found that you have to adjust the size of your cover. Ideally you want to have around half of the rim showing and half of it covered as this will make life easier in the next step.

Next we run a line of electrical tape all the way around the outer edge of the rim. Covering the new cover and then folding some over the edge of the rim and in to where the tyre sits. Electrical tape actually seems pretty good at going around the circumference of the circle and with practice I can always do it in one long pass. A piece of tape neatly covers that overlap line and then we finish the wheels with some black hot glue just to hold the centre to the hub. It can be worth doing both sides of your wheels, just remember you still have to have access to the valve holes! I just cut a small window on the inside covering to allow access to this.

Recently some friends and I decided to enter the Leamington Spa Rotary Club Raft Race. With a few weeks to go we were still struggling to source some kind of buoyancy we could easily get a hold of for free. Eventually this led to the realisation that the roof from a van would make a perfect boat, all we needed was a roof. A quick call later and we had secured one (nice work Robin and the guys at Truck Busters), and that Friday we went equipped with angle grinders and a large trailer.

After scouring the lot we settled on a lovely white LDV convoy. This had the advantage of the side door not interfering with the roof (one less hole to fill), a high roof line and, most importantly, it was fibreglass so wouldn’t be too heavy for us to move around. An hour or so later and we were done.

Some cutting and re-welding was done to remove as must rust/weight as possible. Shawn and I then cut a piece of 6mm ply to make a stern (where the rear doors went). This was sealed on with silicon sealant and screwed in to the metal frame around the back. Although this mostly held water out during testing we later added some ‘sticks like shi*t’ to fully seal the joint.

A floor was added to spread the load of the 9 occupants of the raft across the fibreglass, then decoration was all that we required to be ready for the race. This lead to the problem of how to move a 4m long and 1.8m wide ‘boat’ 2 miles down to the river. None of us owned a big enough car trailer and it was too big to go on to my roof rack.

Enter the Bike Trailer

In the UK there are plenty or laws about towing by car. However, I could find fairly little relating specifically to towing by bike (note: this does not guarantee this is legal). A giant trailer could double up as a launch for our boat and be towed, albiet slowly, behind my mountain bike. Fortunately I had some spare steel left over from our Ambulance Adventure so I managed to whip together a bike trailer fairly quickly.

This thing is huge. Brakes have been added but with such length on the cable (over 4m) I struggle to achieve any actuation with all the cable stretch and outer flex. As a result the brakes are set so they are permanently on very slightly. Due to the weight of this thing (I estimate around 90kg) the constant rubbing of the brakes just about keeps the trailer in control, believe me I tried without.

This was just a quick project to get the boat to and from the race and we had some great comments about it, despite being in the way a LOT on the road. Despite how quickly it came together it has survived around 12 road miles with the boat on the back, going up and downhill through Leamington town centre. The next iteration will probably have servo actuated brakes and I will get rid of the camber on the front wheel.

We had a lot of fun with the race and ended up coming 4th out of 15 rafts. We also managed to raise some money for Help for Heroes, and we are still collecting on justgiving if you are feeling generous. Hopefully a race video is still to come.

This post is a little late in coming and the eCumbent trike is now a lot further along in development, as usual I find myself forever playing catchup.

The first step was to harvest some donor bikes for parts and build the rear forks. Every time I use my chop saw on steel I realise just how useful it is, this thing cuts through so easily and saves me from hacksaw hell. The forks were jigged up and some axle holding tabs fabricated to hold the wheel in place. These have been made from much thicker (5mm) steel plate: heavy but they should hold down the torque provided by the rear hub motor that I plan to add.

Once the fork came together I started building the spine of the frame. The pictures below show a crude setup using toolboxes of convenient heights to get the frame geometry correct to the plans. It took several attempts to get the spine of the frame true. After using a myriad of distinctly average measurement techniques I eventually settled on a set of tack welds I was happy with and managed to weld the spine together.

I also received the components to build my front wheels, 20″ with 20mm axles. I haven’t laced wheels in a while and the small rims with 3cross lacing proved a challenge but as always it all pulled true in the end. Lacing wheels is a very therapeutic part of building a bike and I would recommend it to anyone who has some level of patience.

More progress to come.