summer project 




bike solar,    








a fantastic    









  • 13TH jULY 2016
  • 14TH jULY 2016
  • 15TH jULY 2016
  • 16TH jULY 2016
  • 17TH jULY 2016
  • 18TH jULY 2016
  • 19TH jULY 2016
  • 20TH jULY 2016



Is it possible to convert my mountain bike into a solar power assisted bike?”



Driven by my love to design machinery, my personal interest in solving this dilemma and my constant need to challenge myself I have come up with the idea to develop what will hopefully be a successful summer project.


After some serious research I decided to buy a bike conversion kit, a solar panel, an MPPT, other bits and bobs, cross my fingers and hope for the best…


Driving is a fantastic thing, being able to get to one place to another in a relatively short question of time without physically exerting oneself is one of the most essential factors we take for granted in our lives. But, what happens when you don´t have a car but still want to get to places without having to depend on public transport? Wouldn’t it be great to be able to do that without having a license and with a budget of 700 pounds? Answer: Yes! Have you thought about an electric bike?

Okay, but that means I have to charge it…

Answer: No need! Make it solar!





“Is it possible to convert my mountain bike into a solar power assisted bike?”


The first challenge to the project was to convert my old mountain bike into an electric bike, using the bike conversion kit I had purchased:  The kit cam with pretty clear instructions, so I set up a space in the garage and got to work.  Things went pretty smoothly, despite needing to find alternative solutions when my bike turned out to be not-so standard.


The first three steps were pretty straightforward.  It was just a question of knowing how to relate the instructions to the reality in front of me. As an example, “remove handlebar grips” may be easy for a bike mechanic, but for me, it begs the question “how?”.  There was a slight hiccup when the battery did not fit into the designed position, solved with a power drill, file and sweat, but overall things went well. 

Mounting the magnetic pedal-monitor in step 4 was more challenging.  First, removing the pedals was not trivial – both due to 6 years of neglect of the bicycle and also not realising that screw mechanism works backwards – second, my non-standard frame meant that there was nowhere to actually mount the magnet.  Cue glue-gun and a slightly “home-made” mounting.  


Finally, step 5 was relatively painless – though, I had to improvise again as, given that my bike is a girl’s bike, the space where the cable connector was supposed to fit was much smaller than expected – something maybe for the kit providers to take into account!


The conversion consisted in five major steps:

(1) Replacing the back wheel with the new one integrating the 250W electric motor

(2) Mounting the battery

(3) Swapping out the existing brakes for brakes with built-in electric cut-off

(4) Mounting the thumb-throttle and the magnetic pedal-monitoring system

(5) Adding the LED and wiring everything together.


Finally, the moment of truth – the first test ride. Again, despite a minor scare when the brakes did not work (thank goodness for quiet roads and thick-soled shoes) and the need to adjust the pedal sensor everything went well and I was amazed at how liberating it was effectively riding a moped with absolutely no engine sound!  


After more test rides than were strictly necessary, I started to experiment with charging the battery from the solar panel.  In this first stage, the solar panel was not mounted on the bike.  Instead, I simply connected the panel to the MPPT, and then calibrated the MPPT before connecting the MPPT to the battery.  The job of the MPPT is to convert the voltage from the solar panel into a stable output, while also acting to keep the current below certain levels if needed.  My panels produced a target 12V, but I needed a higher output voltage to charge my 36V battery.  I experimented with a number of different values as I needed to be careful not to overload the battery as my MPPT (cheap from China through e-bay) does not have a safety cut-out.  In the end, I settled on 40V, which is lower than the charger that came with the bike (42 V) but still seemed capable of fully charging the battery.

After charging the battery with solar power, I took the bike on a long test ride, designed to drain the battery completely to see if I got the same mileage as with a mains charge.  However, after about an hour, disaster struck.  After a long downhill, followed by a speed-bump, the motor suddenly stopped working.  The battery still had plenty of charge, but the opening the throttle produced nothing more than an occasional grinding sound.  I could not see anything obviously wrong and I was scared I had burned the motor out by somehow overloading it and pedalled the bike home, cursing the manufacturer for dodgy parts and scared that something I had done had caused the problem.  When I finally got  home I decided to leave it until the morning before contacting the manufacturer to see what was covered on the warranty.  Lucky I did: “new day, new eyes” and as I looked around the circuit in the morning, I noticed that a small wire had become disconnected in the connector box.  A simple reconnect and I was back in business – though I still have no idea what the grinding noise had been.  Next step, attach the solar panels to the bike to make it truly solar!

While there are a number of alternative ways to fit solar panels to bikes, I wanted to do something completely original that would really stand out.   As I planned my project, I found myself sketching a bubble-like canopy over the bike with the solar panel at the top, providing shade for the rider as well as power for the bike.  In retrospect, the design looks quite like the BMW C1 motorcycle (though the design experts at BMW would surely disagree).  Finally, I wanted the canopy to be detachable.  This would allow me to much more easily store the bike, put it in a car if necessary, or use it without carrying the solar panels with me if I was only planning a short ride.    

I had thought through roughly how to achieve the bubble, using flexible tent-poles, secured at each end of the bike.  These should be light, flexible enough to make the dome, and strong enough to hold the solar panel.  So far, so good - however, the big challenge was how to fit the dome to the bike. 


I could not fit the dome directly to the bike, as the the canopy needed to be 60cm wide to support the solar panel, so I would have to add some horizontal struts.  But these struts would need to be wide apart to make a dome that would be high enough and could not interfere with the handlebars, pedals , wheels or any of the electrics that I had just fitted to the bike. 


My solution was to fit two long poles running along the middle of the bike at the height of a cross-bar, reaching from the tip of the front wheel to the back of the back wheel. I would then attach the horizontal struts at each end (making a big ][ )and attach the tent-poles through holes drilled in the struts.


The new poles in my structure needed to be thin, so as to not get in the way of rider, low weight to avoid stressing the motor further, and strong enough to hold the canopy in place.  I did not want to use steel given its weight, cost and challenges cutting/drilling it in my “workshop”, so I decided to use aluminium. In the local DIY shop, I bought a 2m U-shaped 2m pole to be cut into lengths of 60cm for the horizontal struts and two 2m L-shaped poles (cut to 1.75m) to run the length of the bike.  The L-shaped poles seemed to provide a good way of keeping weight down as well as simplifying fitting, while providing plenty of strength.


Putting the structure together was pretty simple.  I was able to sandwich the long poles to the bike and could take advantage of two pre-existing screw-holes in the bike frame (for a rack, I think) to fasten the structure securely.  Adding the tent-poles was a little tricky, and needed a second pair of hands, but then adding the solar panels was straightforward. 


Oh, oh...

What now?!



Looking at it, it was obvious what the problem was: the lightweight L-shaped aluminium poles may have been strong enough to easily hold the weight of the panel, but they were twisting due to torsion whenever the panel was not completely vertical.  My design was flawed.


Over the next couple of hours, I thought through several different ideas to fix the problem.  These included minor fixes, such as strengthening the poles or adding additional diagonal struts from the bike to the cross-struts, though to more radical solutions such as abandoning the dome and constructing a rigid, rectangular, canopy instead or finally, moving to a trailer-based solution for the solar panel.  Throughout the process, I was conscious that I was running out of time and low on money to finish the bike. 


In the end, I decided to take a risk and try to fix the existing plan by upgrading the long rods for ones with better torsional strength.  Back in the DIY shop, I looked at a number of different options, surreptitiously testing their “twist” and finally decided to stick with aluminium, but move to square-based poles.  While this doubled the amount of material, the increase in weight was not huge and the effective volume was the same as my original design.  I also decided to buy some see-through Perspex screen to add to the front and back of the canopy to increase the rigidity of the structure. 


Back in the garage, I replaced the L-shaped poles and fixed the Perspex into place, using a mixture of pop-rivets with the strut and wiring to the tent poles. The result was a success.  While the canopy structure does take some getting used to when riding the bike, it was now completely rigid and did not sway from side-to-side.  I had a workable solution!


In the final steps of the construction, I mounted the MPPT behind the saddle on the long poles and feed the electric wires from the solar panel, down the tent-poles to the MPPT.  I then attached the MPPT to the battery, continuously charging the battery whenever there is sun, regardless of whether the bike is being ridden or not.


I then conducted some test rides to check the stability of the bike and its ability to charge itself up. Up until now, I can honestly say the bike has exceeded all my expectations with the biggest problem being other drivers taking their eyes off the road to stare at me as I glide along in my silent bubble. 





However, at this point, it was clear I had a problem.  While the bike was fine when perfectly vertical, any movement or vibration would cause the canopy to wobble from side-to-side in an alarming manner.  There was no way I could ride this.

In my opinion, the bike conversion has been a very ambitious project which has worked very well. So, to answer my title: “Is it possible to convert my mountain bike into a solar power assisted bike?” I can happily answer: Yes.


I am extremely happy with its outcome and for EPQ prospects I can gladly say I have reached my objective and its end. 


However, for my own interests I have not finished developing it and plan to upgrade it in the close future when I have more time by adding a second solar panel (easy enough to do, I just need to shift the first solar panel down the poles and fitting the second one behind and do some rearranging with the wires to join them in one single circuit). I also plan to add LED lights and am considering the possibility of installing a music system, though I need to assess how much power drainage from the motor these “luxuries” would entail.  More prosaically, I also intend to add a couple of baskets on the long poles over the wheels, so that I can carry things without needing to wear a backpack.








13 July 2016


Test the solar panel by using an ammeter and voltmeter

Find an initial overview of the effects of solar panel inclination Check functioning of MPPT

Start installing electrical components to bike

Test solar panel

Connect MPPT to solar panel

Remove back wheel

Find most efficient incline for solar panels, first values

14 July 2016


Continue installation of electrical components to bike


Create a clear analysis on the effect of solar panel incline and the effect on power output efficiency



Replace back wheel with integrated 250W motor wheel

Mount battery


Exchange brakes for brakes with built-in electric cut-off and mount throttle

Find most efficient incline for solar panels, complete analysis


15 July 2016


finish converting the bike into an electrical one and test it

Fit pedal assisting and magnetic sensors

Install multifunction LCD meter

Tidy-up wires

Connecting wires to LCD controller

Test ride

Formatting MPPT

16 July 2016


Connect the bike’s battery to the solar panel and charging it with it

Drain the battery

Program MPPT

Connect MPPT to solar panel and battery and leave to charge

17 July 2016


Compare mileage covered and average power to drain battery when solar charged or charged by mains


Take the bike out again to drain the battery (charged from solar panels) and record the mileage covered and average power for this to happen

Charge the battery using mains electricity

Take the bike out again to drain the battery and record the mileage covered and average power for this to happen

18-22 July 2016


Build the structure of the canopy holding up the solar panel


Make adjustments judging on test rides until I have obtained acceptable result




In addition to being environmentally friendly, electric bicycles are an extremely economic form of transport, particularly when compared to cars or motorcycles. 

Given how cheap electric bikes are to run and the relatively high price of solar panels, it is not clear that a solar bike makes sense economically.  In this section we look at the economics of both forms of bicycle, discuss what payback period you would need for the solar option to be a cheaper alternative and then also discuss why, despite the unfavourable economics, there may still be a practical case to make for solar bikes, beyond environmental considerations.


The following table shows the costs first of simply converting the bicycle to an e-bike and then converting the e-bike to a solar bike.




e-Bike conversion







Solar panel









Screws and additional material



Tent poles



Perspex sheet






As can be seen, the e-bike conversion costs £350, while the solar conversion increases the cost to a total of £608 – ie an additional cost of £258.


The motor on the bike is rated at 250W and the battery holds a charge of 13Ah at 36V. In theory, at least, the battery is able to run the motor at full power for close to 2 hours (13*36/250 = 1.87 hours, or 112 minutes). 

Charging the battery from the mains costs around 10p, so each minute of ride costs £0.00089. 

If instead of using the mains charger, we use solar energy, we would save this cost.  However, it would take a total of 289 785 minutes of ride to recover our investment of £258 (258/0.00089) even without assuming a discount rate.  That is over 200 days of constant usage of the bike, or if we assume 2 hours a day, is 6.6 years. 


While 6.6 years is a long payback period, in reality, the actual payback period is likely to be much longer.  First, you should expect more wear-and-tear on the bike from the canopy and second, you are unlikely to get 365 days of sunshine per year!






Economic and business study


Against this, though we should take into account that the components for the solar e-bike were separately bought based on my own, unoptimized design.  A quick comparison of the prices of the elements in the e-bike conversion kit, versus the kit, suggests that uying peices individually adds just over 30% to the cost of items.  This is before and bulk discount or steps to optimise the design so it is reasonable to assume that it might be possible to reduce the investment required from £258 to something closer to £150, which would shorten the payback period substantially. 

Similarly, we should expect the cost of solar panels to continue to decrease over the next few years, while the cost of mains electricity might be expected to continue to rise, which further improves the relative economics of the solar bike.


Despite the above, it is clearly hard to justify the solar bike from an economic perspective.  However, the solar bike does provide a significant advantage in terms of the range that you can travel before you need to stop and recharge the battery. 


The bike’s motor consumes 250W at full power, while the solar panel produces 100W.  This means that running at full power, the solar panel increases the bikes range by 67% from 112 minutes up to 187 minutes. 


Of course, you are unlikely to run the bike at full power for any period of time. At the same time, the solar panels rarely give the full 100W in practice.  If we assume that the bike runs at half-power and the panels work at 75% efficiency, adding the solar panel increases our range from 224 up to 562 minutes – an increase in range of 150%. Over nine hours of autonomy is likely to be sufficient for most uses, especially as this is based on constant travel. Every time you stop for a break, the bike continues to charge, increasing the range further. 

This increase in range is especially relevant for:

-          Countries where mains electricity is not readily available

-          Bike touring/camping holidays, where you may be away from a mains power socket for a number of days.   


In summary, while the economics of the solar bike are not compelling, the increase in range that the solar panels allows means that there are a number of applications where the technology could be a reasonable solution.


13 July 2016

Today’s plan is to

-          Test the solar panel by using an ammeter and voltmeter

-          Find an initial overview of the effects of solar panel inclination Check functioning of MPPT

-          Start installing electrical components to bike


  • Testing solar panel

Checking functionality of equipment in case of need to purchase new equipment, I am doing this first due to time pressure, in the case I would need to buy new parts it’s better to find out at an early stage to make sure I get it in time.

  • Connected multimeter to solar panel, obtained voltage and current- it works


  • Connecting MPPT to solar panel
    • Striping solar panel wires by taking off original connectors (maybe standard in China, but not in Europe)
    • Manually reshaped wire to adapt into MPPT Solar Charged Controller

Wire not properly connected- it came out, so I then took a different approach and crimped connector to the wires, correct connection I then had to take MPPT apart as screw at an awkward position and easier to put in by taking structure to bits.


  • Removing back wheel
    • Take off back wheel
    • Take out inner tube and check it is not punctured splinters by putting in water and applying pressure to it
    • Check for splinters inside tyre by passing hands through

Easy task which the result of the knowledge that there are no holes or splinters.



  • Finding most efficient incline for solar panels, first values:


By inclining solar panels with respect to the ground and placing them against the wall I was able to identify the highest power output angle inclination.


  • Place top of solar panel against wall
  • Vary distance from bottom of panel and the floor, creating an angle with the horizontal
  • Take readings of height of panel as you vary distance from wall
  • Take readings for voltage (should remain constant) with voltmeter for each variation of position
  • Take readings for current (should vary) with ammeter for each variation of position











Dear Dairy...



My value for power, in watts was found by finding the values for current (amps) and voltage (volts) with a standard multimeter at a “constant” light intensity source (the Sun) at a more or less constant incident angle (readings were taken from 11.03 to 11.38 AM). Temperature was maintained constant throughout (19°C according to The Weather Channel app).

From my results I have concluded that greatest efficiency for solar panels is between the range of incline 0° and 50°. I will continue narrowing this gap by taking further readings at hour intervals and with known angles.

Today’s graph: (for detailed individual readings see excel doc page 1)


I am aware of the fact that this is only a draft of what will be actually found tomorrow. In tomorrow’s measurements I will include lux., take several readings at constant time intervals, have a constant point of reference and equally spaced controlled variable spacings. Today’s mistakes were not having a fixed point of reference to compare to, only having one set of values, I cannot compare it to something else and draw conclusions from it. Today’s experience has helped me develop a routine and get the first steps of the project into action.


14 July 2016

Today’s plan is to

- Continue installation of electrical components to bike

- Create a clear analysis on the effect of solar panel incline and the effect on power output efficiency


  • Replacing back wheel with integrated 250W motor wheel
    • Attach original tyre and inner tube to new wheel with the help of 2 teaspoons
    • Pump up inner tube
    • Attach the new wheelonto bike including lug nut
    • Place inner tube inside tyre

This stage was quite a straightforward stage and did not cause any complications – though lug nut was tricky as there was no hole on the bike to fit screw in and lug nut was very rigid requiring the use of an extension.

  • Mount battery
    • Remove water bottle holder
    • Mount battery holder

Failed as the bike mounting was not consistent with the battery holder.  Need to rethink battery mounting.

  • Exchange brakes for brakes with built-in electric cut-off and mount throttle

This was quite a challenging stage as handlebar grips were not easy to slide off and required the extra stage of adding alcohol to work as a lubricant. Once they were removed there was a problem when trying to attach the left new brake due to poor instructions, which lead to the need to dismantle of the structure to reattach brake in the correct position.

  • Finding most efficient incline for solar panels, complete analysis:
    • Find values for current and voltage at measured angles of incline with respect to the horizontal
    • Angles are found by inclining panel by placing it at known heights and distances form the wall, using arctan function.
    • Find angle the sun makes with the solar panel by placing a pole of a known length into the ground and measuring the length of the shadow it makes.






From these values we know the adjacent and the opposite to a right sided-angle, so we can find the angle by using tan.

  • By comparing both sets of data we can identify the angle at which the most efficient voltage is acquired.
  • Measurements are taken every hour, so the angle with the Sun and the output power will vary each time.
  • Take readings of lux (light intensity) with the iOS Lux Camera app
  • Take measurements of initial length of shadow and final length of shadow as it will change as time passes whilst you are taking readings. Assume the rate at which the shadow changes is constant (even though it will make our uncertainties greater) so that you can therefore take the mean value of the two.
  • Find value for power (from relationship P=IV)
  • Find value for angle (from relationship tan x = opposite/adjacent
  • Assume the rays of light hit the middle of the panel


This activity today was quite tedious and stressful, as I had to stop doing any activity I was engaged in at the time to take the readings every hour. The first readings took longer to find, but as I took more readings the activity became more automatic and faster. There were no apparent problems except for the fact that the cables came out from the MPPT and it had to be rewired and that I had to move the setups place because of the Sun´s movement and the shade formed by the wall against which I was leaning on.

From the data collected I have been able to produce tables on excel to find the average most efficient incline with respect to the angle made by the Sun. Details can be found on the excel spreadsheet but to summarise, the solar panel must be perpendicular to the rays of the Sun (my found value was 85°, which I will round up to 90º), this is hardly surprising, what is interesting however, is to see he angles at which the efficiency works best at certain hours.

As an evaluation of my analysis I would say I should have done the measurements for a greater range of angles, instead of restricting the domain due to the data collected yesterday. If I were to do it again, I would also collect data from 9 AM to 9PM, as the angle variation would be greater. If I were to do this I would also have to take into account the time at which the Sun sets and maybe rethink my time intervals. 

15 July 2016

Today’s plan is to finish converting the bike into an electrical one and test it

  • Mounting battery
    • Resolve positioning issues

After yesterday´s fail, today I have spent a fair bit of time trying to find an alternative position for battery, as the original place did not have compatible holes between the battery base and the water bottle holes.

I decided to try out an alternative method. My options were:

  • Attach a rack on the back wheel, and position battery on it
  • Finding alternative way of attaching battery on original position
  • Placing battery on pole directly below seat by drilling holes for screws (not an ideal option as I do not know the strength of the material and the effect drilling holes will have on it)


After research online I purchased a suitable rack to place battery on back wheel, and I returned to work. I then rethought the design and cancelled my purchase. I then remodelled the base of the battery by drilling extra holes (this resulted in being quite tricky as the water bottle holes on bike were quite small and didn´t leave much margin for error). I successfully attached the battery on the original plan´s position.


  • Fitting pedal assisting and magnetic sensors
    • Remove pedal bun
    • Remove pedal crank arm crank
    • Remove the cover outer ring and fit electric sensor, then replace the outer ring.
    • Fit magnetic sensor
    • Replace pedal

Removing the crank required a lot of patience and strength, as it would not shift from position.

Also, my bike is not standard, so I could not remove the cover outer ring. I resolved this by attaching it directly onto the cover with silicone using a glue gun (as super glue was not working properly)

Fitting the magnetic sensor and the pedal back on was quite straightforward.


  • Installing multifunction LCD meter
    • Attach in between both handlebars
  • Tidying-up wires
    • Join all 4 loose wires (2 brake wires, wire from display and thumb throttle) by binding them using Zip Ties
    • Join battery wire in later on, as it is further down the bike
    • Secure bound wire onto bike frame

This stage was not supposed to be complicated, but poor instructions resulted in me leaving out a wire out of the 4 when binding them together, having to unbind them to join extra wire, and rebinding it. I then left out the battery wire and had to unbind the cables for a second time, attach second added wire and bind them all back on. This was an extremely tedious and unnecessary stage on the project which showed poor performance by my part. Looking on the bright side though, I am now an expert on manually binding wires with Zip Ties.


  • Connecting wires to LCD controller
    • Open a small slot on the kit bag and feed all the wires through
    • Match the cables up to the colours and clip them together

I joined up the cables before feeding them through the bag, which meant I had to undo them, to avoid repeats of this. 


  • Test ride!

Success! The bike engine works and travels and is much more efficient than I had anticipated going at a reasonably fast speed.

The brakes need adjusting, as their effect is not as immediate as would be expected, though the break cut-off does work.

Also experiencing problems with the pedal assistance system as the motor will not activate when you start pedalling and requires the use of the thumb throttle to start it.

Minor problems which will be adjusted tomorrow.


  • Formatting MPPT


-          Follow algorithm and input values



Voltage, V

Current, A








 Not an easy task. The instructions used extremely technical terms which made the algorithm very hard to follow and program the MPPT. It gave no information on how to store programming functions, which means the input values must be repeated every time I wish to charge the battery with the solar panel.


I have set the output voltage to be 38V even though the battery requires 36 because in order for there to be a potential difference and charge to flow from the MPPT to the battery the voltage must be greater on the MPPT than the battery. The adaptor allows up to 42V, however, as the solar circuit does not have a cut-off system I have decided to play safe at 38. This means the battery is unlikely to ever charge to full with the solar panels only. However, this is obviously better them causing long-term damage to the battery because of overcharge.




  •  16 July 2016

    Today’s plan is to connect the bike’s battery to the solar panel and charging it with it

    -          Draining the battery

    -          The LCD screen displays the amount (in bars) left of battery

    -          In order to check the battery is being charged by the MPP we must be able to see a difference in the charge the battery has.

    So ride bike until battery reaches low levels (1 bar)

    -          Disconnect battery from bike and connect to the programmed MPPT

    -          Program MPPT and leave connected for several hours

    -          Place battery on bike and check change in battery levels

    Draining the battery actually took longer than expected, it required 10.4 miles of alternating full power (going up steep hills) and gentle power (going down) with an average speed of 24km/h before reaching the 1 bar.

    As it got quite late I became aware the need to attach LED´s to the bike in order to be seen by traffic at night. The bike will not be recharging itself with solar energy when there is no sunlight, but it will continue to function as long as it has charge remaining on the battery.

    The battery did not seem to be charging on the programmed settings even though the display did show a charge output from MPPT to battery. When the MPPT finally did show an output of 0 amps I switched the voltage output to 40V, at which point the battery showed a significant increase in charge over a smaller period of time.

    I will adjust MPPT settings when charging battery to:


    Voltage, V

    Current, A







    This means that the chances for overcharge have increased, which is a negative. On the plus side though, the battery is actually charging, and the greater potential difference means the rate of charging will be faster as there is a greater potential difference.

 17 July 2016

Today’s plan is to:


  • Take the bike out again to drain the battery (charged from solar panels) and record the mileage covered and average power for this to happen
  • Charge the battery using mains electricity
  • Take the bike out again to drain the battery and record the mileage covered and average power for this to happen


Could not achieve any of the two experiments, as back wheel got punctured at early stages of ride due to glass from several broken bottles left in the middle of the road. I had to return without recording results. Instead I took the back wheel off, removed inner tube (quite tricky), placed it in water and immediately found the hole. As the hole was too big to repair it with a puncture repair kit I simply replaced it with a new inner tube.

I then checked for class in the tyre, as I found none I placed the inner tube in the tyre and put it on the bike once again.


18 July 2016

Today´s plan is to start building the structure of the canopy holding up the solar panel

There was no defined plan to this stage and I sort of worked on the go trying to adapt the structure

  • Plan 1- (using materials I had at home)

-          Manually sawed circular metal poles (2x7cm 2x27cm lengths)

-          Covered bike pole in tape to thicken its structure

-          Attached small metal poles to front of bike using silver tape

Problems: only had 1 long pole (old swimming poll cleaning appliance part and front small poles wobbled, not remaining fixed in place. Discard poles, unmount structure, and look for alternative method.


  • Plan 2- (going to Leroy Merlin and buying extra pieces)

-          Manually sawed U-shaped aluminium poles (2x60cm lengths)

-          Manually sawed L-shaped aluminium poles (2x170cm lengths)

-          Manually sawed circular metal pole (6cm length)

-          Manually sanded down aluminium pole slimming it down for it to fit into the holder

-          Manually sanded circular metal pole to fit into gap

-          Drilled central and lateral holes into aluminium poles

-          Mounted structure and fitted screw through

-          Joined 8 tent poles on each 2 sides to make canopy to support solar panel and fed string through to give more support

-          Fed string through tent poles (very fiddly, strings came out of poles towards the end in several occasions)

-          Mounted joined tent poles onto built structure and crossed over at the top

Problems: canopy very wobbly because of wobbly tent poles on base. Crossed pole structure unstable.

  • Plan 3

-          Removed 1 pole from each side

-          Uncrossed poles for canopy

-          Cut out and sanded wooden blocks fitting them into canopy bases (1.6x1.8cm)

-          Drilled holes in wooden blocks for poles to be pushed in

Problems: canopy still very unstable due to torsion of L-poles alongside bike.

Tonight we took the bike out for a ride again. After making some concerning sounds the motor on the wheel shut down. We had to push it back home and have decided to leave it until tomorrow morning. Hopefully it won´t be a major problem.


19 July 2016

This morning I went through the controller box where all the wires are connected. To my relief one of the wires had come undone, so I reattached it and turn the battery on. It worked. This was very good news, as I had already taken out the warranty papers and was checking out the shipping process for returns, and had gone through the terms and conditions with the company policy. I would have had to unmount all the pieces, return them and remount the whole bike. This would mean I´d be losing a minimum of a month in the process of returns and receiving the new pieces, making the project unlikely to be finished by the end of the summer. However, the odds were on my side and the bike is fully functioning again. What is more, I have been able to partially correct the assisted pedalling system, which occasionally does not respond, but now works much better.

Today´s plan after seeing we didn´t have to make any returns is to

-          Find a solution to the unstable canopy (requires planning and trial-error)

-          Fix MPPT onto bike (requires planning and trial-error)

-          Replace screws on unstable LCD screen (easily done)


  • Plan 4

-          Reduce torsion of L-poles by placing one end of the left-over length of L-poles onto stationary point on wheels and the other end onto the U-shaped pole resting on L-shaped poles

Problems   : difficult to attach onto required position on wheel. No obvious way to dismantle canopy easily in this case. Poles not long enough.


  • Plan 5

-          Reduce torsion of L-poles by placing another square shaped pole under each L-shaped pole

-          Cut 3x10cm pieces of wood to attach to sides of L-poles to fix MPPT box on top

Problems: we only have 1 pole that fits required characteristics


  • Plan 6- Back to Leroy Merlin

-          Purchased 2m long square aluminium poles

-          Purchased 125x100cm sheet of Perspex

-          Purchased standard screws, butterfly wing screws, bolts and lug nuts

-          Cut square aluminium poles to 170cm

-          Cut Perspex sheet into 62.5x100cm shapes

-          Drilled holes all holes on L-poles on square poles, as they will be taking their place

-          Drilled holes on Perspex and u-poles and fixed sheet onto metal by using a rivet gun

20 July 2016

Today´s plan is to continue with yesterday´s work

  • Plan 6 continued

-          Placed wooden blocks in front to prevent wobbling of square poles.

-          Rivet Perspex to lateral supports u-shaped aluminium poles

-          Fixed poles onto Perspex by drilling holes between poles and feeding fence wire through them

-          Mounted Perspex and arches onto bike fixed with 4 retaining bolts

-          Added solar panel to arches

-          Mounted MPPT onto structure by adding 2 L-shaped poles riveting them on and bolting on MPPT

-          Redid wiring on solar panel to extend wire length in order for it to be able to connect to the MPPT fixed in place

-          Attached MPPT to battery


  • Test ride!

Test conditions were not ideal

-          Cloudy conditions

-          Extremely windy

Test ride completed successfully over 1km circuit by 2 different riders. Results:

Solar panels function as expected.

Arch and particularly Perspex made bike significantly less stable in the windy conditions. On sharp bends structure experienced some torsion but remained within acceptable limits

Evaluation of test: No obvious modifications required