Tuesday 19 June 2012

No Camber Tests

From the Sunday Times 17th June 2012

We had a nice bit of publicity from the World Championships at Fowlmead in the Sunday times last weekend, although not totally complimentary about Quattro. “a four-wheeled recumbent of such comical appearance that you’d think it belonged to a low-budget 1950 ski-fi film". But there is no such thing as bad publicity and a two page spread in a national newspaper can only help our cause.
Kingpin Bent to Eliminate Camber (Do not try his at home)

New Fatter UJ Ring

UJ with Eight Bearings Fitted
I made some new double depth UJ rings that can now take eight bearings rather than four and a new slightly shorter and hollow drive shaft. I was still running the same 19mm diameter solid steel one that was made for me half way across America.
So Steve and I set off to Reading Velodrome again to test the new setup.

No Camber and 145mm Ground Clearance
Interestingly I have ended up with about 145mm ground clearance, any less, and the wheels would rub on the fairing.
The initial conclusion from the testing was that there was little change; I was able to maintain 30mph (48kph) reasonably comfortably with a number of laps at 32mph (51kph). We did try taping up the nose vent and that did seem to help the speed a little. I left Reading feeling disappointed that there wasn’t a dramatic increase in speed but when we got back to work, I noticed that yet again, one of the tyres was completely flat and just like at Fowlmead, it was caused by a piece of metal and I have no idea what effect it had on the testing. This means that a return trip is needed to repeat the tests.

Lee Wakefield sent me a link to an interesting looking device that wirelessly monitors tyre pressures; it is designed for trucks but would work just as well on Quattro.

Tyre Presure Monotor
The trouble is that at the moment, our tests on Quattro are not very scientific; it may be possible to see a large improvement but any small changes in performance could go un-noticed. I do have a very old pair of SRM Cranks that are in Germany, being updated and calibrated. I am not sure they will be of much use for my Quattro because of the very narrow custom chainset. Graham Sparey-Taylor’s Quattro may be easier to adapt as he has a more standard bottom bracket arrangement and he has also made provision for an electric motor, which could also be used to give a constant power input.
STOP PRESS
Following his recent success, newly crowned World Champion Steve Slade has been offered a lucrative sponsorship deal with Raleigh Cycles and is seen here modelling his new team strip.

Friday 15 June 2012

Camber

Laser Alignment of Wheels Showing Camber
Having noticed the grease exiting my UJ bearings after the Worlds, I dismantled the front axle and found that half of the bearings were already quite rough. This is after only about 150 miles at the most. Dr Graham suggested that this could be caused by cornering forces. I probably did more, high load cornering over that weekend than the whole crossing of the USA during ROAM.
I ordered and now have received some sealed rather than shielded replacements but I am not sure that the problem is only caused by the loss of grease, I am sure they are overloaded as well. The bearings I am using at the moment are 696ZZ which are 6x15x5mm and have a static load rating of 655N. Unfortunately, I can’t increase the bearing size easily without a re-design, so I am stuck with them for the time being. The only interim solution could be to double up the same bearings within a modified housing.
UJ Assembly
While Quattro was in bits yet again, I decided on some more drastic action. For testing purposes, by adjusting (bending) my kingpins and modifying the lower wishbones, I am going to eliminate the camber on all four wheels. This will result in a narrower track of 600mm rather than 650mm but with the four wheels and low centre of gravity, Quattro will still be very stable.
From the excellent “The World’s Most Fuel Efficient Vehicle” (PACII) book.
The original idea for cambered wheels was to give maximum stability within the narrowest body. The 650mm track was chosen as it is one of the regulations required for pedal car racing in the UK. I was also influenced by the authors of “The World’s Most Fuel Efficient Vehicle” (PACII) book. In the section on tyres, they show a graph of rolling resistance versus camber angles. The curve is more or less flat between 4deg and 8deg, so I decided to go for 8deg. What I should have noticed was that this flat section of the curve indicated a 15-20% increase in rolling resistance. In fact, when I measured the camber on Quattro, it was closer to 9deg.
The other advantage of eliminating the camber will be that the UJs are running concentrically when Quattro is going straight, which should reduce friction losses considerably.
Camber Eliminated
Thinking positive, this could save me up to 20W in my tyre rolling resistance power at 35mph as well as reducing the side loads and therefore the friction in the wheel bearings. The power saving in the UJs is difficult to quantify but the angular speed variation I mentioned in an earlier blog should be minimised.
A couple of people mentioned the lack of shelter behind Quattro compared with other vehicles while racing at the weekend. Mike Burrows tried drafting me during the two hour event and said that there was no noticeable shelter. This is hopefully a sign of good aerodynamics and leads me to think and hope that any big gains in speed may well come from improved mechanical efficiency.
I should be able to get the modifications done over the weekend, so off to Reading for a testing session on the velodrome which is free Tuesday morning next week.



Monday 11 June 2012

World Championships Fowlmead

Firstly, a big thanks to all the organisers, helpers, marshals, timekeepers, first aiders etc who made this a very smooth running and thoroughly enjoyable weekend.

Steve and I arrived at a very windy Fowlmead circuit early afternoon on Friday 8th June. With assistance from the Reltons, lee and other helpers, we attempted to erect Steve’s large dome tent. Although we eventually managed this, there was no way the pegs were going to hold it down for more than a few minutes. Fortunately, Mike Burrows had a spare, smaller tent that he loaned us for the weekend.
Lower Campsite
Quite a few people set up their tents in amongst the trees and bushes up nearer the circuit to try and find some shelter from the relentless wind.
Fowlmead
Fowlmead is a excellent venue for cycle racing; there are two loops that can be joined together by a testing hairpin bend to make a 3.5km circuit. The first race was held on Friday evening, and was one lap standing start, anticlockwise around the smaller, Lydden Wood Loop. There was a second chance, for late arrivals and anyone else who was keen to have another go on Saturday evening as well.
Although Quattro was wonderfully stable in the ferocious Friday night wind, I probably should have had a go on Saturday as well, when the conditions were a little calmer. Steve went four seconds faster on Saturday but the event was eventually won by Lee Wakefield in his Beany with an average speed of 36.0mph (61.7kph) on his fourth attempt. I came 21st with an average speed of 29.5mph (47.5kph).
Start of One Lap Flying Start

Saturday morning’s race was a one lap flying start, clockwise, using the whole circuit. The wind was still very strong but because it was coming from the Southwest, it gave a welcome push along the main uphill straight. I was very pleased to get 9th place with an average speed of 32.7mph (57.9kph). Steve won with an average speed of 38.3mph (61.7kph).

Eddie Robbins in his Rome

We were all very impressed when Eddie Robbins in his untested Rome took 7th place. Rome is based on a tiny pre-Beano, K-Drive moulding that I was doing nothing with. Eddie has added a spacer along the length of the faring so that he can use standard length cranks. With a few refinements, still to come, it should turn out to be a very swift machine.
Faired Ice Monster


Tim Parker at Speed

Another new machine and a very talented junior rider is Tim Parker in his very elegant Faired Ice Monster, 12th place with an average speed of 32.1mph (50.4kph).

Saturday afternoon was still very windy and the one hour time trial on the full circuit was started with the slower riders going off first, at 30second intervals. This worked very well from a safety point of view, as there were never any riders with vast speed differences on the track at the same time.  I went off about 75 minutes after the start and was pleasantly surprised not to be overtaken by another machine until I was nearly ¾ of the way through my hour. Quattro was running smoothly and my legs were feeling good, I was delighted with 7th place and an average speed of 31.1mph (55.2kph). Surprise, surprise, this was also won on by Steve with an average speed of 37.8mph (60.8kph).
Quattro, Wayward Cousin Wobbly Bob & Spiderman
The two hour mass start race was held on the Sunday, with the faster riders going off first. I started off well and got to a similar running position as the one hour time trial. Unfortunately for me, the un-faired and part-faired bikes were running as a large group, doing bit and bit, which gives them a higher average speed. Being faster up the less wind assisted hill, they caught and passed me. I put my loss of speed down to lack of training and old age but towards the end of the race, I realised I was running with a puncture. Having four wheels means that a puncture is unlikely to cause any danger but the downside is that a slow puncture can go un-noticed. The first sign I had was that the steering was starting to pull to one side with about 20minutes to go and then a couple of laps later, it was completely flat. I had two spare wheels at the side of the track and could have stopped and changed one in a couple of minutes but I decided limp on to the finish. I ended up 18th with an average speed of 29mph (46.7kph), which seems a little optimistic. Guess what, Steve won with an average speed of 37.7mph (60.8kph).
Overall, I am delighted how well Quattro performed. The only mechanical issue other than the puncture that needs looking at is that some of the grease has already come out of my new UJ bearings. I think that part of the problem may be that they are only shielded bearings and they are spinning at 90deg to their normal rotation, so the centrifugal force is pushing the grease out.
UJ Bearing Leaking Grease
Champion, Steve Slade, Second Howard Yeomans and Third Lee Wakefield

Oh I nearly forgot, Steve is World Champion again!

Wednesday 6 June 2012

35mph!

Quattro Reeling in Free Wheelie
So it was back to Reading last Monday with the new UJ bearings fitted to Quattro. This time, I left off the spats and locked the rear steering.
Keith and Chris from Wycombe Abbey School met Jim and I there as we were planning to do some tests on the Greenpower car (Free Wheelie) at the same time. We were also met by Graham Sparey-Taylor who was hoping to have his Quattro there but unfortunately ran out of time to complete the latest modifications.
The tests we can do at present are still not very scientific but the bottom line is “how fast does it go mister?” Well the answer is a bit faster than last time. The weather conditions were much better with a temperature of about 24deg and very little wind. I was able to maintain 30mph (48kph) reasonably comfortably with a number of laps at 32mph (51kph) and after a pep talk from Dr Graham a sprint lap at 35mph (56kph)!
At the very beginning of my Quattro development, 35mph was the magic number that I was trying to achieve as a racing speed. This is generally about the average speed of the winner (Steve in the Beano) of most BHPC races. Unfortunately one fast lap around Reading Velodrome is not the same as a 40 minute race on a twisty cycle racing circuit but at least it is a step in the right direction.
Forgive me for the next bit, as it is largely a repeat of my original Quattro article for our club magazine but with some of the numbers updated.
The power required to maintain 35mph is made up of two main elements.
Air Resistance Power = CdxA x½ρV³
Where Cd is the drag coefficient, A is the frontal area in m², ρ is air density (1.2kg/m³ at 20°C) and V is the speed in m/s.
Rolling Resistance Power =CrxVxN
Where Cr is the rolling resistance coefficient, V is the speed in m/s and N is the normal force in Newtons (kgx9.81) on the tyre.
There are other mechanical losses such as bearings and chain friction but these are small and I will ignore them in the following examples.
35mph Beano Power
The Cd for the Beano is probably about 0.12 taking into account gaps, wheels etc. The frontal area is 0.35m².
 Air Resistance Power = CdxA x½ρV³      gives 0.12x0.35x0.5x1.2x15.5³      =94W
The Cr is about 0.006 for good cycle tyres, Steve weighs about 65kg and the Beano about 17kg giving a total weight of 82kg.
Rolling Resistance Power =CrxVxN        gives 0.006x15.5x82x9.81              =75W
Therefore Steve needs to put out about 169W to maintain 35mph in the Beano.
35mph Quattro Power
The frontal area of Quattro is about 0.6m². Let us assume the drag coefficient is the same value of the Beano at 0.12.
Air Resistance Power = CdxA x½ρV³       gives 0.12x0.6x0.5x1.2x15.5³        =160W
If I was to use my Solidworks CFD drag figures of 6N at 15m/s this would give only 90W which I don’t believe, as there is no account for gaps, rotating wheels etc. More on this another time.
Again assuming a Cr of 0.006, I weigh about 76kg and Quattro weighs 34kg giving a total weight of 110kg.
Rolling Resistance Power =CrxVxN        gives 0.006x15.5x110x9.81           =100W
This makes a total of 260W which is the high end of an attainable figure for me and gives an idea of what I am up against!
35mph “Upright” Racing Bike
Putting things in to perspective, a racing cyclist in a crouch position has a frontal area of about 0.3m² and a Cd of 0.9.
Rolling Resistance Power =CrxVxN        gives 0.006x15.5x82x9.81              =75W
Air Resistance Power = CdxA x½ρV³       gives 0.9x0.3x0.5x1.2x15.5³           =603W
This makes a whopping 678W!

Greenpower Car Tests
Free Wheelie at Reading
The Greenpower events are aimed at secondary school children, are all four hour endurance races and are generally held at motorsport venues around the UK. All teams have the same motor and six 12 volt batteries to be used in pairs. During the race at least five team members must drive the car, and up to six further members can act as pit crew. Other than the motor and batteries, teams are free to design a car from scratch, or build one of the kit cars. All cars must follow a set of strict safety regulations but there is quite a variety of designs sitting on the grid before a race.
http://www.greenpower.co.uk/

Free Wheelie's First Race at a Soggy Silverstone
The same equations can also be applied to Greenpower Electric Cars but at the moment the target speed is 25mph (40kph) for the Wycombe Abbey School car.

25mph Free Wheelie Power
Again there is a bit of guesswork for the Cd of Free Wheelie but it is probably about 0.25 taking into account gaps, wheels etc. The frontal area is about 0.35m² and the speed is 11m/s.
 Air Resistance Power = CdxA x½ρV³      gives 0.25x0.35x0.5x1.2x11³         =70W
At the moment, the car is fitted with 16” (349) Schwalbe Marathon tyres. These were one of the tyres tested by John Lafford.  http://www.hadland.me.uk/rolrec10a.pdf He found they had a Cr of 0.01, Keith weighs 80kg and the car about 95kg giving a total weight of 175kg.

Rolling Resistance Power =CrxVxN        gives 0.01x11x175x9.81                 =189W
This makes a total of 259W.
So it seems that at 25mph the rolling resistance is dominant.
If the tyres were changed for ones with a Cr of 0.006, and the weight reduced with a 20kg lighter driver, the rolling resistance power could be reduced to 0.006x11x155x9.81=100W.

As can be seen from the Reading picture, teacher Keith is a little large for the car as it is designed for smaller and lighter young ladies. His shoulders are causing extra air drag and a pupil would probably weigh 20kg less.
The aim of the trip to reading with Free Wheelie was to find the optimum gear to use at the next race. Greenpower cars have the opposite problem to human powered vehicles in that, the motor spins at about 1850rpm rather than the 90rpm of leg power.  To get the correct wheel speed, the car has a 14t freewheel on the motor and a 5 speed block on the axle with 47t, 50t, 53t, 57t and 62t sprockets which are selectable with a standard rear mech. At the moment, the gear ratio is selected before the race, based on battery capacity, race conditions and duration; it is not touched during the event. Most of the circuits are flat and the motor becomes horribly inefficient if it is run too slowly, which would flatten the batteries before the end of the race.
The two 12V batteries can provide about 17Amps to the 24Volt motor for the 80minutes before they are changed. This gives a motor input power of 24x17=408W, taking off a bit for controller, contactor and wiring losses probably leaves 400W.
Motor Performance Graph
At 17Amps, the motor is running at about 73% efficiency, which is near its peak. This means there should be 292Watts at the output shaft.
Working backwards from this power, using the same equations, with a lighter driver and better tyres (0.006), gives a theoretical speed of 14.5m/s or 32.4mph (52kph) with the rolling power at 132W and air power at 160W.
Splashing About at Silverstone
The next Greenpower event is on Sunday 10th June at Goodwood Motor Circuit and the aim is to complete the four hour race without any major breakdowns or flat batteries. The team will be using the Schwalbe Marathon Tyres and starting on a low gear. The gear could be increased at the first battery change, depending on the battery condition.

I'll not be there, as I am going to be splashing about at Fowlmead in the World Championships. The weather forecast is not very good, which is a shame but in Quattro, it may give me an advantage over some of the bikes.