Tuesday, 22 May 2012

Quibell Park Velodrome

Quest, Quattro & Wobbly Bob
Firstly a big thanks to Jeff (blogwat) for all the photographs, I have not used any of the ones I took, as his are so good.
We arrived at a cold and overcast Quibell Park Velodrome at about 10am on Sunday morning after a three hour drive from High Wycombe. Having said our hellos and signed on for the race, I did a few test laps. The handling was still horrible, just like it was at Reading and my low slung spats were catching the timing loop wires that were taped across the track. I stopped to check the wires and they seemed OK, so I carried on for a while.
We then all lined up for a three lap standing start time trial, which is not my favorite event, in a hefty Quattro but I gave it my best. The first bend on the track feels horrible from a standing start because the speed is low and you feel like you are going to tip up on the relatively steep banking. By the time the second bend comes along, the speed is high enough to make it more comfortable. I rode the whole event in my middle chainring to give me better acceleration, but it did mean that I was spinning rather fast on the final lap.
Main Race
After the time trial, Martin the race organizer asked me if I could do anything with my spats to stop them hitting the timing wires. The wires are probably only about 5mm diameter, which shows how little ground clearance I was running. He was quite reasonably worried that I could un-stick the wires and drag them down the track, potentially damaging them and causing an accident. As my time trial performance was not very good and the handling was poor, I was quite happy to take them off completely. This also gave me a chance to look at the steering to see if there was anything obviously loose causing the bad handling, which there wasn’t.
Lee Wakefield's High Speed Crash
The main race was due to be a 40 minutes mass start but was stopped after 10 minutes because of the 42mph (67kph) puncture and crash of Lee Wakefield’s Beano. The Beano lost a lot of paint but Lee was fine and the race was re-started with him in his Quest.

Although it was my fastest race yet (about 28mph-no official results yet), I found it a little depressing as the handling was rubbish and Lee and Ian Fardoe both beat me easily in their Quests. Ian Perry was the fastest multi-track with his exposed knee Wobbly Bob. The bad handling of Quattro seems to have appeared since I locked the suspension. Ian Perry had a go after the racing and was amazed at the way Quattro kept changing direction and didn’t dare try and go fast.

Watching Ian Fardoe’s video of the event clearly shows me struggling to hold a line around the velodrome.
Ian Perry Flashing His Knees
I was talking to Jim at work the other evening about the efficiency of the drive chain in Quattro and it occurred to me that what I had assumed about the Sprags being locked is not actually the case. Because of the slight speed variation in my UJs, unless they are lined up, the drive shaft will drive one wheel and then the other, depending on the relative position of the UJs. I think the speed variation is only about 1.2% but it could create quite a bit of drag as each Sprag locks and unlocks under torsional and side loads. A cure would be to have the aligned UJs fixed to the drive shaft and the Sprags on the outside. Having said all that, on a velodrome, I am probably driving the inside wheel for 80% of the lap.
After the 30min race I did stop quickly and turn Quattro over to see if there was any heat in the bearings etc. The only thing I could detect was a warm disk but that was probably from stopping quickly.
Steve in his Beano
I suppose I should mention that (subject to publication of the race results) Steve won the time trial and the main two races despite his chain falling off in the 30 minute race. He had to stop to put it back on but still managed to beat a newly spatted Howard Yeomans in his Great White.
Howard Yeomans in his Great White
I am beginning to think these spats are over-rated....!
I had a useful chat with Geoff Bird after the weekend and he mentioned my long connecting rods from front to back maybe vibrating on the rough track. I think he is probably right about the rods. When I looked at them yesterday, the rear ones which are thin and kinked seemed to be working in compression, which is not good. He also reminded me that I had similar problems on bumpy roads around High Wycombe. Hillingdon and Castle Combe circuits are smooth enough not to be a problem. Back to two wheel steering for a while I think.
UJ Outer & Bearings
While putting Quattro back into two wheel steering mode, I decided to check all the front bearings as well. I stripped down the front axle and found that my UJ bearings were knackered, one side slightly rough and the other side very bad. I know one of the bearings failed completely on the last day of ROAM but I can't remember how many bearings I changed when I got back to the UK. I have got 8 new ones here so I am going to change the whole lot. Definitely need to go up a size on the re-design, at the moment they only seem to manage about 1500miles.
After getting back from the frozen North on Sunday evening feeling a little depressed about Quattro and my performance, I have again got reasons to be optimistic. I will replace the bearings, sort out the steering and then it’s off to Reading again for another testing session........

Tuesday, 15 May 2012

Wheel Spats Part II

Having lowered the front wheels and replaced all the suspension springs with nylon spacers, I trimmed the spats to give 10mm ground clearance. This may prove to be insufficient even for testing at Reading Velodrome, but I thought it was a good starting point.
Underside of Quattro
Wheel Spats Trimmed
Once the spats were trimmed, I turned Quattro over to check the tracking. It was way out. The front was very toe in and the rear was slightly toe out. This would suggest that my new Ackermann compensating rods are not doing a very good job at controlling bump steer. Hopefully this will not be a problem for the spat tests, now that the suspension is locked out.
Foam Strip
I used some adhesive foam strip to locate the spats before I taped them in place.

Ground Clearance
All four spats are taped into position and have very little ground clearance.
Spatted Beano
Testing at Reading Velodrome was a bit tricky as the weather conditions were very blustery. Steve quickly got going in the newly spatted Beano before the rain started. He managed ten or so laps at about 38mph (60kph) but was having trouble with the wind and stopped when the rain came down.
I got going just as it started raining and was being blown about quite badly by the gusts. The spats were hitting the ground on a few of the bumps and the handling seemed horrible. I did about 20 laps, including three roll down tests from 28mph to a stop which took more or less two laps (950m). Graham Sparey-Taylor asked me to try and gather some data from my Garmin that he could analyze to verify the rolling and air resistance of my Quattro. We then took the spats off Quattro and I taped up the wheel cut-outs to reduce the gaps. I then repeated the 20 laps with roll downs.
The conditions were not good enough to draw any real conclusions from the testing we did. My gut feeling was that the spats made me a little faster. Also, I was expecting the handling to be better without them but in fact it was still horrible. So I will give them a try at the next race. Steve didn’t like them and felt they didn’t help his speed and will race without them. The banking at Reading is also a bit flat for Beano speeds and he was grounding the spats on the bends. Our next race is at Quibell Park Stadium, which is a large outdoor velodrome with a good smooth surface and steeper banking than Reading.

Wednesday, 9 May 2012

Wheel Spats Part I

My idea for the spats is to try and make a single oversize mould, and then make four mouldings that can be cut down and stuck on to Quattro. I don't want to spend too much time on them, I just need something rough and ready that I can test on the track.
I have tried to compare Quattro with and without spats in Solidworks CFD and there seems to be little difference in the drag figures. This is actually quite encouraging from the point of view of adding the spats as I have not modelled in the gaps around the wheels, there is no rolling road and the wheels are not rotating.
If they work well, I want to incorporate them in my Quattro re-design. My idea is to have the major portion of all four spats as part of the bottom moulding. There would then be a lower removable section that can be taken off to change a wheel or access other components. By doing this, I am hoping that I can get rid of the removable panels underneath; they are heavy, the thin edges are prone to chipping and they make the mouldings rather complicated.
The quick and dirty mould is made from two MDF female formers with a thin sheet of plastic expanded and clamped into position. The formers are simple aero shapes, the top one 900mm x 180mm and the bottom one is 300mm x 60mm and they are 225mm apart. This will make oversize, conical and symmetrical parts that I can trim to fit around all four wheels.

Wheel Spat Mould
The aerofoil section I have used is simply an elliptical nose and a radius tail section. The radii are tangential to the ellipse and where they meet at the tail, they have an included angle of 32deg.

Aero Section
The main body and head fairing of Quattro are designed using similar “homemade” sections. I experimented with more sophisticated NACA profile but couldn’t measure any significant difference in CFD.

First Moulding as Template
I moulded a couple of wheel spats that I am using as templates but then ran out of cloth, so my moulding man is making me six. Four are for Quattro and two for experimenting on Beano.
I am locking my suspension at the moment for the spat test, I have pushed the front wheels out a bit further to match the rear wheels and should end up with a ground clearance of about 125mm, with a horizontal bottom. There are three 5mm section O-rings fitted on each of the suspension struts to give a little cushioning.
The reason I am locking the suspension is so that I can bring the spats right down to the ground to get maximum benefit. If there are any benefits that is, hopefully we can get to Reading Velodrome at the end of this week or early next week to give them a try…..

Tuesday, 1 May 2012

Castle Combe April 2012

Quattro I and Quattro II

The second race of the year was at Castle Combe motor racing circuit. This has generally been an early season event and as it is a wide race track, it is normally run as a single two hour race.

It is a simple triangular track, three main comers and three straights with kinks in each, plus the addition of chicanes in two of the straights. The lap distance is 1.85miles (3k) with a slight downhill along the main straight, when going clockwise.

The weather forecasters predicted the worst day of the year so far and they were not wrong. There was continuous heavy rain and very strong winds. When Steve and I arrived, there was a small group of competitors huddled on the sheltered side of one of the buildings. Unfortunately we don’t bring enough people to the race track to justify them opening the cafe. When we joined the group, there was a general discussion going on about whether to abandon the meeting or to shorten the race.

A number of people then volunteered to brave the conditions and do a test lap. Ian Fardoe’s video can be seen here. http://www.youtube.com/watch?v=xCZe4d2q6gM

This was the first time I had seen Quattro II in a working condition. I had initially made enough parts for two and sold an almost complete pile of bits to Graham Sparey-Taylor, for him to build up. The main thing he was short of was a head fairing; I had chopped the other one down for the ROAM trip. He made his own, rather lovely, carbon fibre version from the original mould, as seen on the right of the picture. Graham has also taken out my original moulded pedal box and replaced it with a tubular steel bottom bracket support. This was because there was insufficient adjustment in the leg length and he also wants to add an electric assist at the front.

The conclusion from the lap testers was inconclusive, with some saying it was too dangerous and others wanting to go for it. I took Quattro out for a lap and was pleasantly surprised how well it handled the conditions. In the end it was decided that there would be a half hour mass start race for anyone who wanted to compete. Steve was keen to have a go in Beano but I quashed his enthusiasm as I didn’t want the job of mending it if he did crash.

We lined up on the track and started the race, not realising that Ian Perry and Ian Fardoe were still out on a warm-up lap. For the first time in many years, I found myself leading a race. As the wind was predominantly blowing up the hill, I found my pace was much more even than normal at this circuit. My forward vision was OK as I can look through the vent hole at the bottom of the screen, trying to find the apex of the corners was a little more difficult because of the raindrops on the screen. Inside Quattro, I could tell the conditions outside were bad, but I was amazed how little I was being blown about. Before the race, I was worried about the effect the side winds would have on my rear steering, but if anything, I think if may be compensating for the side gusts.

I maintained a high speed but not flat out for the majority of the race and kept checking the mirrors as best I could. I was consistently passing other riders but had no idea where the two Ians were. Unfortunately for me, the first time I saw Ian Perry in my mirrors, was on the last corner when he came sprinting past. It was a fair result however as he had to make up a lot of ground on me due to his late start. I am also happy because his machine, Wobbly Bob is a cut down version of my Bubble and Squeak.
Ian Perry and Wobbly Bob at Monza 2011
Wobbly Bob 2012

These are the provisional results for the day.
Ian P.                8 laps. 36.13.   m/o
Miles K.            8 laps. 36.16.   m/o
Ian F.                8 laps. 38.11.   m/o
Andy F.             7 laps. 40.55.   u/p/o
Guss N.            7 laps. 41.15.   st/sp/u/p/o
Geoff B.            6 laps. 38.20.   Sp/u/p/o
John L.              6 laps. 38.41.   m/u/p/o
Yoshamora.      5 laps   36.27.   st/sp/u/p/o
Sam K.              5 laps. 36.51.   m/u/p/o
Graham st.        5 laps. 42.31.   m/o
Martin D.           4 laps. 38.59.   st/sp/u/p/o

Clare k.             1 lap.    5.51      l/st/sp/u/p/o
Barney h.          1 lap.    6.13.     u/p/o
If I am not mistaken, the points from this race added to my Hillingdon points puts me first in the championship! It won’t last unfortunately.
My average speed was about 24.5mph (40kph), which I am quite pleased with considering the conditions. Graham was not quite as lucky in Quattro II as his pedal fell off halfway through the race. He can’t blame the designer for that one though!
My Quattro was un-changed from the tests we did on the velodrome in Reading. I did leave the extended tail off as its benefits were dubious and I spent some time finely tuning the tracking the day before the race.
Our next event is on 20th May on a velodrome in Scunthorpe and I am hoping to add some spats (wheel fairings) before then.
CAD Image Quattro with Spats