September 2020
30Sep
Fractory parts
The front wheel trigger disks have been manufactured by fractory.com and for my first laser cut order, I'm very impressed with the quality of the parts. Makes you wonder why we bother with hacksaws and drills, when the edges cut by the laser are so perfect. I had 20 disks made, and I'm selling them for £20 a pair, if you are interested in buying them, please drop me a line, or take a look at my ebay listing. I fitted them tonight, and they work perfectly against the XS608B sensors. I've set the ECU to expect 16 pulses per wheel, and the quality of the signal received by the ECU looks good. This should finally fix the issue.
29Sep
ECU changes
The radiator cooling fan is now set to turn off when engine rpm exceeds 3000rpm. ie if the fan comes on when I'm about to launch the car, the fan will now switch off as I launch, which will give me a bit more bhp to get the car off the line. I've removed several channels from the logging configuration, which I added for Blyton, but they're not useful so I've removed them again, and I've upped the logging frequency for launchRPM and DrivenSpeed to 100Hz to give me more granularity when checking the launch control phase data. Talking of which the base engine speeds now have an S curve, to try and get the car to accelerate more quickly from the standing start. I found an article discussing S curves, and I have matched the base engine RPM's to give the same profile. I need to be more aggressive with the % wheel spin to see if the acceleration can be improved. Its all really down to trial and error.
I've also multiplied the traction control spin target values stored in the ECU, by 0.88, so I can return the switch back to the x1.0 position rather than the 0.88 position its in now.
The front wheel trigger disks should be with me tomorrow, and I'll fit them for testing during the week. I've had a look at designing a 3D printed plastic reluctance ring for the Hewland output cups, and slotting magnets in to the ring to activate the GT101 sensors. But disappointingly when I tested the sensors with a set of 3mm diameter x 1mm thick neodymium magnets, I could hardly get the sensor to detect their movement. So I've had to abandon that idea. I'll look at modifying the steel reluctance rings instead. They dont have the required 5mm step heights, for the GT101's to detect their movement, and I think 44 teeth might be too many for the X10 to react to. At 150mph, 44 teeth equates to approx 1633 pulses per second (Hz), which the X10 should cope with, but I cant find the specification of the maximum input frequency. With two sensors running at 1633 Hz, that might be too much for the X10 to handle.
25Sep
4K video from Blyton
The cooling fan was running when I started my run, and I've found a setting in the ECU that allows the fan to be turned off when the vehicle speed exceeds the set value, so I'll change that to 5mph and next time I start a run, if the fan is on, it'll turn off as I set off.I took a microphone intending to plug it in to the Kodak 4K camera, but when I went to fit it, discovered the Kodak camera has a 2.5mm socket, and the Gopro, 3.5mm, so the microphone wouldnt plug in, hence the wind noise. I've an adaptor on order for next time.
I've designed a new front wheel trigger disk, to replace the sink drainers that work OK, but I was seeing dropouts at Blyton in the front wheel speeds. I drew it in Fusion 360, and fractory.com are making them for me. Delivery is around October 7th. They have trapezoidal holes, 16 in total, and should present a perfect straight edge to the sensor. I was going to use circular holes, but the straight edges should give a better surface for the sensors to work from. I'm selling them for £20 a pair so if you need any, please get in touch.
21Sep
New PB at Blyton
A good performance by me on Sunday at Blyton, with 0.6s shaved off my 2019 PB with a 58.37s run in the afternoon, finishing 6th overall. We had 2 practice and 5 timed runs, and during the day I was focused on trying out differing traction and launch control settings, since it was a good opportunity with no pressure, and that allowed me to take my time making adjustments and seeing what difference they made. With the launch control I was trying different launch RPM's, and on the five timed runs I made five consistent 2.1s 64ft times, which was pretty impressive. On the final run I made a mechanical adjustment to the car, which meant that almost all the harsh load was eliminated when I released the clutch, resulting in a smoother take off, and my quickest 64ft all day. Thats going to allow me to increase the rate of acceleration, and maybe increase the torque a little in 1st and 2nd gear, which again should reduce the 64ft time further. The fastest 64ft all day was that of Graham Porrett in the mighty 600bhp V8 Lola, with a time of 1.87s, and that was mostly down to the new compound Avon tyres that they were trying. They are quite impressive and it'll be interesting to see how they cope over time against the popular Pirelli Ultrasoft.Traction control was also working (finally) with the reduced spin targets. I ended up running with a gain of 0.88, and the adder was set to -2, which gave me great confidence accelerating out of the corners. I could be really aggressive on the throttle pedal and the car just shot off like a rocket. I've aquired lots of data from the seven runs so I can have a good look and see what other gains can be made. Yet again though, the rear speed sensors were showing large (~100mph) drops in the readings they were producing, and that was triggering faults. I tried enabling Filter on Rear Left A and Right A, but the ECU said that that wasnt supported. I changed the number of teeth per reading from 1 to 2 on the rears. I increased the spike delta wheel speed from 75mph to 120mph. The spike count decrement interval was dropped from 2.5 to 1.0s (so if no spikes occur for 1s the spike count is decremented). And I changed the spike count limit to 100. That in part helped. But I really need to sort this out properly, so its back to square one again. I'll have another look at the reluctance rings, and maybe remove every other tooth so there are 22 instead of 44, since the X10 expander really doesnt seem to be able to cope with the number of teeth at speed.
I didn't take a spanner to the car all day. I started the day with 10l of fuel onboard, and after three runs, topped it up with another 3litres, so I now know that a lap of Blyton uses exactly 1l of petrol. Once I topped it up, for the remaining 4 runs I just let the level drop lower and lower and there was no fuel surge at all. Now I know that I can let it fall to a much lower level, and continue to find the minimum amount of fuel before surge sets in. The Radium surge tank and Gemzoe tanks are a good combination.
I also ran some yellow wool on the near side barge board so I could video the direction that the air was flowing at speed, and it appears as though the boards dont produce a vortex down the side of the car to seal the floor. I'll look at adding maybe a right angle strip to the top edge to try and encourage the air to stay attached to the boards, which I can try out again at Castle Combe in October on the EcoBat sprint.
17Sep
Side skirt fitting complete
I've fitted the offside skirt, so thats both sides now swapped for the deeper strips. I'm looking at how well the barge boards are flowing air at the weekend, so I've stuck strips of yellow wool down one side, secured using magic tape. I'll fit the gopro to the bodywork so it can record the movement and direction of the wool.
16Sep
Side skirt fitting in progress
Direct Plastics has supplied the deeper Tufnol strips, and I've fitted them to the nearside, after spraying them with matt black paint, the gap to the floor is now 40mm on the Dunlops, so will be slightly higher on the correctly inflated Ultrasofts. I'll fit the other side tomorrow evening.
15Sep
Vinyl removed
I peeled the remainder of the matt vinyl off the engine cover, then polished the sticky marks off and applied wax to seal the surface. I replaced the steel M6 bolts that were holding the GT101 sensors on the 3D printed brackets, with aluminium bolts, which combined with the titanium nuts have saved a little bit more weight.
14Sep
Carbon shims fitted
The carbon fibre camber shims that Andrew O'Malley produced for me, are now fitted to the hubs. They were cut from 1mm and 2mm carbon, and are based on my drawing I produced in Fusion 360. Swapping them over takes seconds, and the Mygale shims are now back in my spares stash. Swapping from aluminium to carbon saves a few more grammes of unsprung weight.
Shims fitted
My Fusion 360 drawing
Speaking of unsprung weight, I've made a replacement brake pipe cover for the nearside front lower wishbone, the previous one was removed by the low pressure area behind the front wing, at Combe in July. They're supposed to be secured with a strip of tank tape, but I'd forgotten to stick any on thinking they'd never come off. Well, they do.
Aluminium brake pipe cover fitted to the lower front wishbone
The numbers are on for Sunday. Ages since I've had to stick numbers on, its still a faff. I've got some more Tufnol strips on order, to lower the skirts on the floor. Since the geo was setup professionally by Triple M in August, the car now has a slightly higher ride height, with more rear rake, so the gap from the floor to the 50mm tall strips was actually 55mm not 40mm, so 65mm tall strips are being delivered Thursday, and I'll get them sprayed and fitted for Sunday.
12Sep
TC updated
I've updated the TC Spin Target Map 1, to halve the values that were in the map I ran at Snetterton. I've also built a spreadsheet that runs through the switch positions for TC Select and tuneSwitch, so I can get an idea of the amount of wheelspin that the ECU will allow before the system takes over. I wasnt sure whether the tuneSwitch value was added to the result of the multiplier, or added to the spin target and then multiplied, so I connected the laptop to the ECU and displayed the all TC channels, and confirmed that the tuneSwitch value IS added to the result of the multiplier. So this value in effect offsets the spinTarget%.What that means is the driver can make adjustments to the offset, whilst the car is being driven. So if the TC is kicking in too early, but the overall clamping is still in the desired range, then the tuneSwitch can be selected to move the spintarget above the current wheel spin levels, to allow more wheel spin before TC takes control.
These three graphs show the vehicle speed (blue), the wheelspin% at Snetterton (gray) and the spinTarget% that the ECU will allow at Blyton (orange). The tuneSwitch allows the orange line to be moved with respect to the actual wheelSpin% (gray).
tcAdder -5.0
tcAdder 0.0
tcAdder +5.0
My tuneSwitch on the steering wheel is set to provide from -5 through to +6, so I can adjust the offset easily if I feel the TC is interfering too much, orif it isn't helping as much as it could.
11Sep
Trailer mods complete
I trimmed the plastic side panels on the nearside and offside of the trailer, fitted them both back on, the cutouts allow both the motor movers to operate. It was easy enough to trim the abs plastic. I used a stanley knife and steel rule to mark them out, and a hacksaw to cut them. I then tried the motor mover on my driveway, and its very smooth and easy to operate. The next job was to collect the van from storage, and return the trailer, and it was so much easier to manouvere the trailer back in to its parking bay at the storage facility. For next weekend, I will order some 70A cable, and a battery box, and I'll permanently install the battery in the trailer to make it easier for me to collect and return it. I believe the battery can be charged via the trailer electrics whilst the engine is running on the van. I'll do a bit of research to see if that is easy to do.
I'm very pleased with the result. The install looks really professional.
I am car 104 at Blyton next weekend. I've a plan in my head of what I need to do to improve my times. I've still got the traction control settings to adjust. I need to power up the ECU and work out if the tuneswitch adds to the slip before or after the slip % is multiplied by the multiplier.
9Sep
Motor mover installed
The motor mover is now installed on the trailer, following a big favour from Will Locostbuilder, who made me a pair of brackets for the chassis using some scrap he had in the workshop. They're made from stainless steel, and were tough to drill, but after putting the pillar drill to good use I drilled six 12mm holes in the plates, and using a power drill on the drailer, drilled another six holes for the plates to bolt on to. With the brackets in place, it was a simple job to offer up the motors, and clamp them to the them. Once all the bolts are torqued up, I can refit the side panels to the trailer, and work on the wiring inside.
Bracket bolted to the chassis using 8.8 M12 bolts and nylocs
Motor mover suspended on the bracket
How the motors sit on the stabiliser bar beneath the stowage on the RS2 trailer.
7Sep
0 to 60 in 3.2s
The rear wheel speed sensor brackets produced by Andy Laurence for me, were perfect at Snetterton. Both GT101 sensors behaved far better than the XS608B's, but, I'm still seeing random speed drops of 70mph, so for Blyton in two weeks time I shall turn on the left and right rear speed channel filtering in the ECU to see what difference that makes. The 0-60 time at Snetterton was improved by an impressive 0.7s with a 3.2s on my last run. To launch, I'm now holding the throttle open for around 1.5s whilst the boost builds to the required launch boost target, and when I released the clutch, I could feel the wheels spinning initially, but the car really shoved me in the back, and the data from the ECU shows that 2nd gear was selected 0.5s earlier than the best of my runs at Castle Combe in July when I had the other sensors fitted. This is in part due to the improved sensors, but I've also programmed the ECU to report the wheel speed per pulse from the wheels, rather than per revolution, and thats made the closed loop launch algorithm more effective, since the wheel speeds now react more quickly to RPM changes from the ECU than before.My carbon fibre camber shims have arrived. Andrew O'Malley cut them to my Fusion 360 design, using his CNC Router, and I have two sets, in 2mm and 1mm carbon sheet. I'll fit them to the car for Blyton, replacing the aluminium counterparts. Triple M said to be careful they dont come out, so I might place a small amount of gorilla tape on the top of the shims to keep them from escaping.
Carbon vs aluminium 2mm Mygale camber shims
This weekends job was to trial-fit the pair of Reich motor movers to the trailer. I've got to the point where moving the trailer in and out of storage is causing a real headache. The trailer is parked on grass, and it cant be hooked to the van to allow me to tow it out, so I finally dipped in to my pocket and bought a proper motor mover. On Saturday I removed the rear side panels on the RS2 trailer, and placed the motor movers behind the rear wheels. With the motors then moved up in to their fitted position, there is plenty of room beneath them to allow the trailer to be tilted, so I've got a pair of brackets being made that will allow them to be clamped to the chassis. The RS2 trailer doesnt have a C section chassis rail, like you'd find on a caravan or traditional trailer, so a pair of L shaped brackets are required to attach the motors. Brian James Trailers said that the trailer wasnt suitable for motor movers, but I reckon with the 4mm brackets I'm getting made, it should work fine.
I've had to remove a pair of rusty coach bolts from underneath the bottom of the trailer, to clear the square tube that passes across the width of the trailer and joins both the motors together. The coach bolts held the gorilla board to the trailer floor, and arent essential. I've also trimmed both the black BJT mudguards, so the motors can be extended forwards to contact the tyres. That was relatively easy and just needed a stanley knife to trim the ABS plastic.
The next decision to make is where to place the electronic control box for the motors. I might place it inside the plastic cover on the near side, ahead of the axles. I'll see how much cable run I have, as I also need to provide a 70A 12V battery to supply the motors. The two plastic side panels will need trimming to clear the motors, but the added advantage of that is that the motors are partially hidden when the panels are back on.
4Sep
Fressingfield Oily Rag Club Sprint
Following the visit to Triple M Motorsport on Saturday, Sundays trip was a very early 4.30am departure form Leicester for the 126mile journey to Snetterton. I was impressed with the new section of the A14 which must have opened last year. Miles and miles of 3 lane motorway, which cuts out the old section of A14 which used to wind its way around the hills and villages, rejoining the original route on the approach to Cambridge. With the weather deteriorating, I arrived at 6:50am and was allowed straight in to the 100 paddock where I parked up next to Tony Beesley and I started unloading. The day was cold, and cloudy with a strong wind, following very heavy overnight rain, the track certainly didnt provide much grip during the day, as track and air temperatures remained below 20C for the whole time. At Triple M, we totally reset the rear suspension, and tried a few changes to see if more grip could be extracted from the Pirelli Ultrasofts. Back on the F3 size tyres, I now have 1300lb springs on the rear, and Alan removed packing spacers and camber shims, altered the rear droop, antiroll bar settings, and increased the rear rake. After several hours the car was then corner weighted, and gave a satisfying 50:50 left:right split, and 40:60 front:rear. The chassis weight is some 3kg ligher than last year, despite carrying 5l of fuel in the tank. With me sat in the car, minus 3 pieces of bodywork and the front wing/crash box, the scales tipped 540Kg. My 12kg weight loss since the start of the year was finally paying off.
Restoring the Idle Control/Target Engine Speed from 1000 to 1100rpm when above 60C, appears to have cured the limp mode issue I had at Combe, when the revs dropped whilst manouvering the car.
The car wasnt running smoothly at all during the convoy run of the circuit. But she was cold, and stopping and starting in convoy isnt really her thing. I wondered if changing the engine tickover may have been a bad move, but it soon settled down for the remainder of the day. However, for the first timed run, whilst sat in the car, I struggled to start the engine for about 10seconds. It span over but wouldnt fire. It did eventually start, these tiny Lithium batteries really are impressive. After the first timed run finished, out came the 240V petrol generator and I charged the battery between the remaining runs using the Lithium battery charger.
Charging between runs maintained the health of the Lithium battery.