2nd pic shows display dimmed when the car's lights go on.Bike engined Locost
Overview
If you have a sensible car and a toy car, the toy car doesn't need to be sensible at all. So the replacement for the V8 MX5 need not be practical in any way! Into this category falls the lotus 7 type of vehicle. The Locost is a vehicle based on a set of plans you buy in halfords in a book by Ron Champion. There's a whole industry now to serve the home builders of these cars and a number of forums on the web for information (locost_builders and Yahoo)
The major plus in a Lotus & style vehicle is light weight. A motorbike engine takes this advantage further. Using standard suspension/brake parts in such a vehicle results in unacceptibly high unsprung weight so I'm going to use MX5 uprights and wheels and independent rear suspension together with alloy based brake calipers and non- vented discs with alloy bells to keep unsprung weight down. Wilwood market suitable calipers weighing only 1.4lbs each! It should be possible to end up on the road below 500kg. This is 300bhp/ton - supercar quick - equivalent to an MX5 with about 330bhp!
Starting out
Big project - major milestone is STARTING. So (august 2003) I've bought a Yamaha R1 engine off ebay, £940 gets me a 150 horse unit (5 valves per cylinder!) with 6 speed sequential gearbox weighing only 60kg. And a set of 4 old MX5 alloys for £40. I've done a spreadsheet looking at gearing, looks like I need a small sierra diff (I'll get a quaife LSD)(£400+ I think) and a diesel crownwheel and pinion 3.14:1 ratio.files/gearcalc.xls
Brakes
Here is a spreadsheet looking at brake biasing.files/brakecalc.xls If I use the same calipers and discs front and back I'll get a natural 50:50 front rear split. A proportioning valve will reduce rear effort when I press harder on the brakes giving ideal overall characteristics. Note a bias bar assembly will only give ideal braking on one surface unless it is constantly adjusted - something SVA take a dim view of! So for safety the bias bar is set for maximum traction on a perfect surface, then locked at this setting by SVA. On ice braking performance is therefore reduced by about 10%.
What about handbrake? My current plan is to have a mechanical caliper on a disc in the middle of the prop (you need a 2 part prop for a BEC) I'll cut gear teeth on this disc for a starter motor to reverse the car - should be cool. This means that the discs at the wheel are hydraulic only for an easy life.
Dashboard
I've designed an LED based dashboard using a bargraph in the shape of the engine's power curve for rpm (non- linear scale - most accuracy in the 7 to 12k range). mph is by a 3 digit 7 segment display. A further 7 segment display shows current gear. 3 analog bargraphs show oil pressure, fuel and temperature. A shift light indicates when an upshift will release more power from the engine.
Response rates for the various displays: rpm instant - new display every spark! mph 3 readings/second. gear instantaneous. oil & temp 10Hz update. Fuel 10Hz update further filtered to give time constant about 1 minute (so display doesn't wave around on bends)
Indicator tellbacks and mainbeam tellback also on the dash - these therefore will brighten and dim along with all the other displays.
Oh yeah - brightening and dimming range 64 to 1 in 2:1 steps - full brightness, 1/2, 1/4, 1/8, 1/16, 1/32 and 1/64. The display powers up at full brightness then dims 2 steps when sidelights go on, 2 more for dipped beam and 1 more for main beam. But there are buttons to make brighter/dimmer too.
The "intelligence" is in a xilinx FPGA with code in VHDL (93 + numeric.std library) So it's relatively easy to tweak. Mine will be running on a prototyping board from xilinx (spartan II demo board) but only about a quarter of the FPGA is used - and that includes all the unnecessary reset logic. So a more "self contained" version could use a really cheap FPGA....
Here are a couple of snaps of the unit under test - looks OK, just 1 cut/strap required on the display PCB
2nd pic shows display dimmed when the car's lights go on.
Stop press - added a "charge light" for BECs - The middle shift light comes on when the 12V supply is below 12.5V
Other electrics
I'll do my own "loom" using bike stuff. Only slight worry is that if I use the caterham heated front screen (+ wipers at same time) the bike alternator won't be able to keep up. I'll cross that bridge when I get to it. May have to do my own regulator to get a bit more out of it - we'll see. Certainly the bike battery should be small neat and light.
Bodywork
I don't like the cycle wings (+ unsprung weight + lots of "nut covers" for SVA) & I reckon the "clamshell" wings look naff. I guess I'll end up going the cycle wing route though there's a chance of getting some aluminium morgan- look wings made - I'll check it out.
More Chassis Notes
Got notes on (nearly) doubling torsional stiffness of chassis by FEA weilding geezer on one of the forums (thanks cymtriks). Got the tiger avon book by Jim Dudley to get some ideas on IRS design - not very impressed - poor or no triangulation. There's a single photo in RC's book which appears much better than the Tiger but details are scant. I'll have to do my own. Also most folk bolt bike engines and IRS diff direct to the chassis - that's got to buzz a bit I'd say - I'll get some void bushes in there!
Here's a solid model I'm building up of an IRS chassis and rear suspension using Mazda MX5 uprights. With the mazda "minilite- look" wheels (£40 for 4 with tyres) I can fit a 10.5" brake disc on an alloy bell with the wilwood dynalite 2pot calipers (£50 each and 1.4lbs weight). Pretty much minimum possible unsprung weight & not too silly money
Early Feb 2004
Purchased sierra diesel diff - 3.14:1 £70
Purchased 2x 14" with 150lb 9" springs and 2x 16" dampers with 125lb 10.5" springs from "dampertech" £316 delivered
I've tacked up the main frame of the chassis and tacked up the front suspension pickups - had to move LB & LC tubes & move FU inwards a bit to get book dimensions. Alan will weld it up properly next Saturday. Got the U-bracket things from dozracing, I'll probably get front upper bones from him too. The chassis looks the business, esp the back end where the IRS will mount.
Photos show front suspension alignment jig in place on tacked up chassis
I've made a pukka control board for the electronic dash, this has a XC2V30S spartan2 FPGA and a 18V512 flash configuration PROM. Since doing the board I've added a couple of features - a "charge light" for BECs which don't have an ignition light - this lights up when V less than 13.5V. Also I've put a speedo and odo adjuster in so if people change their wheels or diff they can adjust the speedo right again. The control board went together pretty well, only problem was getting a smooth and reliable 32.768kHz crystal oscillator - still not totally satisfied. I'm putting the PTH stuff on the underside so spades will plug onto the back of the unit and the 2 boards will plug into each other with only a small gap.
Mid March 2004
Chassis now fully together (except still working on rear suspension geometry - BTW thanks to P Bura from the Locostbuilders site for his help). As the pictures show, the differential fits (bit tight actually - I have 8.75" vertical gap in the "box" where the diff sits - other builders should increase this to nearer 9")
I've bought more suspension bits from www.gtstuning.co.uk and scrap mx5 suspension parts which should arrive next week and enable me to calculate my suspension roll centres accurately - I can't even finalise the track until I get them!
I also spent far too much money on a "wiring module" from www.nfauto.co.uk - they seem to have an interesting range of products some really good value - the "wiring module" ain't so good but will definitely simplify final wiring by a shed load!
Easter 2004
Ist prototype of the dash (after the one I killed that is) now installed on Gordon Reynolds MK-ninja. Still displays 2x rpm it should & the minor instruments all go backwards but these are very minor problems which will be fixed next visit, when I have a laptop! Otherwise - it looks FANTASTIC, really pleased! dash & switchgear all mopped up by sierra steering column stalks & my dash to give a very clean minimalist overall impression. Gordon has mounted it in the middle with a printed overlay which may give reflection problems in the sun - time will tell!
Here is some extra data about the dash module, including installation and commissioning notes. files/dashbunf.pdf
Lots of design work done & suspension geometry now finalised so suspension pickups are stuck on the chassis & I've started making wishbones (except my brother forgot to order the material). Here are the measurements to give optimal control over roll centres & camber control when using mx5 uprights:
Front suspension - chassis exactly like book Top wishbone length 11.25" Bottom wishbone length 16.25"
Machining required on front uprights: 1) ream out upper balljoint taper to take transit drag link end 2) track rod end pickup must be made thinner and its angle reduced by machining a "wedge" off top and bottom. Fill tapered hole with a plug then drill & ream taper FROM ABOVE to take escort track rod end mounted on top.
Rear suspension - chassis, bottom pickups18.3" apart and 2.5" up from bottom of chassis. Top pickups 25.8" apart and 10.2" up from bottom of chassis
Top wishbone length 9.1" Bottom wishbone length 14.2" Note bottom wishbone inner pivots are 14mm rose joints to give camber & toe adjustment. Outer pivots are mazda mx5 parts - I'm going to use powerflex poly bushes (their part no. PFR 36-111)
This gives rear roll centre 2.7" below chassis, front roll centre 4.5" below chassis. RC does not move wrt chassis for moderate bump/droop or roll. Camber control: effective swing axle lengths 80" rear 100" front. Track 52"
This image shows rear suspension pickup locations for my design with fore/aft measurements (in inches) Other locating dimensions as described above.
I'll put links on here to "solidworks" 3d solid models of the wishbones when I've finished them, so anyone who wants to do the same can just download them. Front top wishbone is just a book wishbone made 2" longer!
Ford escort steering rack will require considerable shortening of the middle bit and lengthening of the track rods. Details of this work and the rack location are not yet fixed. Because I'm aiming for very soft suspension & will use all the travel I need to get rid of the bump steer inherent in the "book" design.
Mid May 2004
Not as much progress as I'd have liked. - 1st job, I've machined the uprights -
1) make tapered plug with 3mm hole through middle
2) saw ~3mm wedges off steering knuckle bits to reduce the ridiculous angle mazda put on there
3) hammer the plug into the old hole
4) saw plug off flush then use drill tomake small holes round join so the weld has somewhere to go
5) weld, do other side grind flattish
6) drill through 6.7mm hole from top using 3mm hole to locate - then tap M8
7) bolt to bar of metal in lathe toolpost & use flycutter to mill flat
8) mill other side flat & parallel leaving 14.5mm thickness
still have to drill 'em out & get the tapers altered. Bet you're jealous of that hacksaw....
I've also made the front wishbones. Top I used 19mm 16 guage tube but bottom I used 27mm OD stuff with a 2.2mm wall - should be strong enough! I'm going to include the drawings as solid model "edrawings" from solidworks - if you want to look at 'em or measure 'em you'll have to download the (free) edrawing viewer from solidworks. I believe it works just fine! A few more weeks should see me with a rolling chassis (always a good moment) though there's the rack to do before then....
files/botwish.EPRT files/topwish.EPRT files/frbotwish.EPRT
note botwish is rear bottom wishbone for back right hand side. The pointy thing is a 'vertex' to allow a 'spherical bearing' in solidworks model of rear suspension - this is where the spring damper attatches. note it's forward of the half shaft. Also note ball joint U brackets show bolt hole in line with swivelling axis - in real life they will be rotated 90o so the bolt is vertical.
topwish is rear top wishbone. The round tube is forward side to allow clearance for spring.
frbotwish is front bottom wishbone with bolt holes for mx5 lower ball joint.
Mid June 2004
The front wisbones are on and the front uprights fitted. Rear upper wishbones are made. Lowers I've started work on the jig & turned the big bushes for the outer bushes - am using aftermarket PU mx5 bushes, these need bushes 40mm ID and 47mm long (allows 2mm for a big washer on the inside as well as the outside). I've borrowed a laser gunsight (really) to stick on the front flange & checkout bump steer. So this weekend I've lots of stuff to take home, including all the dampers - if I get the rear lower wishbones made I'll have a rolling chassis - bit of a milestone!
OK wishbones are made, the rear slots together perfectly, room for all nuts & bolts & plenty of camber/toe adjustability at the back.
Wheels
point forwards too - always a good thing.
Front suspension and rack (+ 4" extender) put together & a laser pointer elastic banded to the frront flange to check bump steer. The wheel is pointed straight ahead at a wall & the suspension moved up & down. The laser dot on the wall should go straight up & down. The height of the rack is altered until the laser dot's curve is near vertical then the track rod length altered till the curve is straight. A final tweak on the rack height and - bingo - no bump steer. That was surprisingly easy. Result? my escort mk2 rack is 3.5" too long (exactly) and needs to be at a height so that the top of the rack tube is 7.5" down from the top of the chassis.
Check
out the laser gunsight. The mx5 gear looks pretty good.
So I now have a rolling chassis! OK I'm painting the bits now but I'm still pretty chuffed. ( note I'm still well behind where I'd start if I'd bought a kit )
This week I've sawn the 3.5" off the new "quick rack" I bought off rallydesign for £49 and machined the end to take the joint. I had a bit of drama with the screwcutting on the lathe - the table of gears on the lathe is 5% out - I think it must have a metric leadscrew. Crazy. Anyway I've bodged it & it feels OK now. I've made a sleeve for the main rack body tube so it'll weld back plumb with just 3" off (worried about how far apart the mounts would be, but they don't look too bad in the flesh).
Early July 2004
Plenty of work done at the weekend but not much to show for it; the front bottom wishbones have been remade to use 13" (extended) shockers to correct the ride height. At the rear the lower wishbone inner pivots have been remade of more substantial material and the shock upper mounts have been attatched, but these will have to be moved to put the bolts at the right angle.
I started making the rack mounting, need to get hold of rack clamps. The rack has gone back together nicely, the mounts are exactly 7" apart & the rubbers are on (I was sure the little one was going to snap...) I'll get photos of the done parts this weekend - maybe even one of it on its wheels!
Pictures show car on its suspension. Note dodgy rack extenders - don't worry I'll make some proper ones! Remade shortened rack looks cool. Today I bought a load of bolts and some peugeot void bushes part QH EMS 1859 (it's what was in the parts shop...) which I'll use to mount the diff. And I have to start constructing the prop centre bearing c/w reverse, handbrake and rubber doughnut drivetrain cush thing. I've got a load of bearings and a starter motor... If I can mount the diff next visit to the forge I can get the half shafts and flange adapters made too.
September 2004
Where did all that time go: OK the diff mounts have been stuck in - now there's 1000Nm or torque on the diff so it will be interesting to see if it's OK, here's a photo
The driveshafts have been shortened to 16" and 17.625". I did this myself in the end, tapered the driveshaft ends to 2.5o and made a corresponding sleeve from CDS tube. I then cut 3 long 'v's into each end and ground a corresponding flat on the driveshaft taper. I then welded these up & hope the result will be OK, if not I'll get some specials made. The picture below shows the bits before welding. The shortening above allows 0.625" for a turned driveshaft flange adapter. A contractor was identified who'd do a friction weld on the driveshafts, but he wanted £60 a side + VAT - not very LOCOST! By the way, there's a potential problem with my plan to use a quaife ATB diff - you can't clip the driveshafts in if you do. No answer for this yet, needs a bit of thought! (note - sorted with extra clips & a bit of groove machining.)
Engine mountings - I've made 3 mounting points for the engine - all rubber isolated, shown in the pictures below. note a long 10mm bolt is shown where the cortina mount with a 10mm stud on each end will go on the right and rear mountings. The front mounting uses the same suspension bush I used for the diff.
I've also dressed the welds where necessary ready for the ally body panels. I need to make the propshaft centre section (handbrake+reverse) and the pedal box, then I can transport it home and start skinning the chassis.
Meanwhile I put a sump baffle in the engine (picture below), made the relief valve tulip out of a sawn up electrolytic capacitor can in the end! I fit an extra diaphragm spring in the clutch for good measure, seemed to go together OK & clutch still works!
December 2004
Hmm - just spent months doing up my kitchen, & now it's Christmas... Never mind - I've spent a small fortune on a quaife ATB diff to go in my sierra diff and a set of wilwood dynalite single callipers. I've turned up the driveshaft adapters ready to rebuild the diff to its new spec - just waiting for new taper roller bearings from Classic Conversions. Pictures will be posted as & when. The calipers are SO light - 1.4lbs - I have a good feeling about unsprung weight etc. esp when I make up the discs with alloy bells - should look good too.
I also have to work on the design of the prop centre bearing - mind you it won't be in the centre, it'll be much nearer the front where the extra tunnel width should let me get an acceptable handbrake and reverse cog diameter. I'm thinking about a stainless disc for this - wouldn't want rust to stop my reverse working....
Today (9th) I've finished machining the driveshaft adapters to mate the mx5 CV joints to the sierra diff output flanges. These also had to have a groove machined at the splined end to clip nicely into the quaife ATB (& not fall out again). The picture shows the job very nearly finished (and the ATB itself)- I need to finally torque up the bolts & dremel the socket cap bolt heads so that the CV joint still fits - they just need a whisker scraping off. On the plus side it'll stop 'em ever coming undone!
The photos above show the crownwheel being torqued onto the ATB unit, and the whole differential assembled with the driveshafts in place. You can see where a small amount of material has to be filed off the inside of the diff casing to allow the ATB to go in (just behind the left output bearing carrier). The bearing carriers are preloaded to 40 ftlbs with the pinion mesh set to zero lash at the highest spot on the crownwheel - it all seemed to go together very nicely in the end with new bearings, seals and O rings. One fly in the ointment, the filler plug is seized up solid - I'll have to take it to "a man" to get it off, 'cos I want it full of oil ASAP.