Yelwoc Motorsport

Building the RX-7

Building the RX-7

The car is a 1982 Series II Mazda RX-7, which is almost the same as the Series I but has soft bumpers replacing the square steel bumpers of the previous series, though they are actually still there, behind the plastic and it has rear disk brakes.

The VIN is JMZFB13120066308, that translates as follows:-

  • JMZ = Mazda
  • FB1 = Model code for Series II RX-7
  • 3    = Body Code for 2-door coupe
  • 1    = Engine code for 12A
  • 2    = Transmission code for Manual 5-speed
  • 00  = Factory code Hiroshima

60001 - 70000 were made between 05/81 and 06/82. This car was first registered in UK 26 August 83 so spent along time at the docks or in the showroom.

The car was built to the Classic Saloon Car Club (now Classic Touring Racing Car Club) Pre-1983 Group 1 Technical Regulations.

There is a significant learning curve building your first car with subject such as corner weights, safety cage standards and even Weber carburetor parts.

The car is quite well balanced with the engine completely behind the front wheel axis; a front-mid engine car. The cornerweights are not too bad. Grassroots do some great technical articles including cornerweights.

Top | Beginings | Preparation | Engineering |

The Beginings

After spending nearly ten years supporting Jon Owen’s racing career, Jon gave me the chance to run his Sunbeam a few times, enough to give me the bug. So I started considering acquiring my own car.

Our experience told us:

  • Never convert a road car; because your adding new bits to an old and unproven vehicle
  • Never run a Japanese car; because the parts are hard to come by and expensive
  • Run a Ford; because they're cheap to run and there's Roland Hayes!

Looking through the old Group 1 Championship information the Mazda RX-7 did quite well so long as TWR were running it. As soon as they jumped ship to the Rover SD1, no one else continued with the RX7, so it never featured in the results again.

Taking scant notice of our experience early 1995 I bought a £900 1982 RX7 road car. 6 month later, a definite rotary convert (after years of being ‘Rover V8 man’), an accidental 9000 rpm in 2nd gear around a Slough roundabout blew the engine and the car’s road life.

We dragged the car back to Worcester with a view to repairing it. After some considerable time sitting in Jon's drive and under risk from his Missus, we returned from a session in the local and started on it!

I gave myself a budget of £8,000 with a possible stretch to £10,000, with an anticipated breakdown of:

  • Engine £2000
  • Chassis £3500
  • Suspension £1000
  • Trans £1500

(The actual costs (for ’96, ’97 and ’98); £15,000 to build (at least £5,000 was wasted!) and £6,500 in running costs, race entries and testing. Just as well I budgeted!)

Top | Beginings | Preparation | Engineering |


In Sept ’95 I stripped the car and prepared the shell (using 20kg Dry Ice to remove the tar) and gave it to a ‘prep shop’, with the intention of running it in semi standard form for ‘96. The ‘prep shop’ was going through difficult times, and it wasn't until October ’96 I got the car back, incomplete!

Thoroughly dispirited, I decided to put a ‘real’ engine in it and asked Pip Gardner of WGT Auto Developments in Nanthwich to install a ‘works’ 12A that would be good enough for a beginner. I also asked him to complete the car. Pip also supplied the Weber IDA 48, LSD differential, stabiliser/roll bars, exhaust system and Greddy EGT gauges. The LSD costs caught me out, as I thought £1,000 was expensive, but I failed to realise that the Crown Wheel and Pinion was not included and would cost a further £800.

I had also tried to make the car as Ford-like as possible to keep the costs down! (Should have started with an MKII Escort in first place!).

The custom LEDA suspension was built on the original RX-7 hub but based on Ford RS2000 settings (TWR used BMW E30 bits). The front brakes were sourced from Gareth at West Wales Rally Spares and are RS2000 AP Forest Calipers on Capri 3.0L discs.

Pip completed the works in about 2 months, but the car was still not race worthy. By this time the ‘prep shop’ had sorted out their problems and were back on line. Since they were local the car returned. Of course they spend most of the time redoing Pip’s work, which wasn't up to their ‘racing’ standards!!!

The Wankel engine has very high combustion chamber surface areas compared to a reciprocating engine and pumps out a massive amount of heat into the coolant and oil system. A custom copper 4-core radiator and two standard oil coolers were fitted.

We then discovered from various US and Australian sources that the RX gearbox casing had the strength of wet cardboard. With input speeds of over 10,000rpm possible, I pursued the Committee to allow me to use the Ford Type 9 box with Quaife internals. This box not only comes from the same era but has exactly the same ratios as the impossible to acquire RX competition unit. Again in hindsite the gearbox out of Toyota Corolla, that has more shaft bearings and strength to support the Supra, would have been better albeit initially more expensive.

One major problem with the car was the seating. Eve though the Japanese drive on the left, the car was designed for the USA market; LHD and there is a reasonable amount of space on that side of the car. The right hand side had severe transmission tunnel intrusions. The RX7’s roof is much narrower than the width of the car at the top of the main door panels. The result is the upper roll bar tubes are about 30cm in widest part of the 'B' pillar and get in the way of the driver’s helmet.

The car finally tested at Mallory in August 1997


Safety Cage

This was the first component installed in the stripped shell. Based on Jon Owen's experience and the fact that this 900kg piece of Japanese tin would be racing with Chevrolet Camaro's and Jaguar MkIIs a significant welded in cage was custom built. To this day, apart from the failure to maximise the head space, its the best feature of the chassis.

Suspension and brakes

The front suspension of the car is based on LEDA struts which are remanufactured RX7 items with 60mm (2.25") tubes and 22mm (7/8") rods. The valving is based upon a Ford Escort Mk II Circuit racer. The dampers are provided with a 20point combined bump/rebound adjustment.

The front roll/stabiliser bar was a 1 1/8" (28mm) Racing Beat item. The rear RB bar couldn't be fitted with the car lowered and so runs without it.

The car is quite well balanced with the engine completely behind the front wheel axis; a front-mid engine car. The cornerweights are too bad. Grassroots Motorsport Magazine do some great technical articles including cornerweights.

The first brakes were based on a Tarmac Rally Ford Escort/RS2000 MK II that would fit under a Championship regulation 13" wheel. Solid back AP 4-pot "Forest" alloy calipers clamping on a 10 1/4" (260mm) disc/rotor.

RX7 Front Suspension: LEDA Struts, AP Calipers, Racing Beat Rollbar The discs were cryogenically treated to extend their life.
A split braking system is used with 0.625" cylinders to both front and rear. Initially, there was little feel with the original 4.75 pedal lever ratio, so this was modified to about 5.7 which gave more feel. The pedal plate was enlarged to reduce the pressure on my foot!
Initially Mintex pads were used; 1177 at front, but difficulties obtaining pads for the rear resulted in the used of Hawk pads all round, with Black/Blue on the front and Blue at the rear.

The Leda struts are based on the RX-7 unit and use the standard stub axles and hubs. The TWR cars apparently used BMW E30 axles and hubs but if I did this again I'd use all MkII Escort components. The advantages offered are mainly cost, stronger stub axles and bigger bearings, though I never had a problem.

The roll bar shown is part of a Racing Beat front & rear set, though I never used the rear one as it interfered with the lowered tail.

The open undersides of the pressed steel TCA's were eventually plated to strengthen them.

Front Brake Upgrade

The brakes have been completely overhauled in 2007 with what are now 'really' the largest discs and calipers that can be fitted. 285mm x 30mm (11.25" x 1.25") front disks with AP "Palmersport" CP4567 radially mounted calipers. The discs are from Peugeot 406 M16 and are machined and fitted to a custom bell.

The standard rear caliper mounts finally sheared and so the opportunity was taken to replace the whole lot with calipers and Ford Granada III Cosworth rear disks. Hispec An hydraulic handbrake was also installed. - Note Hispec now do a handbrake level caliper.

For Goodwood this all had to be removed and replaced with period components!

The Palmersport AP Brakes; never used!

Rear End

View under tail of car, created by stitching two photos together (not entirely successfully!). The loose cable the 'Aeroquip' brake line that is normally secured!

RX7 Rear Suspension: LEDA Dampers, Holley Fuel Pump, Custom Exhaust
The fuel pump seen to the top left is a Holly 'Blue' capable of delivering 135GPH (US) at the Weber carburetor fuel feed pressures (4-5psi). The fuel tank used to drop through a hole in spare wheel well. One of the recent modifications is to complete re-engineer the fuel tank by cutting out the wheel well and fabricating an 80L tank that only has a small presence in the passenger cell. The car uses in excess of 70L/hr around Pembrey.
LEDA adjustable dampers and springs are clearly visible. Rear axle is at full droop here and the springs shown were found to bind after 20mm of travel from normal ride height! Another recent upgrade has been to use Spax coilovers using standard axle damper location albeit strengthened.
The asymmetric Watts Linkage can just be seen ahead of axle with the axle pivot point on the right side as we view it. While its design is a little eccentric it works well enough.
The rear axle has had a very small camber and toe-in induced; small enough not to require fully floating bearings at the hub ends. The twin pipe exhaust system joins just before the rear box. The two Cherry Bombs are well flattened by running over curbs and their contribution to noise reduction is suspect. These have been replaced by sturdier and heavier Racing Beat Pre-Silencers.

The exhaust has been re-fabricated to accommodate the pre-silencers. Recent experience at Goodwood where they were left in on an otherwise open exchaust suggests they do nothing but add weight!

Exhaust and Routing

The following image shows the routing of the exhaust system beside the gearbox and prop shaft and over the rear axle. Some recommend running the pipe work under the axle, but the extra bends are said to help reduce noise and ultimate performance requires 121 inches (308mm) from the exhaust port to tail pipe. RX7 Exhaust: Over Axle, Repackable Tailbox
The exhaust is mild steel with the front section from the header plate made from 2"ID 3/16" wall thickness water pipe! It is now very blue (800+C).
Tail box is re-packable. We use a ceramic blanket; used for insulating Nuclear power station coolant pipes; which holds together better than wool and just manages to survive a race day. We experimented with various combinations of this 64/96/128kg/m3 Rockwool, but it breaks up into fine filaments.
It would appear that one option is to use specially woven steel wool to wrap the perforated tube and then to use woven Rockwool around that.

The Exhaust temperatures were about 960C on full throttle and based upon the colour of the steel, above 400C at the tail pipe.

The following picture shows the routing of the exhaust system beside the gearbox and prop shaft.
The extra bends are said to help reduce noise and help to route the heat away from the driver on RHD.
However, this may be heating the gearbox and prop shaft as the EGTs indicate the gas in the front section runs at about 920-950 deg.C.
RX7 Exhaust: Cherry Bombs well past useful life
The routing of the exhaust is too close to the transmission and is probably a contribution to the blowing of the engines at Eau Rouge (Spa Francochamps, Belgium). This very steep hill requires full power in 4th, then a lift as the wheels are feet about the ground over the crest. It is believed the gearbox expanded longitudinally a few mm and pulled the selector out of gear. Land..back on full throttle...brown rain up the windscreen and a odd noise from engine!!!

Inside of the car was typically 45C and the floor pan is at least 100C!

Top | Beginings | Preparation | Engineering |