Centrifugal Clutch p2

The internet is a wonderful thing – Facebook has a Riley enthusiasts group and the folks in that group have been jolly helpful  over the last few months. I have now found someone who has a spare clutch drive plate to replace my grotesquely worn one *and* he is very kindly putting new friction material on the trashed friction discs for a very reasonable price. So life is positive again.

Whilst I am waiting for the parts to be delivered to me – I continued my work on freeing up the centrifuge weights and fettling all the worn/damaged metal back to a working state. I noticed that one of the square locating pins was free to rotate slightly and seeing as that sits in a rectangular guide slot, rotating (and possibly snagging) is a bad thing. All I had to do was remove the flywheel to get to the securing nut. This was a Haynes instruction manual moment – “simply undo the 6 bolts and remove”. Except the ‘remove’ part took several hours and much correspondence to the Riley forum. Apparently hitting it hard with a large hammer is an acceptable technique (which worked). The pin is now in its right position.

All I had to do now was clean up the faces of the flywheel and the pressure plate so the new friction plates had a clean surface to work against. I was rather pleased to find the surfaces were pretty good and it was only the damaged drive plate that took all the wear.

I now wait for the parts to arrive and the engine can go back in.


Bit of a setback

Although the car could be driven the gearbox was not working as it should. This car is blessed with having the ‘pre-select’ gearbox which is probably the best mix of a manual and an automatic gearbox that was ever designed. In theory it is impossible to stall the engine when coming to a stop with the car still in gear. There is no clutch to depress… which is how I knew that something was not quite right. Because in our car you *could* stall the engine which means that the automatic ‘traffic-clutch’ was not doing its job. I was a little hopeful that this was due to the car not being driven for 40 odd years and that it would free up naturally. Turns out I was miles away from the mark!

Fellow Riley enthusiasts advised that I could access the clutch unit through a small access panel which is sort of true and I could see that something was not right as the moving parts…. didnt!

After a fair amount of soul searching because I had only just got the car on the road – I decided the best way forward was to to take the engine out and take a proper look. Surprisingly this didnt take as long as I thought and I am glad I did because at some time in its life the clutch had failed completely and shredded itself AND a fairly chunky piece of metal.

The car looks a bit ‘hot-rod’… sorry about that VSCC…

Here we have the engine out and turned sideways so that I can access the clutch unit

…and this is the Amstrong-Siddley clutch unit. Around the outside are 3 heavy weights that swing out due to the centrifuge effect and thereby clamp the 2 drive plates together via the friction surfaces. When the flywheel speed reduces at tickover the weights fall back to their normal position – releasing the pressure on the friction material and thereby removing the drive. Quite clever really but it is soooooooo heavy.

And here is the clutch unit taken apart – you can see that both friction plates are destroyed and the metal plate (bottom left) has had its bunny ears snapped off…

I rather suspect that something would have exploded had I not taken it apart at this time!

The search is now on for replacement parts / a conversation with the racing boys who actually remove the entire unit and dont use it at all… apparently the factory grandprix ERA’s (English Racing Automobiles) which had far more powerful engines had the same gearbox but they threw away the clutch unit because of the sheer weight. At this juncture I have not decided what to do…

Boarding Step

Originally, like so many 1930’s cars, the Riley had running boards. I think  most people know what they are – a flat surface between the front and rear wheels that you could stand on to get into the car. These were one of many things that were missing from the car when I bought it but  I do not need them. However our car doesnt have any doors so all passengers have to step over the side into ‘the cockpit’ and take their allocated seat. This is fine whilst we are young and limber (coughs theatrically) but we will get older and less limber and will need that step. Getting out is easy, you can just jump.

The challenge was to fabricate something that was elegant but very strong. I have a large sheet of 10mm thick steel which should achieve the strength and provide a suitably stable plate to then bolt the more aesthetic ‘Riley’ Aluminum step onto. My design is to bolt one end of the steel plate to the chassis *but* that means that weight on the step will send all the leverage onto the threads of the bolt – effectively standing on the plate will literally (over time) lever the bolts out of the mounting holes. The solution then is to have a vertical plate welded on so that the weight stress is now transferred to a shearing force through the bolt. That is significantly stronger.

Below you can see the plate cut and drilled to shape – the curve on the front face is actually symmetrical, the camera angle has foreshortened it.

As you can imagine, cutting and shaping 10mm thick steel takes some time, as does drilling and filing it to shape. You can therefore consider that fabricating these components consumed much of my day.
In the next picture you can see the concept of the design

The supporting plate was then offered up to the chassis rail, the mounting hole position transferred and then drilled and a thread cut into the holes to take 5/16th high tensile bolts. This was to enable me to unbolt the step at any time in the future should I want to. Welding it to the chassis would have been much stronger but would mean I was stuffed if I find over time that it is in the wrong place!
Below you can see the support plate bolted to the chassis and a hole drilled and threaded for the vertical support. This was bolted firmly into place and then a small weld was applied to the edges of the vertical support to set it in the right position ready for a full weld to be applied.

That done, the whole ‘bracket’ was sprayed black and left to dry. I fully expect the paint to erode off the bracket over time BUT it is nearly twice the thickness of the metal used in the chassis so I imagine it will last.

Finally, the picture below shows you the finished step. The Riley plate (custom made by a chap in Holland which I purchased a long time back at a Beaulieu Autojumble) is simply bolted to the bracket using stainless steel machine screws. It too is 10mm thick but it does not take any actual load because the bracket is almost as wide as the Riley Plate

Now all I have to do is make another one for the rear passenger which I may put on the other side of the car. But that can wait for a while until I know whether it will ever get used or not.

Engine tuning

I had setup the engine well enough to start and tick over but I could tell it wasn’t really happy as it seemed to splutter above 1500rpm. Cutting a long story short, I had to reach out to the VSCC boys for some guidance as I *thought* I had set everything up as per the factory settings. I also discovered that water was leaking into the inlet manifold which means that it was mixing with the fuel as it was sucked into the combustion chamber – not ideal. This meant that the fuel/air mixture was severely weakened and that turned the exhaust pipe blue because the gasses were excessively  hot.

Resolving the water leak required the carburetors and manifold to be removed, the issue(s) fixed then everything reinstalled. The ignition timing it turns out was setup totally incorrectly. To enable vintage cars to start, the driver used to manually adjust the distributor so that power was sent to the appropriate spark plug when the piston was at the absolute top of its cycle. Then once started it was adjusted so that the spark plug started to fire around 8 degrees before the top of the cycle (which allows time for the spark to fully ignite the fuel). The factory information doesnt mention how to configure the ‘engine running’ setting and I was leaving it at the ‘starting’ position. It turns out that this okay to start the engine but to obtain driving power its an order of magnitude out. The nice boys at the VSCC advised me to simply move the distributor when the engine was running so that it performed smoothly and then to lock it into that position instead. So thats what I have now done until the car can be driven on the road and more accurately tuned. It has completely changed the sound of the exhaust interestingly – it is no quieter just a lot smoother.

The brake shoes have been sent off for new friction material to be fitted and I await their return.

Below are a few pictures of the car as-is as I think many people have not seen how the car currently looks. There is a long way to go but it definitely looks like a car… I have since polished the blue out of the exhaust manifold…



Fuel things

Today I was going to look at the charging system but changed my mind and instead focused on the fuel side of life. The petrol tank was custom made when the body was crafted and it mimics the original tank in looks but aside from that it is unique being made from aluminum with a few customisations. The fuel sender is a marine item whereby the fuel level in the tank is measured by a bobbin that floats up a vertical sensor shaft. Realistically, there are no moving parts to the sender as the bobbin is not actually attached to anything – the shaft senses where the bobbin is and sends a specific resistance down the sensor wire to the gauge within the range 0-180 ohms (0 being empty). Trouble is vintage fuel gauges had totally different resistance ranges and in all the spares I had the range was 250-0 ohms. Completely opposite in functionality. For a while therefore I am forced to use a modern alternative which sort of looks alright but sadly not really what i want… hey-ho I need to know the fuel level so this is put back on the job list for another day. On the positive side I know that this part of the new wiring does actually work.

My other non period customisation is to insert an inline fuel filter as a just-in-case. The tank may have tiny pieces of aluminium dust still inside which would not be great inside the combustion  chamber and it also gives me a visual that fuel is actively being pulled from the tank to the engine.

Below is the engine bay with the newly provided fuel system. It finally looks like a proper vintage racing engine. One day I will get round to polishing both the rocker covers on the top of the engine.

I may remake the copper pipe that runs from the petrol pump (top left) down to the first carburetor – I think I can improve on the route and I sense the 360 loop is not actually necessary. Sadly whilst I am making huge strides forwards I do have a few issues to resolve:

  • Cylinder 3 is noticeably running hotter than the other cylinders (really weird)
  • The rear carburetor has a slight fuel leak (which may account for the above issue)
  • There doesn’t appear to be any oil getting to the inlet valve rocker shaft

I also took a look at the fuel/air mixture and I found the front carburetor was set too rich so I spent a happy hour ensuring the fuel levels were the same in both carburetors. That will make a difference to the tuning and it also meant I had to remake the mechanical choke connections because the length of the original ones were too short resulting in their enriching the front carburetor without my knowing. This is all resolved now.

All of the technical issues need to be addressed before the car can be driven – oh…. and I need to rebuild the entire braking system which had slipped my mind for a while 🙂

We are getting ever closer to driving the car after which I fully expect to unearth a whole host of other challenges….

Wiring and firing

Most of the day was consumed working out how to run the wiring from the new fuse panel under the dashboard to the various components that required power.

First priority is the brake lights. I definitely need those before I take the car on the road but because of the cars age, the UK laws mean I am not required to have any other lights or indicators fitted or working as long as I do not use the car in the dark. To be honest I will fit night lights and indicators later as other drivers are used to seeing them but the point is that I don’t *HAVE* to fit them to be road legal if I use the car in daylight.

Rolling under the car, I could see the remains of the brake light switch. Naturally, it was seized solid but I had purchased a replacement a long time back envisaging that if the switch still existed, it was likely to be un-salvageable. Engineering wise the switch is very basic and is operated by the brake rods ‘pulling’ on a spring which is connected to the ‘switch’. When the spring is under sufficient tension, it pulls the two connections together within the switch. What I didn’t know initially was how much tension was needed to make it operate and how much physical movement was required to obtain said tension. But I did at least had a working start point. The old switch was unbolted from the chassis and found to be riveted to a mounting plate which is not supplied with the new switch. So carefully drilling out the rivets to preserve the mounting plate, I transferred the new one into position and locked everything into place with small stainless steel screws.

The mechanics of all this are the easy part. The tricky bit is manufacturing a wiring loom. Running cables underneath the car means they are exposed to the elements and somehow all the individual wires would need to be covered in a single protective material. I therefore decided that I would cheat and take advantage of a multi-core cable designed for use with a trailer/caravan. This is a heavy duty 8 wire cable to safely provide driving lights (and charging) on a caravan and more importantly it is pretty cheap to buy as I only need about three meters. I worked out a route for the cable to follow then considered future wiring needs (night lights, indicators etc) specifically leaving myself sufficient lengths of wire for that purpose.

The brake and rear lights you will remember are now built into a removable panel that mounts in front of the two spare tires. Basically I needed to extend the cable from the chassis rail out to the lighting panel leaving myself enough free cable for the lighting panel to be removed whenever I need access to the spare wheels. You may note that the cable is run to the panel through flexible stainless steel tube (like a small scale shower hose). This is a period accessory on luxury cars back in the day so I thought I would add a little luxury to the car even if this is the only bit.

The main multi-core cable is  fixed to the underside of the floor safely against the chassis rail. All the wiring is routed down one side of the car and the fuel is routed down the other side as I thought I might as well keep them separate. I did consider fixing the cables permanently to the chassis rails so that I had the future option of removing the floor boards, but I cant think of an occasion where I would ever need to do that. Once the cable was routed I simply applied 12 volts to the appropriate wires to check everything worked before locking it all into place. The brake switch spring fitment did need some adjustment in order for the lights to come on when the brake pedal is slightly depressed but this was not difficult to achieve. The key part is that I now have working brake lights.

More exciting was refitting the newly machined exhaust manifold which I am pleased to say now nicely seals against the cylinder head. The radiator was refilled with water to double check there were no leaks (I have one small leak to fix) before I refill it with Fernox and a temporary fuel feed was provided to the carburetors as the real feed from the fuel tank is not quite finished yet. The engine started on the button (which still surprises me) and ticked over at 1000 rpm (way too high) so I connected my strobe tuning light and pointed it at my previously manufactured tuning disc…  Aha! The timing is around 3 degrees out so the distributor was rotated until the strobe flashed when the rotating disc was immediately below the pointer. SUCCESS!!! My removable timing device does its job perfectly and is sooo much more accurate than guessing where top dead center is. With this adjustment the engine speed  dropped to 700rpm but it is supposed to be between 350-400rpm.

I then noticed that both carburetors were partly operated at idle so it was a thing of moments to loosen the brass linkages allowing the carburetors to return to their natural closed position. I then tried to start the car but it refused to play. I wasn’t quite sure why but knew the engine started before I closed the throttles. Opening both throttles very slightly resulted, I guess, in the engine being no longer being starved of fuel and air allowing it to fire up and immediately settled into ticking over at 200rpm. That was a surprise as I have never had a car that could tick over that slowly but I am told there a whole bunch of vintage engines that run slower than that.

Using my 1960’s “Crypton Synchro Check” balancer I adjusted both carburetors to have the same air flow rate increasing it slowly until I achieved the desired tickover speed. I still need to get the fuel/air mixture right, which will require further adjustment of the air flow but to ascertain the right mixture – I have to drive the car (wow how exciting is that thought).

The picture below shows my trusty ol’ Crypton then below that tool in position but the  the engine is not actually running in the picture – the dial showed 3.5 on both carburetors when it was ticking over.

One final picture to show the engine ticking over around 400rpm, reasonable oil pressure at idle (45psi at 1000rpm) and the water temperature coming up. Its in Fahrenheit (which is not very British) and ‘boiling’ is of course 212F. For me a gauge that reads 100C as a maximum makes far more sense. I will replace the oil pressure gauge at some point as it will never read more than 50psi so its pretty pointless having a gauge reading up to 160 psi. I believe this one is for a Riley 9.

Next ‘essential’ job is I think wiring up the dynamo/regulator etc so that the engine charges the battery whilst driving… poop poop.

Petrol tank protection

The fuel tank sits out in the open in between the chassis rails just behind the rear axle and is also close to the rear wheels. It is therefore at risk of damage from stones being kicked up by the rear wheels into the sides of the aluminium tank. The tank itself is constructed from 3mm aluminium but a bit of protection won’t go amiss.

Vintage Bentley’s placed a strong mesh on the sides of their fuel tank so I am using their concept. When I had the tank made early last year, I already had this idea in mind and asked for mounting tabs to be built in. Conceptually the mesh would have a frame that wraps around the edges of the mesh. I tried bending a U channel out of sheet steel but just couldn’t produce anything I liked the look of. Next idea was using aluminium U channel and bending it to shape. I discovered that I can’t bend U channel and was forced to form it using straight lines. This took a couple of hours but sadly again I was not happy with the aesthetics. I grant you people are unlikely to notice the mesh but that’s not the point.

Next theory: the stainless steel mesh is pretty strong and doesn’t really need a frame so I tried it against the tank ‘naked’ and placed some shaped washers onto the mounting points. This looked far more elegant and is my current solution. The fuel tank and the mesh will sprayed body colour so the whole thing will be quite subtle.

To ‘complete’ the fuel tank, I turned my attention to the drain plug which needed to be drilled to allow locking wire to be fitted. The locking wire simply prevents the drain plug from coming undone by accident. This is a racing requirement.

The fuel tank is now ready to be refitted and a pipe fabricated running into the engine bay. Something for another day seeing as it’s -5 outside the garage at the moment.