Aeroscreen glass

Well this evening was different and another step into the unknown…

The replacement glass for the Brooklands screens (actually called aeroscreens because they were fitted on WW1 aircraft) was of course new, the period 1930’s Brooklands frames are not new. Ipso-Facto naturally, the replacement glass doesn’t fit. It is very slightly too big for the original frames. Deep sigh of frustration – what to do?

I cleaned out the channels in the frames until they were spotless – but the glass still didn’t fit. Also, I discover that strangely one Brooklands frame is fractionally larger than the other. Strangely? Perhaps not, I have to remember they were made in the 1930’s when the quality mantra was probably “near enough is good enough”. Actually I should adopt that phrase at work, life would be far less stressful. But this doesn’t help me –  i need custom fitted glass.

I then had a thought – the people that created the replacement glass have ground the edges smooth. So if *they* can grind the edges smooth – then I can grind the edges back until the glass is smaller… This for me is a pretty scary idea, I dont even know if it is actually possible in a home garage. But when did that ever stop me?

My weapon of choice is of course my trusty angle grinder fitted with the most evil,  sanding disk I could lay my hands on. Spinning at 14,000rpm and with a light touch, I approached the glass with a degree of nervousness and gently brushed the leading edge  with the grinder… to my surprise, it *did* leave a sanding mark… so I continued gently stroking the glass with the evil grinder over quite a period of time until I had removed close to 3mm off one edge of the glass.

Below you can see the laminated glass held securely in a vice so that I could sand the offending edges. The ground areas are not visible so they dont need to be pretty.


I then did the same on the horizontally opposed side. Although this was jolly boring (and noisy – sorry neighbors) the glass now slid about 75% of the way into the frame. Fast forward a few hours and I now have 2 fully fitted Brooklands aeroscreens.


And in case you ask – yes the passenger aeroscreen *IS* supposed to be at an angle – If it was mounted horizontally true – the gap underneath would make it nigh on pointless. So nobody does that.

So far, i am properly pleased with how the car is progressing – our vintage sports car is beginning to emerge. IMG_3706

Poop Poop!


Exhaust Mounts

A short and sweet update – the exhaust system is now held on the specially made mounts with springs – in theory springs were used because it makes the systems quickly and easily removable, although I struggle to think why I would want to remove it regularly…

Below is the exhaust header mounting


And here is the tail end mount. There is another one in the middle of the system that looks the sameIMG_3701

And the instrument panel is coming along too. You can see the four holes that produced the disks i used to make the wind deflector mounts. The picture below was taken before I fitted the Brooklands Windscreens.


Wind deflector fabrication

The problem with having bespoke bodywork on your car is that a lot of parts available from retailers were not designed to fit the car which leaves you with a bit of a problem.  Take the two wind deflectors for instance (I can’t really call them windscreens). In period they come with a very simple 2 inch high mounting bracket on either end which whilst adjustable to a certain degree are not able to accomodate being mounted on anything other than a fairly flat panel. The bodywork on our car doesnt have *any* flat panels… The big butterfly nut allows you to adjust the angle of the glass so that in theory the wind is pushed up and over your head.


I dont like these mounts and they dont fit my car anyway so what is needed is an elegant engineering solution.

I have spent many a happy hour across several months looking at how other people had crafted a way of bolting down the wind deflectors. Some had simply used the bracket shown above and lived with whatever angle the screens ended up at (in some cases it looked horrible) but most had come up with something else. Some people had removed both the bracket and the fixture it is clamped onto replacing them with a simple alluminium panel folded to conform to the shape of the bodywork. In some instances this looked quite nice but it wasnt really the look that I wanted.

Some people weld a vertical tube onto a spreader plate. The plate then being bolted to the bodywork with 3 or 4 bolts and the screen end is inserted and clamped into the top of the tube. This was more elegant but I didnt like seeing the mounting bolts and wanted something cleaner. Having learned to braze-weld alluminium recently, a design concept formed in my mind as I really wanted the whole structure to be crafted out of alluminium:


Step one – investigate the challenge.

By far the most complex challenge is the highly shaped cowl in front of the driver, it has steep angles and complex curves. Having worked out where I think the best place for the screen will be – you can see below that the new screen vertical support needs to be cut at quite an angle in order to mount flush onto the curved panel

IMG_3682The round disk is a two inch ‘useful throw away’ from when i cut out all the holes on the intstrument panel – as luck would have it, I need 4 disks – 2 for each of the wind-deflectors and I had cut cut 4 holes.

You can see in the photo above that the disk needs to be slightly curved to match the profile of the bodywork and the bar needs to be cut at an angle somewhere around 45 degrees but its impossible to measure what that angle actually is. Rather than risk wasting lengths of alluminium – I cut a piece of wooden dowel (which is much cheaper) and sanded the angle until it fitted.

IMG_3683This particular support I then knew needs to be cut at 48 degrees. Before making that cut, using the lathe I drilled a deep hole into the end of the bar in readiness for fixing the threaded rod. This way when I cut the bar at 48 degrees – I would know for certain that the hole in the bar and the cut-off end piece would line up perfectly. In case you were wondering, the cut-off end piece would be used on the oppostite side of the body panel as it matches all the angles of the body work and allowing the nut to be fitted square to the threaded rod (see the sketch if my wordsd confuse)

A thread was then cut into the long bar and a length of studding fixed into place. Below you can see how all the pieces go together although for the photo I merely slid in a smaller bolt so you could see it as a unit.


Transferring all the sections to the car means I now had this:


The bar is purpously far too long to enable me to adjust the angle on the end until it  bolted in perfectly vertical. I needed the length so that a bubble level could be held against it. The bar was now cut to size and another hole drilled in the top to accept the screen fittment. Below is yet another trial fit and you can see where I am heading.


The final stage was to braze-weld the disk to the bar so that they became one unit. The completed support will now be cleaned up and all the machining marks removed.


I followed the same principle and fabricated the screen mount for the other side, the angles and lengths were all different but the process was the same.

And this is the finished result (before cleanup of the parts and fitting of rubber sealing washers etc). The broken glass will of course be replaced its only there for the photo.


Now onto the passenger screen.

So much to do…

As much as I don’t want to rush the creation of our car, sometimes the enormity of the construction work is over whelming. Fabrication of components takes absolutely ages as does the making decisions on where things would be best placed for aesthetics and function. Of course there is also no manual advising on which parts to fit first so sometimes you can fabricate something then realise that once it is done, you cant fit it because something else has to be fitted first.

I had pretty much made my mind up that i was going to get the engine running in stages. I.e. connect the minimum amount of fuel and ignition components as possible to check they worked before expanding down the line. Fuel wise, the pump is fitted but not connected to the fuel tank, the pump can suck petrol out of a temporary can. The starter motor doesn’t need the battery to be in its final position, it can simply be connected via an old cable etc.  If when connecting the battery to the starter motor, it turns the engine over, then I will know that part works. If when connecting 12v to the petrol pump it pushes petrol to the carburetors, then I will know that part works. \

The problem is that life is not as easy as that – the temperature gauge that i fitted I now remember has to be inserted through the front of the instrument panel so I had to remove that and work out where the sensor tube will run through the engine bay, through both firewalls to the instrument panel. Skip on a few hours and that is mostly done – except I then remember that the instrument panel isn’t anywhere close to being finished yet so I can the gauge yet.

The oil pressure gauge likewise – work out where the copper capillary tube needs to go, fabricate a complex series of bends into it (get it wrong and start again) until likewise you realise that this gauge too is inserted through the front of the instrument panel. Skip on a few hours and the pipe work is now complete.

The Ignition coil needs to be kept cool so I want to keep that out of the hot engine bay and mount it behind the first firewall. Easy enough, except that the main high tension cable now has to go through, over or round that firewall. Skip on a few hours and the coil is fitted with suitable cutouts made and (forward thinking), some protective covering so the cable cant short out to earth on the bodywork. Fabricate the rest of the ignition leads. Skip on a few hours and this is now done.

Ah but if I fit the instrument panel, it is going to be really hard to mount the two tiny windscreens which need bolting down *behind* the instrument panel. Think about how to make the windscreen mounts. I think some 12mm aluminum rod, welded to a plate specifically contoured to match the bodywork and at the top weld it to a machined 25mm spacer with a 20mm hole through the middle that the main windscreen mounting bar can slide into. Yuh that should work. Problem, the machining I can do but i dont know how to weld aluminium. Skip on a few hours and I have taught myself aluminium brazing which it appears is plenty strong enough. Below is my first successful attempt at ‘brazing’ 12mm bar to some 3mm thick plate. This is just a test piece and it is unfinished. Once cool i put the bar piece in the vice and tried to hammer off the plate, it didnt move. I then tried to lever off the plate with some very large pliers and the 3mm plate bent. The joint was completely fine! Bingo. I can now fabricate aluminium components and add a new skill to the portfolio.


Next onto the machining of the 25mm spacer which needs a 20mm hole through the middle. The lathe can drill up to a 10mm hole through the 27mm round bar, that means I have to machine out the inside of the hole until it is 20mm. I have never done this before and it took some time reading up and then getting the hang of ‘how’ to do that. Skip on several hours and I am still doing that… it is taking ages because you can only machine off a tiny amount at a time. Funny thing is I might not actually use the solution if it doesn’t look nice and I may change mind as I have another idea that would be easier.

We shall see… did I mention all this is time consuming? Fun though.

Instrument Panel

Being generally happy with the mockup version of the instrument panel I transferred the outline onto a sheet of 3mm thick aluminium. Not too surprisingly this is a lot harder to work into shape than the low density fibre board version plus it had to have an even gap around the edge so that at a later date I can place a thin piece of trim around the edge to match the interior of the car. 

Shaping the panel took a number of hours and I have no idea how many times I trial fitted it. Eventually I was happy and then worked out where the 5 brass mounting screws would be. I had planned ahead and the bodywork team had welded in thick tabs in perpetration for drilling and tapping. So all I had to do was drill a 3mm hole through the panel and the mounting tab to be sure the hole was in the same place on both parts. Next the tab on the car was drilled out to 4.25mm and tapped to take the 3BA brass bolts. The Instrument panel holes were drilled out to 6mm giving me about 1mm of clearance. 

The panel was bolted in with the protective film still in place. That will provide 2 functions: protection from damage whilst I drill sand and finish all the holes for the guages and the old switch etc, plus also it allows me to use it as a surface i can write in to mark out where all those switches and gauges will eventually be placed.

In the picture below you can see the new panel and the accurately placed guage images, the battery on-off switch and the klaxon button. I quite like the layout and think it should look quite nice once the aluminium has been semi polished. 

Next task is to cut all those holes out and a cutout for the steering column when it is raised back into position. 

Pipe Runs

Making fuel lines is a little frustrating but ultimately satisfying. Frustrating because when you bend pipe you only get one shot at it. Any one bend in the wrong place or at the wrong angle means throwing that piece away because you can’t un-bend it. Satisfying because whilst you know that no one will ever appreciate the work it took, YOU know that no one noticing it means you got it right…

Putting pipe runs to one side for a minute, having managed to get all the engine cooling elements fabricated,  sealed and holding water – I shifted my attention to the electricals and I have I *think* put the distributor into a position where it should be close enough to allow the engine to start, it can be adjusted once the engine is running.  There are no ignition timing marks on these cars plus I have updated the internals of the distributor to be electronic (and not need the unreliable points) so it is now impossible to tell visually or otherwise when the high voltage surge would be sent to a spark plug.

The carburetors have been setup based on experience in what I will call a ‘default’ position which again should be close enough to allow the engine to breathe but will need adjustment for maximum power (coughs theatrically). I have adjusted all the throttle linkages so that both carburetors open together and adjusted all the linkages from the throttle pedal to the carburetors taking out the loose play. I also had the foresight to fabricate a solid pedal stop in order to prevent a heavy right foot bending the carburetor linkages but all of that is now in place and ready to go

So back to the title of this post – pipe runs, more specifically – fuel.

In early motoring history fuel was fed to the engine using gravity requiring the fuel tank to be above the engine, technology moved on and by the 1920’s cars commonly had a hand-operated pump on the dashboard which pressurised the fuel tank with air. (yes seriously) following pneumatic principles, the compressed air pushed the fuel out of the tank towards the engine. When you stop the engine there is a manually operated tap on the fuel tank to let the pressurised air out again. The fundamental flaw in that design is that you needed to rebuild the air pressure over a surprisingly short distance. One of the jobs of a riding mechanic was to operate that pump throughout the journey. Luckily Riley’s were always on the forefront of design and in the 30’s it was common to have an electric fuel pump mounted onto the bulkhead in the engine bay. Thank you Percy.

Where the pump is mounted is not an aesthetic decision – these pumps are not able to ‘suck’ fuel up to a great height and they also need to be close to the carburetors to keep the pipe run short.

In my car, there are not actually many places the pump can be mounted. Once a decision was made, I then had to work out how to route the copper pipe from the pump to the carburetors making sure it would not be in the way of any moving parts such as the gearbox linkage.  I also have to factor in how the pipe from the fuel tank gets to the pump. 5/16 copper pipe doesn’t bend at acute angles so the shape of the pipe run also needs to consider those bend angles. Back to the first paragraph – get the angles wrong and you throw away the pipe. A challenge – I like it.

I first worked out how the pipe needs to route to get to the bulkhead, it had to go up and over the gearbox linkage but not obstruct the fuel mixture adjustment nut on the rear carburetor. Now I had to work out where the pump could sit giving me enough room for the inlet pipe and the outlet pipe. It was soon apparent that this was going to be a complex piece of pipe bending.

The picture below shows you just how complex it needed to be and the tool that pulls the pipe around a mandrel and into shape:

 photo IMG_3420_zpse6dfxscq.jpg
The full 360 degree bend (which was quite interesting to achieve!) is to allow the pipe to have some flexibility – the engine will move when it is running – the fuel pump is bolted to a non moving surface. If the pipe was rigid; over time the to and fro movement would cause a fracture in the pipe and the fuel pump would happily provide a pressurised spray of fuel all over a hot engine. This is something I am fairly keen to avoid.

In the picture below you can see the pump-to-carburettor pipe in its final position. I am rather proud to say this was achieved on my second attempt. The only reason had to reject the first try was because I wasnt paying sufficient attention and kinked one bend because the pipe fed into tool incorrectly. But I am rather pleased with only needing two tries. I suspect the experience gained restoring my Etype Jaguar (that blog is at which had a LOT of pipes that needed complex shaping, had given me sufficient knowledge to progress with confidence.

Out of the left side of the pump will be another pipe that will run straight down, back under the car along the inside of the chassis rail to the fuel tank – I have that excitement to look forward to another day – thats a 6 foot pipe to play with…

Once both pipes are finished I will solder the connections on and then antique the pipe work so that it matches my earlier brass work that can be seen below.
 photo IMG_3421_zpsle8xplr5.jpg


Oh how I laughed… i thought i had fitted all the parts i need to in order for the engine to hold water, sadly I found that was not the case. The hoses needed clamps, the pipe from the waterpump to the side of the engine was loose but most annoyingly – the exhaust manifold studs are it transpires drilled into the water jacket. Naturally when I filled up the radiator – i had 4 nice waterfalls…

Oh. I would normally just put a few drops of stud sealer onto the threads or more often as not seal the threads with PTFE tape but this is the exhaust manifold and it will get jolly hot. Much pondering later and many conversations with other Vintage car owners i opted for both PTFE tape and thread sealer. So the exhaust manifold is now fully bolted into place. I have jumped over the joys of making the exhaust manifold gaskets after the many hours of filing the holes of the 10mm thick stainless steel manifold.

 photo IMG_3409_zpspvkesaye.jpg

So here are all the completed components of the cooling…

The non standard (but period) waterpump complete with a brass bung (crafted on the lathe) to fill in another inlet hole that i have no use for.
 photo IMG_3414_zpsgn4msqzz.jpg

The non standard pipe work from the waterpump to the radiator and the one into the side of the engine
 photo IMG_3411_zpsoo1chov1.jpg

The non standard alloy pipe from the top of the engine up to the radiator – also fitted with the sender for the temperature guage
 photo IMG_3413_zpsf4elchyg.jpg

So now in theory I have a thermo-syphoning cooling system assisted by a mechanically driven water pump.