Jump to content

1967 S3 (Millington XE 2.0L)


Recommended Posts

I make my brake lines from AN4 aluminium hardline and route the rear brake similar to the fuel lines. The front ends behind the radiator, from where I continue with braided hose. I also buy from Ebay a first generation C20XE crank, which is very light with 12.7 kg. But in bad condition, so it needs a full regrind. I remove the timing /trigger wheel, as I'll be using an external trigger wheel because of weight and durability. The rear brake hoses touch the wheels, so I had to rethink quickly  bracket to keep them out of trouble. Its all in the details... and I've never seen these things on a Caterham assembled, so it is what it is... The ECU is now connected to chassis loom and talking with the AIM display, so all set for the engine loom. I also make pieces from scratch to plug the holes I had to make when fitting the engine (as it was more rearward than the original).










Link to comment
Share on other sites

  • Replies 51
  • Created
  • Last Reply

Top Posters In This Topic

I'll be running everything sequential, so I need a cam sensor. The Millington cam cover had a place for it, but of course the sensor I selected (blindly that is) didn't fit. So I had to modify the cover a little. And who could resist some bling... Lotus Racing sign to the radiator made with a laser. The hose angles on the radiator made no sense to me, so they needed some rework as well. Next was the drysump tank. Absolutely no space to fit anything off the shelf. The only space I had was in front of the driver footwell, but that is also where my steering column now was. And the exhaust. And the drysump pump. Of course I forgot to take a picture of the baffles inside the tank, but they exist... The cylinder head visited a machine shop to fit the larger (intake +2 mm , exhaust +1 mm) valves and my wallet was again also quite a bit lighter (the theme is all about weight savings...) 












Link to comment
Share on other sites

The coolant tank is the classic Volkswagen tank, it just needed a mount. The steering column needed a new cover as well. I found a nice oil thermostat from Mocal and I picked a location for it below the steering rack. I was running out of ideas where to place my oil filter, until I discovered a V shaped mount in the chassis below the intake manifold. I still don't have an idea what it was intended for, but with some modifications to my part it fit perfectly and the location is great from both a hose management and oil filter exchange. I had also enough room to fit the larger Ford filter. 










Link to comment
Share on other sites

All oil lines are either AN10 or AN12. 2 lines from sump to pump. One AN12 from pump to oil tank. One AN12 from oil tank back to pump. AN10 from pump to thermostat. 2 x AN10 from thermostat to oil cooler. AN10 from thermostat to filter. AN10 from filter to engine block. It is hard to capture the line routing in images (especially since they all black...)


I don't like long water hoses, so I made some 34 mm aluminium tubing for all long lines, it was also then easy to connect the water tanks, by passes etc to the aluminium tube.

You can also spot my oil pressure sensor versions 2 and 3 in the images... 


The chassis loom is ready and the engine loom starts to take shape, but to finalize it I need the engine in its final configuration.











Link to comment
Share on other sites

Engine build time! My main bearings are ACL Duraglide 5M2327-25, so 0.25 mm oversize. And they don't fit. We didn't realize, that also the thrust flanges are oversize, i.e. the width of the thrust bearing shell is also 0.25 mm larger. The crank needs to go back to the machine shop. My bigend bearings are King CR4017XP0.25 pMax Black. The Saenz ROD-VX-2.0E-149-01S come with ARP bolts and the instructions recommend to torque by elongation and not by torque, so I make myself a simple stretch gauge from a scratch piece of aluminium, which I can use during assembly. The structure of the Millington block is different to the C20XE block and it has a large bearing cradle, my clearances check out ok, they are 0.04 mm for the mains and 0.055 mm for the bigend. The thrust clearance is 0.08 mm. The piston ring gaps check out ok, no grindinding needed there. I forgot to by a conical piston installation tool, so I made ghetto piston ring comressor, luckily I didn't brake any rings... I maintained a record of both the torque and stretch and interestingly enough, a few bolts needed much more torque to stretch the same amount as the others. It was one of those moments, where you are not completely sure if this is right or not, but I went with the stretch. 












Link to comment
Share on other sites

The pulleys for the drysump belt lined up perfectly! Next was to figure out how to mount the crank trigger. I used a VR sensor from Magneti Marelli 352316170849, I chose a bottom location as there were belts all around. I also made a stand for the cylinder head out of scratch aluminium (from my first version of the Tillett seat rails, yes, I make everything look here much easier than it was in reality... ). 


When I trial fit the bellhousing it hangs below the sump. This will catch everything under the car, so I'll need to figure out some protection for it. I also make a plug for the no longer needed distributor hole. 










Link to comment
Share on other sites

I had ordered a swirl oil / air separator cap for my drysump tank from the UK, but they apparently forgot me, so I made one. I couldn't find any details how these actually look inside, so this is just my best guess and of course I again forgot to take picture of the internals... I digged into my scratch bin and off we went. I didn't remember I had used a thicker top plate on the tank to facilitate welding, so I was carefull not to burn thru the thin sheet (1.5 mm) I thought the tank was, and ended up with a slightly cold weld. I've unfortunately lately found one pinhole in that weld. So far after 1000 km I don't have any oil in the catch can, so the top part at least seems to work well. With a 70-80 degC oil tank surrounding your feet it will be cozy warm, so I put some Thermatec insulation around the footwell to prevent as much heat transfer as possible. The valves/springs/bottom plates/spring retainers find their place.


Then a major headache, the lash caps. How to grind them into spec, as I need to create a fixed clearance of 0.020 - 0.025? I come up with and idea to use my Tormek tool sharpener. I use its drill sharpening holder and use a steel rod with a 7 mm protrusion (valve stem size), on which the cap can rotate when I press it agains the grinding wheel. The rotation is created by the fact, that I'm never perfectly straight to the stone. The process of grinding the lash caps takes 2 full days and everytime I measure I need to put the cams back on. Those springs are stiff, so I made myself a press tool from 2 C-clamps to reach the threads of the cam caps. And when rotating the cams, you need to do it in sync, or your valves hit each other... (trust me on this). 


And one important detail - once the grinding is done, the lash caps need to be hardened with heat treating. Red hot and then drop into water. I found out about this almost accidentally and it would have been a catastrophe waiting... I guess every engine builder knows this, but at least I didn't.














Edited by Rosteri
Link to comment
Share on other sites

The I finalized the bellhousing bracket and behold, this was the first thing that I have so far hit with the car from underneath - it was worth the trouble! Some laser etching bling as well to bright up the day. A Cometic metal 0.051" gasket was used for the head, the piston edge protrusion was 0.25 mm and my squish clearance was 1.04 mm, while SBD recommends 0.85 mm. They have considered that the block is planed as well, but with my sparkling new block there was no need. This lowers the compression ratio a little, which ended up by measuring the head volume and using the given piston volumes, gasket thicknesses and my squish clearance in a 1:11.9 compression. 










Link to comment
Share on other sites

C20XE engines have several iterations of the timing belt - the early ones used a short belt (141 teeth) with a round tooth profile and the belt was tensioned by rotating the water pump, while the newer ones had a longer belt and tensioning pulleus. These tensioning elements connect to the metal timing belt cover, which I obviously no longer have in place. Water pumps have not just different tooth lengths, but they have also 21 or 25 teeth and different width belts. Finding the right combination was really difficult, so here are my numbers. 


Cambelt: Bosch 1987949091

Original fixed pulleys: SKF VKM 25211 (usually 2, but I used only 1, more on this later)

Water pump: Magneti Marelli 352316170849 21 teeth, round shape


But the belt didn't fit, it was too short. I guess this is mainly an issue due to the Millington block, but in the end I solved it by using a smaller pulley. I made a plastic pulley just to find a diameter that I could use, then ordered  4-5 pulleys randomly in that dimension to see, which one I could re-use. The best one was SKF VKM 81004 57 mm diameter from a Toyota, so I modified the Millington pulley support for it. The original C20XE pulley was 67 mm in diameter, so quite a difference was needed in the end. 


I also made an assembly of my Weber fixed 3 bar pressure regulator (modified it to take a fuel pressure sensor) and connected it into a neat package with the GM flex fuel sensor, so I can also run an alcohol map for the engine. I found a good place for it in front of the passenger footwell. 

Next was valve timing, this took me one weekend just to realize at the end of the session, that I had accidentally swapped the timings of the exhaust and intake... phuuh. My timing strategy was to place the max valve lift to the right position, as the final opening and close points are dictated by my valve lash. I made a large timing disc (invaluable) and luckily I did check when my valves actually open as a final thing, which is when I realized that things were not quite right.












Edited by Rosteri
Link to comment
Share on other sites

Thanks Steve and Mike! Next was the mating the geabox - and it didn't fit. Type9 has two input shaft lengths and the C20XE uses the longer one cut down a little. Mine was the short one. And the hydraulic clutch didn't even reach the pressure plate of the clutch... this was a bad day.


Next weekend I picked up a Type9 core for small money from a Sierra P100 diesel, it was from the early 90s and had the newest castings. The only issue was that the thread for the gearbox mount was destroyed. I also picked up some 304 stainless for the headers.


I also ordered a full set of gears, a new alloy case, semi-helical gearset, needle bearing mainshaft and some other bits from SPC / Tracsport in the Uk. 










Link to comment
Share on other sites

In parallel I started preparing for the headers, the exhaus flange needed some work. The port is oval and tubes are round, so this required some head scratching. I went for a 4-2-1 setup. I went with black steel for the flanges, as they warp less and used a special tig filler that works with both stainless and black. I connected the headers and the flanges with an inside weld, againg to prevent warping and creating a nice smooth intake. 











Link to comment
Share on other sites

Some soda blasting made the tailcasting of the geabox like new again. I shortened the new input shaft and started with the assembly. No instructions, only this one gentleman in Youtube, who shows the disassembly and assembly of a type 9 - thank you!! 5,5 kg lighter, yey! A friend made me an adapter for the clutch, as he had nothing else to do for the weekend - thanks!












Link to comment
Share on other sites

An4 for the clutch line and the bleed nipple to an easily reachable position. And it finally fits.. The engine/transmission goes in without drama, the engine mounts I attached when it is in place. The alloy case is thicker than the cast iron one, didn't take this into account. But there is clearance - some... and my new mount locates the bolts correctly, nice! And then, the clutch lines leak inside the bellhousing...


Ok, one more lesson. Apparently the copper washers are these days too hard, so from now on I soften them always before using them. Glowing red (happens fast) and drop into water (yes, copper behaves differently). I had to pull the engine back out again, soften the washers and install everything back.










Link to comment
Share on other sites

Next went in the water and oil lines. The belts took a few tries to find the right lengths (not easy to find in the first place). The drysump oil pump belt really has no adjustment, it just takes time to find the right distance (shims or spacers made to fit) and the tension is right, when you can take it off and back with your bare hands. I designed in my mind all kind of tension mechanisms, but there really is no need.


At this stage I took a few months break, then mid June 2021 continued with the headers. Thru bonnet was the only option from space perspective. My exhaust numbers:


Primary dia 44 mm

Secondary dia 57 mm

Collector dia 63.5 mm

Primary length 570 mm

Primary + y-joint + secondary length 850 mm


I really couldn't find a lot of reference numbers for C20XE headers, the Lotus7 forum had an old post of some touring car headers, they had 48 mm primaries and the length was 685 mm for the primary and 940 mm for the secondary













Link to comment
Share on other sites

After waiting for more than half a year my roller body throttles arrived. I tried to fit hoses for the oil breather, but the routing didn't work. A couple of joyful hours with the saw and welder, as I don't have (yet) a pipe bender. I also got a sausage filter from ITG (ITG if you read this, please send me one with a logo that is not upside down!). I destroyed a nice chuck when making the backplate.. argh. Small jobs to complete were done now, like the brake light switch, heatshielding, TPS, throttle cable, alternator bracket... 


The throttle body was of course not aligned with my (ported) intake, so I elongiated the bolt holes a little for some up down adjustment. Good to have a large milling machine!














Edited by Rosteri
Link to comment
Share on other sites

The intake at full throttle has a direct view into the cylinders! My injectors were EV14-580, with a measured output of 590 cc at 3 bar. This should be enough also for alcohol. Coil on plug is troublesome, so we made a quick adapter for the timing light. 


And it is ready for first start... 




Link to comment
Share on other sites

It started at its fourth try, then we ran it for 20 minutes at 2000 rpm to bed in the camshafts. After this it was off to the dyno, which didn't go too well - the throttle body wasn't balanced and then the engine started giving trigger errors... in the end we didn't do a single pull, it was a very expensive day. 


Trouble shooting continued back home and the trigger error was a lack of fuel - I had accidentally switched off the transfer pump, so my small swirl tank ran low and the trigger error was a miss-fire... The throttle balance turned out to be a much bigger problem.


The TB idle bleed screws were not machined correctly and did not bleed, also the other barrel was smaller in diameter and had much more bleed around the barrel than the other. I welded more material and carefully fitted them into the body, it was 2 weeks of work to match them close enough, so that the idle bleed screw could be used to balance them. We did a preliminary base map so we could drive a little and the car passed its inspection - it was road legal again after 14 years in boxes!


The radiator was leaking, so I ordered a new one from Caterham. I wanted to do a core exchange, but could find a core... oh well. I chopped the new radiator a little as well.


Action time again at the dyno, we came home with a nice baseline that made 187 kW / 7800-8200 rpm and 250 Nm / 6300 rpm. The engine really is a 2.0L with a 86 mm cylinder bore and 86 mm of stroke. The map is knock restrained, so next year we will go for ethanol fuel to see if we can increase the power. I'm also considering making a new header with larger tubing to make the top end breathe better, as it feels like it hits a wall at 8k. And a sequential gearbox would be nice as well... does anyone need an almost new Tracksport 5-gear? =) 


The weight of the car is 510 kg with all engine fluids but no fuel. 


This is it - thanks for reading and questions or suggestions are always welcome!! 








Edited by Rosteri
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...