Intake Manifold Replacement

I admit to being a wuss and booking a garage to refit the intake manifold in order to fix the oil leak under the dizzy, I was quoted 5 hours = couple of hundred quid. I thought this isn't too bad really, already have the parts so I said ok. The bloke at the garage asked who built the car and was surprised when I said self-build and I wanted something as straightforward as a manifold gasket replacing.

That got me thinking and I later rang the garage back to cancel, having decided to do it myself. Armed with knowledge from all corners of the internet I got to work. First job, get some supplies from the kitchen, this could be a 4 cuppa job:

Then jack up the car ready to drain the coolant - its too low to get a bucket under otherwise.

I forgot how much I bloody hate draining the coolant, I always get it everywhere as the only way I can do it is to pull off the bottom rad hose as I my radiator doesn't have a drain plug. Anyway, that done, on with the job - disconnect coolant hoses from the inlet manifold, namely the thermostat and heater connection.

Next, all the connections from the carb. Fuel line, vacuum line, throttle return spring bracket, throttle cable.

Labels on all ignition leads

Note position of rotor

Position of distributor housing, complete with marks from previous distributor removal just to confuse things.

I removed the intake bolts and checked that they are all the same length so didn't have to note where they all came from.

Now remove the intake, the lovely sound of sealant coming unstuck. I had to get a long screwdriver in the front end and gently pry it apart but once the seal started breaking I could lever it up just by getting my hand in the carb inlet hole in the top. Quick look in the engine to see if anything has dropped into the valley, looks ok (paper towels stuffed everywhere immediately after taking this picture)

Here's the manfold with the old gasket stuck on. Its a rubber/metal type with lots of rtv all over. The end with the round hole (dizzy end) is where the leak was. There was no obvious break in the seal (assembled with RTV on the front and rear walls) but the sealant peeled off very easily and was saturated with oil on both sides so clearly the seal had been broken at some stage and once the oil has a small path through then its game over. 

To get to this stage had probably taken about 2 hours but that includes jacking the car up which takes a good 20 mins or so as I've only got one hydraulic jack and one scissor jack to get it on axle stands. I had also stopped for 2 cups of tea. The next stage is probably the most time consuming - cleaning everything.

I started off by going straight in with a stanley blade and most of the rubber gasket came off easily. The RTV bits at each end peeled off in one go but left a very thin film behind that even a blade couldn't remove and I didn't want to get any deep scratches in the machined surface of the manifold. I gave everything a good soaking in brake cleaner and found some fine wet & dry, 360 grit and 1200 grit. I went at it with the 360 and finished with the 1200 and it came up looking nice. One final wipe with brake cleaner on a cloth and half an hour or so later had nice clean gasket surfaces. I stuffed paper towels in all the runners to stop debris getting in then got a hoover with brush attachment to pick up the little flecks of silicone that were stuck to the casting.

The heads on the engine cleaned up nicely just with brake cleaner and a blade, no RTV had been used on the cylinder head side of the old gaskets so it was very easy to clean it up. The china walls front and back of the block did need a bit of 1200 paper just to get the thin film of sealant off. Again paper towels were stuffed everywhere and anywhere to stop dirt and debris spreading.

Got a nice picture of one of the inlet valves, I'd never seen any SBC valves before. I knew they were big but good heavens they are about the same diameter as my cup of tea!

Clean and tidy ready for reassembly.

The replacement gasket I bought was an Edelbrock 7201 which comes with exhaust crossover restrictors (not needed) and a replacement distributor gasket. The instructions recommend gasgacinch on the cylinder head side only, RTV around the water ports (extreme left and right holes on above picture) and RTV on the front and rear block china walls. So that's what I did. I ignored lots of the internet where people have done it differently "for years with no problems" and followed the instructions to the letter.

I painted gasgacinch on the gasket (cylinder head side only - the side without the black silicone bead embedded in the gasket) and on the cylinder head. Gasgacinch only holds the gasket in place for fitting, it doesn't do any sealing so there's no need to ladle it all over like gravy on your sunday roast. Just a thin smear then leave for 5 minutes before mating the gasket to the head.

Top tip here: loosen or even remove the rocker covers. Luckily I dry-fitted the gaskets to make sure everything lined up and the rocker cover gaskets stopped me getting the inlet gaskets in exactly the right place.

The RTV needs to cure for 15 minutes until you can touch it without it sticking to your fingers. Just enough time for a cup of tea and some biscuits.

Then confidently place the manifold down flat and level and don't move it too much once its down. Bolts in and torque tighten to 25lb.ft (34Nm) in two stages starting with the central bolts, then front, then rear. The centre ones are a pig to get to with my torque wrench even with a universal joint on my 1/4 drive.

All that remains is to refit everything that was taken off before.

Tada!

Oh and refill the coolant. I know my cooling system takes exactly 9 litres to halfway up the header tank. On my Dax I have found the best way is to unbolt the header tank and balance it on top of the carb, nice and high to get some rho*g*h going on, then just fill, fill fill, occasionally squeezing the bottom hose (lowest point), you can hear the system burping air. I don't know if this is the same for other cobras as some are plumbed differently. There are no other bleed points in my system apart from the small take-off just after the thermostat housing, which is the highest point of the system.

Filling went ok on this occasion, previously I have started the car with the header cap off if I suspected some air was still inside and let it burp the air locks out that way. Another good way I've found is to pull the heater return hose off which gives a good route for escaping air.

When I eventually found the car keys I started it up, it was a bit reluctant but of course the fuel bowls were completely empty and the inlet tract dry as a bone so after two or three pumps on the throttle then hold the throttle halfway while cranking it finally caught. Through all the escapades with this car I've learned not to worry when smoke billows out the engine bay, this was just a bit of oil on the headers that I dripped on when removing the dizzy, and spilled some coolant too which got burned off. As soon as it idled on its own (no choke of course) which pleasingly was about 15 seconds holding it at 1500rpm, I hopped out and got a torch and had a good look at the front and rear block wall seal. I had cleaned it all up to make it easy to spot new leaks. I've seen a trick with flour to show up leaks but makes a mess so I just relied on my 550 lumen bike light. A careful inspection showed nothing so I went for a drive, 30 miles or so. Got it nice and hot, got the revs up and inspected again with the torch when I got back. No obvious leaks so it looks like its good! I'll keep an eye on it of course but the next job now is to clean up the sump and bellhousing which are covered in oil. I didn't see any point cleaning them until I knew the leak was gone.

Total time from opening the garage door with tools in hand to starting the car and rolling out = 6 hours. If I was to do it again I reckon 4 hours is just about possible.

Longest Drive Ever! (so far)

At the weekend we decided to embark on a 320 mile round trip over two days, 2 1/2 hours and 140 miles each way (plus some extra local mileage at our destination) which is by far the longest journey we've done in the cobra so far. It took about half an hour to settle down and stop tensing up and desperately trying to work out what every small sound was - e.g. driving over a dry twig which snapped so loudly we could hear it over the exhaust made me nearly poo myself that something had fallen off!


We waited and waited for it to stop raining and eventually set off with very wet road. Naomi was brave and drove the first bit, flooring it on a slip road to join traffic was interesting - "slip" road is definitely the right term! The fuel gauge had me worried too, It's never been completely full apart from the IVA and I had forgotten that it doesn't drop in a linear fashion. Full to 1/2 is only about 20 litres so that was gobbled up in the first 70 miles and I was getting quite worried but then the needle barely moved for the next 50 miles. Range on a full tank I reckon is about 240 miles but that's pushing it a bit unless you know exactly where the petrol stations are!

Got into the swing of things but 2 hours on the motorway at 70mph must be reaching anyone's limit of fun in a cobra. We were togged up with ear plugs, hats gloves etc. We swapped drivers halfway but I was definitely ready to jump out and have a cup of tea by the end. Sunday morning was fun though, visited a number of friends and family and gave them all a go in it. I was especially pleased to go for a drive with my grandad who was an AA man for about 40 years and has been in the motor trade all his life. For him to say "not bad" was practically the highlight of my life so far. He nodded in approval as we thumped through the villages in 5th gear 30mph, 900rpm, then did the classic muscle car trick of 0 to a nice round number entirely in top gear.Rest of the journey home was ok. Slow traffic on the A1 southbound in the evening and we were crawling at walking pace for about half an hour, car was fine, fan working well and was actually quite pleasant with the heat wafting up from the footwells to combat the cold as it got dark. MPG for whole weekend was about 17, I reckon I can get a little more out, maybe 18 or 19 but not that bad really considering almost 40 miles or so were "demonstration" miles. The whole journey used about 150-200ml of oil but I know I've got a small leak from just behind the dizzy which is on my winter list to sort out so wasn't really concerned by that, everything is nicely rust proofed by oil now!
Just got to clean it all up and do a few cursory checks: header bolts need snugging down again as I could hear it chattering towards the end of the trip and I suspect the tracking is slightly off after catching a massive pothole.

HEI Distributor Maximum Advance Limit Mod

Getting ready for a very brave 300 mile round trip to see some friends this afternoon so went for a little shakedown drive this morning. Had a final check of the timing after doing a slight modification to the distributor.

There is a method for limiting the mechanical advance which includes welding up the slots in the dizzy plate but this sounded a bit dodgy but I happened upon another method which I prefer as its completely reversible, and I did it without even removing the distributor from the car.

In the metal strip that the rotor screws on are a some holes punched at the factory. By chance they are just over 5mm in diameter so can easily be tapped to take a homemade M6 grub screw. All I did to make a grub screw was take a regular M6 hex head machine screw and chop the top off. You can see in the photo below all you have to do is leave enough of the grub screw protruding so that when the rotor moves, the screw will hit up against the centre plate - thus limiting the mechanical advance. Lots of people have experimented with grinding the head of the screw down to various shapes and sizes to limit the travel so I did the same, then verified with the timing light.

I got the dremel out and made a slot in to get a flat head screwdriver. You can then do further grinding to the head of the grub screw to get your adjustability. I cut half the top off, retaining some of the original grooved cut so you can still get a screwdriver on it. Rotating the grub screw then gets the flat edge of the screw either closer or further from the baseplate depending on its position.

You can twist the plate that the rotor fits to by hand to see how the sweep of travel is affected and then by rotating the grub screw its easy to see how the position of the screw can alter the maximum rotation of the rotor.

It's probably best just to experiment and have a guess, then check the timing and make further adjustments. I found it quite satisfying to do it this way actually as it proved to me that I knew what I was doing!

The angle of the grub screw required for a certain amount of total advance may change depending what shape weights and centre plate you have because the starting point of the sweep of rotation of the rotor may be different. Best to experiment yourself. As long as you haven't  changed the base timing position you can't hurt your engine if you start it up and run the revs up to 3000 to check the "all-in" figure, there's no load on the engine so it's no problem.

The inside of the plastic rotor housing needed a tiny bit of fettling to clear the screw but all was well. I'm now confident that the max timing is definitely the maximum and there is no extra bonus timing. We briefly ran the car up to 4000, 4500 and 5000rpm and the timing mark on the balancer was steady so I'm happy that's the true maximum.

We went for a short drive to get some fuel and to do the timing in the quiet industrial estate behind our house so as not to annoy the neighbours and it was fine.

We checked the timing: roughly 16-17 static, 30-31 all in at about 2700rpm so thats 15ish of mech advance. A quick adjustment on the grub screw and my mechanical advance was opened up a tiny bit to 16°. The final setting was 18° static and 34° all-in. This allowed me to close the idle speed screw a bit and I've now locked the dizzy down and will probably leave it there.

One important thing to note - if you limit the total advance so you can raise your static advance, you may need to swap to stronger springs because your "all-in" figure will come in sooner.

The Moroso advance kit gives the following information:

RPM	500	1000	1500	2000	2500	3000	3500	4000	4500	5000
light	0°	10°	20°	22.5°	23°	23°	23°	23°	23°	23°
medium	0°	0°	6°	13°	19°	23°	23°	23°	23°	23°
heavy	0°	0°	0°	6°	10°	16°	21°	23°	23°	23°

The values in the table above are added to your static timing to get the total timing for each rpm interval.

An example: say you originally had 12° initial and a a pair of medium springs. By the chart above, you would get a total of 12° + 23° = 35° advance in at 3000rpm.  At 2000rpm you would have 12° + 6° = 18° mechanical advance. If you whack your static up to 18° at idle, at 2000rpm now you have 18° + 13° = 31° mechanical advance. If you already had an aggressive timing curve with light springs, you may suddenly find you have some pinging issues when accelerating from low rpm as the mechanical timing is now higher than it was before.

What I did was to check the timing at 500rpm intervals and swapped the springs for heavier ones to bring the timing curve back to what it should be. Surprisingly I ended up using two heavy springs. The top row is my original timing curve, without the mech limit grub screw, static set to 12° and two medium springs. The bottom row is my new timing curve with 18° initial, total mech advance limited to 16° and two heavy springs.

Total timing:

RPM	500	1000	1500	2000	2500	3000	3500	4000	4500	5000
Original	12°	12°	18°	25°	31°	34°	34°	34°	34°	34°
heavy/heavy	18°	18°	18°	24°	28°	34°	34°	34°	34°	34°

I felt that this was the best match. I tried one medium spring and one heavy spring but the advance rate was too fast and was giving me 34° timing at about 2400rpm and I was getting some light pinging when accelerating from lower revs.

Went for a drive and it felt good. Didn't notice the loss of a couple of degrees of advance in the 2500-3000rpm range. Perhaps on a dyno there might be a small difference, or maybe in a really heavy car you might possibly detect a small change but in the real world with this engine in such a light car you just can't tell. Idle screw settings don't seem to affect the cruise over 2000rpm so that suggests the idle and transition circuit is working how it should (closing down the idle speed means the transition slot is being used properly as designed) but I still get a little lean surge below 2000rpm so I wonder if the idle/transition circuit can supply enough fuel. It's on the list for things to research. EDIT - see post Idle Channel Restrictor for the solution to this! Over 2000rpm AFR's of 12.8 to 13.0:1 for cruise does seem a little rich to me but anything more than that and the lean surging gets worse so I've decided its best to give the engine what it wants, not just what the internet says it should have (mid 13's)