Description: Originally the MGB used the same breathing method as the MGA. This was a very basic non-positive system consisting of one hose between the top of the rocker cover and the front air cleaner, and another open-ended hose hanging down from the timing chain cover (often called the 'road draught tube'). There will be very slight suction on the rocker cover hose from the air cleaner, varying with throttle opening, and when under way there may also be slight suction on the open end of the timing cover hose from the effect of the air passing its open end. The effect from either is minimal, which way any air would flow is a matter of conjecture, and if there is more suction on the rocker cover hose it will be pulling unfiltered and potentially moisture-laden air into the crankcase from the timing cover hose - not ideal. At least any fumes pulled out of the crankcase in that direction would be burnt in the engine, if the flow is the other way they are just pumped out into the atmosphere.

In February 1964 with engine 18GA a positive crankcase ventilation (PCV) system was introduced. A PCV valve was fitted to the inlet manifold, and the purpose of this valve is to provide a source of a continuous low-level suction under varying engine operating conditions. The port on this valve is connected to a port on the front tappet chest cover, which contains a wire gauze mesh that acts as an oil-trap as well as a flame-trap. The oil filler cap was changed to the vented type, which lets fresh air into the crankcase via a wire gauze filter and a restriction. The purpose of the filter is obvious, the restriction further limits the flow of air through the engine and PCV valve into the inlet manifold to a low level. This has two purposes - one is to avoid excessive weakening of the mixture, and the other is to provide a slight negative pressure in the crankcase to ensure that under normal conditions i.e. a sound engine, fumes aren't lost to atmosphere from any other place like oil seals and gaskets, but are burnt in the engine. The PCV valve consists of a sprung diaphragm and valve. With the engine stopped the diaphragm is pulled back and the valve is fully open. With the engine running air flow causes a depression in the crankcase and under the diaphragm, which with atmospheric pressure on top or the diaphragm, tends to push the diaphragm down which closes the valve. This reduces the flow, which reduces the depression below the diaphragm and inside the crankcase, which tends to allow the diaphragm to move up again, which opens the valve to give more flow and a greater vacuum and so on. In practice the sprung diaphragm continually balances crankcase pressure with atmospheric pressure to result in the relatively constant flow rate through the valve, and hence the engine, with varying inlet manifold depressions. Clausager refers to this as a 'closed circuit' system but it isn't, it is a 'through-flow' system.

In October 1968 with the 18GG engine the system was changed again, replacing the PCV valve with a vacuum source taken from the twin SU carbs. The SU carbs are referred to as 'constant depression' carbs and this refers to the area between the butterfly and the piston. On a running engine as the butterfly is opened the depression in this area tends to increase, but as there is are passages between that area and the top of the piston the higher vacuum appears there also, which tends to lift the piston, which has the effect of lowering the vacuum again between the piston and the butterfly. In practice of course the piston moves up and down with the butterfly opening and closing, and the vacuum between the two remains at a relatively constant low level (see SU Carbs for more information on how the SU carbs work). Each carb has a port that taps into this area, and so provides the same sort of signal as the PCV valve, but with no moving parts (or any parts to fail other than possibly a blocked port) and even more importantly to the manufacturer at no cost! These ports are connected via a Y-pipe to a hose that goes to the front tappet chest cover as before. Both carbs are ported to retain the air-flow and mixture balance between them, and the later HIF carbs are the same as the earlier HS. Again the oil filler cap is of the vented type.

For UK cars the system remained like that until the end of production, but for North American cars the system changed in October 1969 with the 18GJ engine when the 'evaporative loss control system' was introduced (see North American Emissions Plumbing for more information on this system). The main additional component of this system is the large charcoal canister sitting in the right rear corner of the engine compartment with three ports on the top and one on the bottom. Two of the three ports on the top are connected to the fuel tank and carb float chamber vent ports, and any fumes from expansion of fuel or filling of tank or float chamber are pushed into the canister, adsorbed by the charcoal granules, and fume-free air is vented out of the port at the bottom of the canister. As the level in the fuel tank drops while driving air travels the other way through the canister to replace it. The third port on top of the canister is the important one as far as crankcase ventilation is concerned. On these engines the oil filler cap is replaced with a non-vented type, and a port with a restriction is provide on the back of the rocker cover, this port is connected to the third port on top of the canister. The restriction in the rocker cover port provides the same function as before i.e. preventing excessive weakening and to ensure a small negative pressure in the crankcase. Carb vacuum pulls fresh air through the port at the bottom of the canister, through the granules which purges them of any adsorbed fumes as well as filtering particles out of the drawn-in air, through the engine picking up any oil fumes, any fumes from either source being burned in the engine. When North American spec engines changed from twin SUs to the single Zenith/Stomberg in December 1974 for the 1975 model year the situation remained the same, this carb is also a 'constant depression' type the same as the SUs and has the same breather port, although in this case there is only the one and no Y-piece, of course.

V8 engines are slightly different. Each carb has its own hose, with an oil/flame trap, going to a port on its respective rocker cover. On the back of the block near the right-hand side there is a metal pipe pointing upwards, on which is a short length of slightly kinked hose, on top of that a petrol filter held in a clip on the back of the air-cleaner box, and a U-shaped hose on top of the filter (to prevent debris dropping into the filter). In this case the airflow is in the opposite direction to 4-cylinder cars i.e. fresh air goes in via the filter to the crankcase, then up through the engine into the rocker covers, and from there into the carbs and engine. The filter is obvious, I'm not sure where the restriction is - possibly in the pipe on the back of the engine, but there definitely is a restriction as the engine note changes when I remove the oil filler cap (see below).

Problems: The original system has enough problems to begin with, drawing unfiltered and wet air in through the timing cover port, and being very haphazard as to whether crankcase fumes are burned in the engine or pumped straight out to the atmosphere. Apart from that all that can happen is either or both hoses get blocked. With either hose - and this is the same for any of the three ventilation systems - a blockage in one hose will prevent any ventilation. The main effect of this is to allow condensation to build up inside the engine, especially in cold conditions or where the engine is only used for short journeys, which will cause corrosion. This is usually visible as a creamy 'mayonnaise' in the oil filler hole and on the bottom of the cap. If both ports get blocked then there is no path for the relief of excess crankcase pressure, which can blow seals and gaskets, however this is more likely to occur on older engines with some blow-by. Note that contrary to often expressed opinion the blocking of one port, whilst it will stop through-flow ventilation, won't allow crankcase pressurisation to occur, as the other, still open, port will relieve that, whether it be via the PCV valve, carb ports, ventilated oil filler cap or charcoal canister. Blockage of one or both of the hoses is also about the only thing that can happen to the later carb ventilation systems. In theory North American spec cars with the canister could get a blockage in that or its fresh-air hose, but in practice this is likely to cause running problems (overflowing carbs and tank vacuum) before it is noticed elsewhere. On positive systems (PCV valve and carb ventilation) the suction-side hose can be checked very easily, as removing the oil filler cap should result in a weakening of the mixture and a slight increase in idle speed as effectively you have created a vacuum leak. If you put the palm of your hand, or a sheet of paper over the oil filler hole, it should be sucked onto the hole with slight pressure. A blockage in hose between the rocker cover and the charcoal canister is more difficult to detect, removing it from the canister will show very little vacuum, although it should pull smoke through i.e. from a cigarette or other smoke source. A blocked ventilated oil filler cap is even more difficult to detect, but these are probably best replaced at 12k intervals anyway. The one that came with the roadster always tended to leak oil past the seal and down the side of the rocker cover, replacing it cured that. With V8s if one carb/rocker cover hose or flame/oil trap gets blocked the crankcase will still get ventilated via the other, but only the one rocker cover.

The PCV valve has a finite life, when it fails it is usually the diaphragm that ruptures, the effect of which is to apply full inlet manifold vacuum to the crankcase, which can pull significant amounts of oil into the combustion chambers fouling the plugs as we as resulting in high oil consumption and oil smoke pollution. If you have a PCV valve and experience stalling when the cap is removed, or a large vacuum is felt, then the valve has probably failed. That is if you haven't already noticed high oil consumption. Other problems can be oil and combustion sludge inside the valve restricting the movement of the diaphragm or blocking the valve. After-market valves and those used on other vehicles often have a plunger instead of a diaphragm which removes the main failure mode of the MGB valve, but they can still stick open or closed and get gunged-up. The advantage of this type over the MGB type is that under crankcase pressurisation from excessive blow-by the valve will close to prevent air being forced into the inlet manifold from this source so weakening the mixture. If this type of valve were used on the MGB the excess pressure would be vented to atmosphere (via the oil filler cap or charcoal canister), but systems with this type of valve tend to have the fresh-air intake inside the air cleaners, so any fumes emitted will still be burned in the engine. This type of system is a closed-circuit system, unlike the MGB.

Modifications: These range from the simple, like removing the emissions kit from North American cars, to the more complex like replacing the SU or Zenith carbs with something else e.g. Weber. The first thing to say is that unlike the air-injection system (the removal of which isn't covered here) the 'vapour loss recovery system' (aka charcoal canister) has no detrimental effect on performance or economy, and does help to keep the atmosphere a little cleaner than it otherwise would be. The only reason for removing it is to free-up a little space in the engine compartment. And if you have a 73 model or later with the anti-runon valve, interfering with the canister and its plumbing disables the valve, which can be a positive disadvantage. If you do decide to go down that route there are a number of aspects which must be considered. The tank, float chamber and rocker cover vent pipes can be left dangling in that corner of the engine compartment. But if you remove the tank plumbing and separation chamber and seal its vent port you must fit a vented fuel filler cap in place of the standard unvented. Don't remove the pipework running to the front of the car but leave the separation chamber or vent port from the tank open in the boot or it will fill with fumes, and with the electrics in the boot particularly the sparking points of the rubber bumper fuel pump is an explosion hazard. If you remove the existing float chamber vent pipe that runs across the engine compartment you must fit alternatives that run down past the engine and exhaust for safety, neat petrol pouring onto a hot exhaust is not a good idea. With the charcoal canister removed you really ought to provide alternative filtration to prevent the crankcase breathing system pulling dust and moisture into the engine. The best way of doing this is to remove the hose from the rocker cover and fit a small filter to it instead. You could seal off that port and fit a vented oil filler cap instead, but subsequently someone may not realise and fit a non-vented cap again, which will disable the ventilation system resulting the aforementioned condensation and corrosion. Much better and more obvious to leave the cap as standard and fit the filter. If you do all that you might as well remove the anti-runon valve as well as it is no longer doing anything useful, and if you do that you must seal its port on the inlet manifold.

If you fit a fixed-jet carb like a Weber they do not have a PCV port as they have no source of a constant vacuum or air-flow so something else must be done. Some revert to the prehistoric non-positive system used on the first MGBs or just leave both rocker cover and front tappet chest cover ports open, but then you are back to condensation and dust in the engine and pumping out oil fumes. Much better to retro-fit a PCV valve and retain the positive ventilation. If use an after-market PCV valve or one intended for another application these are often smaller and neater than the MGB valve as well as being more robust not having the diaphragm and protect against mixture weakening from crankcase pressurisation. Whichever, plumb that to the rocker cover rear port. Weber air-cleaners often seem to have a breather port, in which case so you can connect the tappet chest cover port to the air-cleaner port. Then you will have a positive, closed-circuit system that protects the mixture against crankcase pressurisation and the environment against fumes.

The V8 engine is installed on the chassis rail brackets using three components each side - the rubber mounts as one might expect, a spacer, but also a plate that fits between the mount and the block. It is vital to be aware that these plates are handed but can be installed on either side. If installed on the wrong side the engine and gearbox assembly is about 1/2" forward of where it should be, which means the crankshaft pulley can rub on the anti-roll bar under heavy braking; the sump can rest on the front cross-member and wear through; and the gearbox cross-member can only be installed by bolting it in the forward position on one side and the rearward position on the other.

If you have occasion to refit these plates, look carefully at the relationship between the holes that are used to bolt it up to the block and the holes that takes the mounts. They should be installed such that the holes for the mount are in front of the centre-line between the two holes that secure the plate to the block. This will ensure that the engine is installed in the rearmost of the two possible positions - the correct position.

When replacing the mounts I found it easiest to undo the nuts securing them to the chassis brackets then jack the engine up so the studs on the mounts cleared the chassis brackets (you will have to tilt the engine to remove first one then the other) then remove the mounts from the engine plates. The rubber mounts have the chassis plate stud offset from the centre when attaching the rubber mount to the block brackets the stud must go in the lower of the two possible positions.

Because the mounts are angled it is not possible to drop the engine with rubber mounts attached straight onto the chassis plates even though the chassis brackets are slotted. Tilt the engine to get one stud in then tilt it the other way so that stud is at the top of its slot and you should be able to get the other stud in. Before the stud goes all the way through fit the lock-washer and start the nut. When you lower the engine all the way aim to get the studs at the same position in their slots as each other to ensure the engine is correctly aligned.

The drivers side is tricky because access is severely limited by the steering shaft passing through the chassis bracket. I wedged the nut into an open-ended spanner and stuck the lock-washer to the nut with grease then offered the nut/washer up to the slot in the chassis bracket, then screwed the mount (with spacer) into the nut a few turns, then secured the mount to the engine plate.

Remember to reinstall any spacers on the appropriate side of the rubber mount. Many cars will have had a second spacer fitted on the drivers-side to stop the exhaust manifold hitting the steering shaft as the mounts age and compress. If you find you need to add one it is easier fitted between rubber mount and chassis bracket, the dimensions are given in the pictures accompanying this section. If you cut a slot in the mount instead of a hole for the rubber mount stud you will be able to slacken the nut, raise the engine slightly that side and slide in the spacer, rather than having to undo the nut completely and raise the engine enough for the stud to clear the chassis bracket.

Bolting up the stud nuts is a long, slow job as you have to turn the open-ended spanner over twice for each flat, so be patient.

Update May 2007: I had to undo these to deal with the exhaust manifolds yet again, and was disappointed to find one of the down-pipe to Y-piece clamps wouldn't undo. I could turn the nut back and fore on the bolt a little way, but as soon as I applied any more force the square end of the bolt turned in the clamp. Because the bolt head is so low-profile I couldn't get enough purchase with grips, so had to grind it off without damaging the clamp. That was OK, and I did have another bolt and nut, and why the other one came undone just fine but this didn't I'll not know now, as the bolt end and nut were destroyed by grinding them off, of course. Maybe I should consider myself fortunate the down-pipe to manifold clamps came undone just fine, and the stuck one was so accessible.

Bearing down against spring pressure whilst trying to get the threads engaged is one of those 'worst jobs' on the MGB. Refit the cap without the spring, slowly unscrew it, and mark the cap and block where the threads just disengage. Now you can refit the spring and position the cap just before the threads will engage and not only will you know where the threads will engage, but also that whilst fighting the spring you will only have to turn the cap a few degrees to do it.

RPM Torque (Lb.Ft.) BHP Fuel Consumption (Pints BHP/hr) 1000 71   0.95 1500 89 26 0.78 2000 101 38 0.69 2500 107 51 0.63 3000 110 63 0.58 3500 108 73 0.57 4000 106 81 0.575 4500 104 88 0.59 5000 98 94 0.62 5500 91 95 0.65 6000   92 0.73

See also this table of North American MGB and Midget performance figures from Skye P Nott's 'MGB Experience'. Note that the reduction in MGB power and torque from 1973 onwards only applied to North American spec cars, UK cars retained full performance to the end.

The later rocker cover nuts, as well as having the stud on top of the nut for the heater return pipe, has a deeper cylindrical portion below the hexagonal section. This passes through a 1/8" spacer, cup washer, rubber bush and rocker cover onto the long cylinder head studs used to secure the rocker gear. The rubber bush performs two functions - it seals the fixing holes in the cover to prevent leaks, and also applies pressure to the cover and from there to the cover gasket and head. Old rubber washers compress and harden with age and so reduce the pressure applied to the cover gasket. Old cork gaskets similarly compress, and can rarely be reused successfully once disturbed, even if it was oil-tight before.

On my engine once the cover nut has taken up all the free play between it and the top of the spacer, it can only be tightened one more turn before the bottom of the cover nut contacts the top of the rocker gear nut. This is with new rubber washers and new cork gaskets. Even gluing the cork gasket into the cover, and using Hermetite Red as the seal to the head, I still get weeping from the rear of the cover. Although this results in only an occasional drop of oil on the floor I still don't like it, and I have had to put 2 or 3 thicknesses of gasket card between the cup washers and rubber washers to get a seal.

In discussions some have avowed that this is because it has been overtightened in the past and buckled the cover, the cover nuts should never contact the rocker gear nuts. But my cover is perfectly square and symmetrical. One poster stated that, with the exception of the oil filler, you should be able to lay a straight-edge across the top of the cover and have no gaps. Now straight away this is incorrect as the cup and rubber washers sit in recesses, that extend as grooves down the side of the cover. I can't for the life of me see why the cylindrical portion of the cover nut should be made so much deeper than the cup and rubber washers, unless it was designed to bolt down to the rocker gear nut and so prevent overtightening. I've no experience of the 3-bearing and 18GD engines but they have a different rocker cover nut which doesn't seem so deep and with those I can imagine they could be tightened so much they distort the cover. I also don't see how on the one hand overtightening the nuts overcomes the natural springiness of the pressed steel cover enough to cause distortion, but in my case at least adding more packing to apply more pressure stops the leaks. The only way the cover could have been distorted so much that there is now only one turn of the nut available before it 'bottoms' is if a lot of extra packing i.e. several thick washers had been fitted in the past. But there weren't any when I got the car, it didn't start leaking until I replaced the rubber washers and cork gasket, and the old ones weren't bonded. But then I only have a sample of one.

The upshot is that for the moment I'm leaving things as they are. I'm not going to try and 'uncrush' the cover as that almost certainly will distort the cover beyond sealing, and at the moment nipping the nuts down as far as they will go gives me a decent seal without any further risk of overtightening and distortion. I'd still be interested to see how much the stud does or doesn't protrude on other engines, though, and get the measurement from the face of the flange the cork gasket sits against to the recess the rubber washer sits in.

Both the Leyland Workshop Manual and Haynes show and describe, prior to fitting, the gear set inserted into the chain with the dimple in each gear being opposite (Leyland) or adjacent (Haynes) and in line with the centre of each gear as shown on the left (click to enlarge) - this is correct.

Both also show the keyway in the crankshaft being at the top, which coincides with TDC for pistons 1 and 4 - and the camshaft keyway at approximately 1 o'clock and this is also correct.

However this places the cam and valves in such a position that it is No.4 piston that is at TDC on its compression stroke, and not No.1. The problem arises when it's time to insert the distributor drive dog. Both manuals say that the engine must have No.1 piston at TDC on its compression stroke, which is also correct. But the Leyland Manual goes on to say the dimples have to be in line, leaving out the word 'opposite' but omitting to mention that the cam gear now has to have its dimple at top-right i.e. remote from the crank gear dimple i.e. the crank has to be turned through 360 degrees, and this is where the confusion arises. However Haynes still specifies that the dimples should be adjacent as well as in line and this is the bit that is incorrect.

I suspect that whoever originally did the drawing and spec for the cam gear got the dimple 180 degrees out and it was never corrected, there seems no justification whatever for fitting the gears with No.4 piston at TDC on its compression stroke. See also John Twist's advice on correcting the distributor drive gear position.

Another problem concerns the gaskets - I have tried three different types so far. Originally they were single, thick, metal-faced sandwich gaskets, which have quite a good ability to cope with a small amount of the flanges not being flush with the head. The next were thinner, green and black composition and were useless. Not only didn't they compress much, but with the very small overlaps between flange and head blew a piece out on the first decent run. The third type (picture, click to expand) come in pairs i.e. one gasket covers two exhaust ports and whilst they are a metal sandwich again they are quite thin.

All have similar sized holes, which are up to 3/16" bigger on each edge than the head ports (picture). I suppose there is an element of not covering up some of the port in this oversize, and also perhaps variations from head to head. But if the holes were quite a bit smaller it would significantly increase the amount of overlap which would reduce the chance of them blowing.

With the new manifolds I decided from the outset to weld struts between all four flanges (picture) so they couldn't turn in to each other. These struts are placed over the link between the two halves of the paired gaskets i.e. in the lower half of the flanges, so as not to obstruct the plugs or dipstick tube. However I also discovered that whilst the faces of the flanges aren't cocked at an angle to the head, the outer two on both my manifolds are further away from the head than the inner two. I enquired about getting them machined, but two engine machinists I spoke to said they cannot hold them securely enough to run a grinder over them like one would when skimming a head or block. I did separate one of the old single gaskets to add to the new double gaskets to give some extra thickness on the rear port of the left-hand manifold but obviously it wasn't enough, as it started ticking slightly on acceleration quite soon after fitting. This year it suddenly got noticeably worse, and so is at risk of failing the MOT.

A few minutes saw the left-hand manifold come off. The good news is that the struts seem to have done their job as all the bolts went back in OK. However with them all in the manifold was 'hanging up' slightly so I did file one hole out a little so the manifold slid in and out easily when all eight bolts were half screwed in. With the outer ports further away from the head than the inner two it was obvious that these had been blowing from the staining on those gaskets, whereas the inner two are fine. So I guess this type of gasket is OK given correct alignment and spacing of flange to head. From the staining I could see that the two outers had been blowing towards the inner ports, so obviously when tightening down these outer ports, because they have further to go than the inners, they turn in slightly, the very thing I'm trying to avoid with the struts. This means that the gasket isn't clamped as tight on the inner edge as the outer, and the inner edge blows.

Another problem is that the alignment of the manifold ports to the head ports is very poor (picture). Clive tells me this wasn't discovered until he had some one-piece flanges made for RV8 manifolds and in an idle moment held these up to the block-hugger manifolds. He was shocked to discover that although the bolt holes lined up the ports didn't, by 3/16" or more in some cases. Even though the holes were oversize so some misalignment would mean the ports weren't partially blocked, the amount of misalignment is so great that the head ports are obstructed to some degree. At that time he was having the flanges stamped out by one metal-basher in Dudley, and the pipes formed and welded by someone else. When he queried this with the stamper they admitted that the flanges weren't made especially for the Rover V8 but were for another application and seemed close enough!. This was some time ago, Clive now has the flanges laser-cut by someone else and they are supposed to be a much better alignment, but at £400 for a new set I'm going to have one last go at getting a good seal on these. Incidentally, someone wrote to me recently saying they had no problem with a set purchased from the MGOC, but Clive supplies the MGOC anyway, they simply charge even more for them.

Changing the gaskets could probably be done by leaving the manifolds in-situ and still connected to the down-pipes and remainder of the exhaust, in which case you could probably reckon on less than an hour for each side. But I wanted to check the surfaces of the flanges, so removal was the order of the day. Even so the left-hand manifold came off in about 20 mins. I can use one of several spanners or sockets (9/16" or 14mm) on most of the bolts but the two inner lower bolts need a specially ground-down spanner as access is restricted, and the lower rear needs a long-reach 3/8" drive socket, or at a pinch a standard socket with the end of the wrench only just slotted-in, not fully seated. The long-reach is fine for the left-hand manifold but on the right-hand the rack shaft is still in the way. I've seen sets of Allen bolts for the V8 manifold and one would probably be useful in the lower rear of the right-hand manifold, and as a replacement for one of the two lower centre bolts (the left in this picture), but the other one is almost completely covered by the end pipe, indeed the bolt has to be fiddled into the hole and started into the head with the manifold clear of the head. If you leave it until the manifold is tight up against the head you can't get it in.

I used a flat-faced whet-stone to run over the faces until I got a shiny ring all the way round, which probably took about an hour,With a straight-edge across all four flanges I could see the faces were still flat to the head, but the rear port was about 1mm back from the others (picture) and the front port about half that. The new gaskets are the same shape and size as the old two-port ones, but slightly thicker, even where they haven't been compressed by the flanges. I decided to use the old ones from the two inner ports - which hadn't been blowing - as extras on the outer two. I'd removed the down-pipe by this time as I wanted to bolt the manifold up to the head without anything getting in the way of it being fully flush, so refitted the manifolds and gaskets, and with the other down-pipe to Y-piece, middle and rear clamps on the exhaust loose refitted of the left-hand down-pipe. Tightened everything up, started up - and still had a tractor in the garage! I had assumed that the left-hand gaskets which had been blowing slightly for some time had suddenly got worse - but no, it was the right-hand manifold!

So, nothing else for it but to pull the rack to enable complete removal of the right-hand manifold, as I wanted to check its faces as well. A couple of hours more work to flat the flanges, check the gaps, and reassemble with old but sound gaskets on the front and rear ports plus new two-port gaskets. More scrawling around underneath to reattach the down-pipes to the manifolds and Y-piece, start her up, and everything was fine :o) Another hour or so to refit the rack, wheels, and the middle and rear clamps and a successful test-drive. Not too exuberant yet as I have the MOT in a couple of days and I'd rather get that out of the way (she passed) before risking blowing them again. I must remember to check the tightness of the bolts at least annually, I was surprised how loose they were when I came to take them off, which may have contributed to the blowing.

To my surprise I was able to replace the starter without removing either the tubular manifold or down pipe on the right-hand side, but I did have to remove the rack in order to get sufficient movement with a spanner on the top nut. Subsequently (I tried an alternative starter for a while but went back to the OE item) I used a pair of 3/8" extensions and a universal joint to get to the top bolt between two of the pipes on the manifold, meaning I didn't even have to remove the rack. The situation with the original cast-iron manifolds may be different.

The alternative I mention was one of the 'gear reduction' starters beginning to crop up all over the place. They are much smaller and lighter, in fact the solenoid is bigger than the motor, and bigger than the on the original starter so should be more robust, it being the solenoid that usually fails on V8 starters. The first time I tried it I thought it was just spinning and not turning the engine over as there was no rocking of the engine and no grinding, just a steady hum, but then it fired up. Because of the gear reduction the motor has a lot more torque, hence the smaller size, and spins faster and so take a lot less current which should take a lot less out of the battery at each start. However the connections were in a different place meaning I had to connect the cables before I could fit it which meant lying on my back under the car holding the motor up in the air with one hand, while I attached the cables with the other. There is also no boost contact for the coil on rubber bumper cars and all V8s on the ones I have seen. This last could be simulated with a relay, but with the lower drain on the battery it may not need it. They are about 50% dearer in price though. The problem with mine was that the motor assembly was attached to the adapted plate with just a couple of self-tappers, and needless to say these came loose after just a few days. On another occasion, and with beefier mountings, I could well be tempted to fit one. Be aware that there are after-market starters available described as 'hi-torque'. Not all of these are geared, Caveat Emptor.

Dismantling ...

... went fairly well, it's a good job it was the GT and not the roadster - plenty of room to lay all the engine parts out in the back out of the way leaving the bench clear for whatever I was working on at the time. The block has a water tap on the left near the rear and a bolt at the front on the right and unless you can open/remove both these (the bolt only after removal of the right-hand exhaust manifold) you cannot empty the block so removing the heads allows water to pour into the cylinders. I could undo the tap but the bolt was no-go, it was well rounded so it looks like a previous attempt had failed. I was amazed at the internal condition of the engine. It was a light golden colour with no sludge anywhere equating to 'low mileage' according to the RPI web site. I reckon it had done 80-100k miles since the last time any work was done, 65k of that in my ownership. I always change the oil and filter every 3k, using a pre-change flush treatment as often as not, looks like it pays dividends. I could still see the honing marks over most of the bore surfaces indicating bores/rings were unlikely to be the cause of the low compression readings. 20 thou oversize pistons fitted so obviously a major rebuild sometime in the past. When not actually working on the block I kept it covered with a clean cloth. I poured a little petrol into each inlet and exhaust port on the heads and only one inlet showed any seepage, the others all being 'gas-tight', again not a cause of low compression readings. The heads looked to be level and flat with just a trace of gasket burning on the edge that faces into the combustion chamber. The rockers and shafts showed no wear ridges at all, just polish marks. Three of the inlet valves had wear ridges and discolouration where the rocker contacts them so I replaced these as a precaution. The hydraulic tappets came out swimming in oil and just showed circular polishing marks on the base indicating they had been rotating properly, negligible indentation. The camshaft looked fine although it is difficult to judge by eye. The timing chain had quite a bit of slack. The oil pump pressure relief valve spring was 1/8" shorter than spec, the clearances of the gears and pocket being in spec. So my order went in to Clive Wheatley for a top gasket set, front cover gasket and rope seal for the crank pulley, oil pump gasket, three inlet valves, timing chain and gears, tappets, camshaft and oil pressure relief spring. Also a set of rocker cover screws as mine came with some incorrect ones, a head bolt as one of mine was an odd-ball with a very deep head with a tapped hole in the middle, and two inlet manifold bolts - I had broken one and the other front one was not as long as it should be, the casting being deeper for those than the rest. While awaiting the parts from Clive I had ample opportunity to deal with the inlet and exhaust manifolds and clean and where appropriate repaint all the parts I had removed.

Heads and valves

Replaced the three inlet valves and removed and lapped all the others (used the stick with rubber suckers on each end had to glue the suckers onto the stick). All the seats were in very good condition and they and the new valves needed very little work to grind them in. Checked them by pouring a little petrol into each port and none showed any signs of leakage. Made sure I could screw all the head bolts right in by hand then put the heads on using a coating of Wellseal first on the block then on top of the gasket and torqued them down. I used the original type of tin shim gaskets, there are composite types available but these reduce the compression ratio due to the greater thickness.

All Rover V8 engines have a head bolt 'at each corner' of each cylinder which gives nice balanced pressure to the gasket. However those made before 1996 have an additional row of four bolts on the outside of the Vee - 11 to 14 in the tightening sequence - that are said to cause a problem. Because they are putting additional loads on one side of the head it is claimed they cause warpage of the head and leakage of the gasket. Whilst an unbalanced force is not desirable I'm pretty sure that a correctly assembled engine won't suffer from the problems described in this article unless perhaps it has been modified for much greater power outputs, if there were any major problem with gasket leakage and oil breakdown in the tens if not hundreds of thousands of units produced it would be well known. After 1996 the engines were produced without those bolts, also demonstrating by now that they aren't required. Probably not a good idea to leave these bolts out altogether and the holes open in earlier engines, so just torque them up to about 25 ft lb, with thread lock as per all the others, to stop them coming loose.

Camshaft and timing gear

The camshaft is surprisingly heavy and the book warns against damaging the block bearings as they are not replaceable. I put its gear back on to give me more purchase, got it most of the way out then it started to foul the fans and oil cooler hose, but by that time I was able to angle it to one side and got it the rest of the way out. Being a rubber-bumper with the underslung cooler I could leave that in-situ, unlike the chrome bumper. When fitting the new shaft I didn't put the gear wheel on first but rested the shaft on the block bearings just before the last section was due to go in to give me a breather. The new one went in easier than the old one came out. Put on the new timing chain and gears, this was more difficult than removal because of the lack of slack in the new items, it is a matter of sliding each gear onto its shaft a fraction at a time, keeping one directly above the other as much as possible. Put the distributor/oil pump drive gear and spacer back on the camshaft, the big washer and the fixing nut. Put the oil thrower on the crank, this time concave side outwards (it was inwards on removal).

Front cover and oil pump

I could only get the one oil pump hose undone from its adapter, with the other (Sod's Law dictated that it was the long one to the cooler) the adapter came out of the body which meant I had to remove the oil pump cover and unscrew that from the hose rather than the other way round. The front cover was a bit tricky - although most of the fittings were long bolts there was a short stud and a long stud on the right-hand side, above and below the oil pump. These had become well stuck to the cover so a bit of judicious levering was called for. Even a couple of the long studs were difficult to get out having partially seized in the cover. Make sure you have removed all the bolts and stud nuts, including the two up through the sump, before levering the cover, I heard of one chap with one fixing lost in the gunge and he wrecked his front cover trying to lever it off. BTW, loosening/tightening the crankshaft pulley nut and camshaft nut are easy if the engine is in-situ and the rear wheels are on the ground by putting the gearbox into 4th. It looked like a chisel had been used to undo the pulley nut at some time. You have to undo the anti-roll bar mounts from the front apron and push the bar down to get the pulley off the crank. Very little paint left on the cover, came off easily with a small axial wire brush, repainted it satin black. Put the new rope seal in its holder which I left in the cover, didn't seem any point in removing it only to have to secure it again. I was surprised how easily it went in, but there was about 1" left over. I thought about cutting it off by decided to try pushing the rope into the holder rather than just laying it in. The second attempt left me with just 1/2" left over, the third got the whole length in, so don't be tempted to cut it short, persevere and push it all in. Made sure I could fit the crank pulley now with the cover off, rather than leaving it until the cover was bolted up then discovering it won't fit. Checked all the front cover bolts and studs would screw in all the way by hand then fitted the front cover. A word here about gasket sealant: I bought some from Halfords in a dark blue squeezy plastic container shaped like a bellows as I thought it would be easy to use, but I found it skinned and started hardening far too quickly for my liking given the size and complexity of the front cover gasket - more later. Packed the oil pump cavity with Vaseline which was a bit awkward from underneath, attached the oil pump cover to the hose (remember the adapter came out off the pump instead of the hose off the adapter) and then to the front cover. Fit a new water pump (I thought the old one was failing some time ago but it turned out to be a squeaky fan belt, but as the pumping-up problem could be the pump sucking in air it seems as good a time as any to use the new one) and gasket - sealed with Hermetite Red.

Tappets push-rods and rocker assemblies

I fitted the tappets dry to check the pre-load. There has been discussion in various places about this recently, and a figure of 20-60 thou has been bandied about and is quoted on the RPI site. I was a bit taken aback to find mine were 110-120 on the right and 40-110 on the left. Spoke to Clive and the MGOC and their opinion is that unless the engine is highly modified there shouldn't be a problem as the whole purpose of hydraulic tappets is to cope with a wide tolerance in the rest of the valve gear. Spoke to someone at RPI, who despite what is written on their site said "you wouldn't want to go as low as 20 thou and 120 thou should be fine". So I left it at that. Put a drill on the oil pump (made a driving spindle out of an old box-spanner tommy bar and used a length of appropriately sized hose to fit snugly over that and the oil pump driving shaft to keep the two together). Started off very slowly with much slurping from the oil pump and in no time at all oil was pumping out of the front tappet sockets, I kept going until oil came out of all of them and from all the rockers. Immediately before fitting the inlet manifold I squirted oil liberally over all the camshaft lobes.

Distributor. When refitting I realised with horror that I had got the spacer and drive gear on the end of the camshaft the wrong way round! So off came the front cover again and I was able to leave it hanging on the oil pipes while I swapped the gear and spacer round. Fortunately the dodgy sealant I mention above hadn't stuck very well and I was able to scrape this off the crankcase, front cover and even the gasket. Not only did it dry very rapidly but it also went very hard and splintered off - not good for a sealant I would have thought. Reverted to good old Hermetite Red which stays soft for ages and doesn't fully harden at all (remember to do the top of the sump gasket as well), and back goes the cover. Follow the instructions in the book on getting the distributor in with the correct orientation, as unlike the 4-cylinder car where the distributor only engages with the drive in one position the V8 can fit in many positions. Now I could crank the engine (plugs out) and get oil pressure and oil flow from tappets and rockers.

Inlet manifold

Managed to snap off the left-front inlet manifold bolt, the shank of which was heavily corroded. I was able to drill and retap the head, but nearly gave myself a heart attack when laying a ruler across the inlet manifold bolts to find that the one I had just drilled and tapped was 1/4" or so out of line with the others. Before doing myself too much damage I then checked the other side, to find that it was the same i.e. they were supposed to be that way! The corroded bolt indicated a leak from the water passage immediately adjacent to that bolt, and is immediately above the only place where I had seen any coolant leakage - down the side of the front cover gasket. I had assumed it was the gasket that was leaking, but it was only occasional and had been doing it for much longer than I had had the problem. It could be the source of the air in my 'pumping up' problem. One of the carb adapter stud holes has been stripped for as long as I have had the car and was loose, also one of the fan switch screw holes which had had a larger screw of the wrong thread forced in. With difficulty I found someone locally who said he could helicoil the two stripped threads in the inlet manifold with UNC, I was not pleased to find he did them in metric. I also snapped off one of the heater valve adapter pipe bolts in the inlet manifold but again drilled and tapped this myself. I discovered that the thermostat bypass pipe that is internal to the manifold (not the heater return pipe bolted underneath) was choked with scale which I rodded out with a long masonry drill. I decided to paint the inlet manifold grey as originally (mine was heavily flaking black) but ordinary paint would have a very rough finish on the casting and I didn't want the hassle of sending it away to be powder coated. I have used something called PlastiKote on household items in the past and it gives a finish similar to powder coating, but obviously isn't as tough. It is good for 150 degrees C, and they do it in mid-grey, looks good freshly painted (and after a couple of hundred miles), time will tell if it lasts. Fitted a new core plug to the bottom of the inlet manifold while it was off as a precaution. Before refitting I checked I could screw the bolts all the way in to the heads by hand, cleaning out any oil, water and other detritus. The book says to put sealant round each water passage - four in all as although there aren't any rear passages on the manifold there are on the head and in the gasket so these must be sealed too - on both sides of the gasket. Because of the earlier problems with the blue Halfords sealant I compared it with another 'instant gasket' translucent blue sealant also from Halfords and a clear one from elsewhere. Squirted out an inch or so onto a piece of cardboard and left it in the sun. The clear stayed tacky and soft much longer than the other two so that was the one I used. Now the gasket is fairly stiff metal that starts off flat but has to be deformed into a curve when fitted, and I reckoned if I put sealant on then just pushed the gasket into place it would disturb the sealant before everything was in place. I could have fitted the gasket first then put sealant under it but I didn't want to risk buckling the gasket by pulling it up too high. So I fitted the rubber seals to the crankcase with sealant both sides, then put sealant on the heads round the water passages. I laid the gasket flat across the space between the heads and this allowed me to loosely insert a couple of the bolts on one side just to hold that side of the gasket in position. Then I carefully bent the gasket into position keeping it away from the sealant on the other head, until I could insert a couple of bolts that side too. This also ensures that when you have the inlet manifold in position you know the holes in the gasket are in the right places. Then I fitted the curved plates that clamp the gasket to the rubber seals and crankcase in the centre of the valley (most of the way but not tightened), which allowed me to remove the bolts that had been holding the gasket in position. More sealant on top of the gasket round the water passages and lower the inlet manifold into position - only to realise in the nick of time I had forgotten to refit the heater return pipe (repainted satin black) underneath, fortunately only a moments work. A little wiggling and I get all the bolts in and torque them down. Install the Otter switch using Hermetite red and fit the P-clip for the vacuum pipe.

Carbs adapter and air-box

I removed these as a single unit. Removing the adapter for repainting (satin black) from the carbs while the carbs were still attached to the air box avoids disturbing the linkages, even though these are simpler than on the 4-cylinder car. My adapter has an additional bolt on top near the fork of the Vee. I started unscrewing this but realised it had been stuck into the hole and there wasn't a thread, the bolt was a UNF in any case when it should have been UNC if the hole had been threaded. The hole is almost (but not quite) above what appears to be a casting passage between the two throats of the adapter, possibly as a balance pipe, but it is extremely narrow in the order of a couple of mm. Other people have this passage but not the hole and bolt on top, maybe this was a POs attempt to fit a vacuum gauge. Thought about stripping and repainting the air-box which is currently in what looks like Hammerite silver, but by now the parts had arrived and I wanted to get on with rebuilding, I can do the airbox anytime and it isn't that bad anyway. On refitting I used a little clear sealant around each port on top of the inlet manifold (no gasket) then fit the adapter and carbs, leaving the airbox off until I had balanced the carb airflow. Sealed the mystery bolt back into its hole and refitted the choke and accelerator cables and the fuel pipe. Leave the carb overflow pipes off for the time being so if one leaks I can easily see which. Set the mixture to the starting point of the top of the jet being flush with the bridge then two full rotations clockwise (being HIFs) ready for a full setup later on.

Exhaust manifolds

Although the book says you can leave them in-situ and move them to one side or the other to remove the head bolts it is a false economy. Remove the down pipes (slackening the rear mounting right off makes this much easier) and the manifolds altogether and get them right out of the way, but to get the right-hand one off you have to pull the steering rack. The manifolds have always been a pain on this car - they are tubular and warp in use which means when you take them off you can't get them back on without filing out the holes, but then the flanges are at different angles and distances from the head reducing still further the likelihood of a good seal. I noticed that the ports in the manifold were much bigger than the ports in the head, meaning that although the centres might be off-line you could still avoid any overlap which would tend to choke the port. I welded bars in between each flange to push them back to where they should have been and stop them warping any more, put wooden blocks in the head ports and could fit the manifold over these blocks showing there was no mis-match in the ports. I left them with the same person who 'did' the inlet manifold to go on a belt sander and flat all the flanges, he did but one flange on each was still not aligned with the rest. Shan't be going there again. It was about 15 thou out which was too much to ignore but not enough for an extra gasket, the thinnest of which was about double that. So I took an old gasket constructed with metal facings both sides and a compressible core and split it into two halves. This gave me just about the correct extra thickness for the one flange on each manifold. The usual struggle to get the down pipes on the manifolds and engaged with the Y-piece on the main exhaust. The gaskets were of a type I hadn't used before being one gasket per port with black composition both sides and an inner metal core. As you will read below although these sealed initially they were blowing after a short motorway run. Removing them showed that although the ports in the head were not mis-matched with those in the manifold, because the centres were not aligned some of the gaskets were only clamped over a very narrow area and this had allowed the composition to be blown out. Fortunately I had two pairs of double gaskets left over from a previous go at sealing the manifolds and these have the metal facings with an inner compressible core and are much stronger. Even so the manifold is *still* blowing slightly, and I think this is due to the very narrow overlaps on some ports rather than the faces not being flat or flush. I'm thinking now that the only thing left is to remove it again, build up the flange on the *inside* of the port with weld to thicken the overlap, making sure it doesn't go so far as to start restricting the flow.

Final refitting

Loosely refit the radiator - just in case I have to take it out again. I used new hoses all round, all except the bottom hose having been on the car since I bought it eight years and 65k miles ago, being careful to position the clamps so I could remove or tighten them without any dismantling. The clamp on the rear of the heater return pipe under the inlet manifold is the trickiest, but it is possible to angle it so as to reach it with a set of 3/8" extension bars. Using clamps with hex heads instead of screwdriver slots - even cross-heads - makes life much easier. Fill up with plain water and leave it overnight for leaks. No leaks so I fit a tyre pump adapter to the expansion tank hose and pump the system up to the cap pressure of 15psi. The only immediate leak is where the bottom hose goes on the pump, which stops on retightening. Three days later (banned from the garage for family visits) no more leaks and plenty of pressure left in the system. Fit the alternator and fan belt - a new one as I noticed a crack on the inner face when removing the old one. The alt has always had a spacer washer fitted to each of the head adapter castings, and this seemed to me to move its pulley forward of the pump pulley, and the belt often emitted a rhythmic squeak. Careful measurements using a straight edge showed this to be the case and removing the washers seemed to put them right in line. However I then discovered that the adjustment bracket fouled the clamp on the new heater return hose where it joins the pump, and had to put spacer washers both here and on the alternator lug. Fit the distributor cap, and notice that the distributor must previously have been incorrectly installed as all the plug leads are now one position out, so correct them. Spin the engine on the starter, plugs out, till I get oil pressure. Install plugs, fit HT leads.

Nothing for it now but to start up. Turn on the ignition, pull the choke on by hand (air-box removed) and use a jumper lead to apply 12v to the starter relay winding, whereupon it starts almost straight away. Two things are immediately apparent - no tappet clatter but it sounds rough. I had been warned that fitting new tappets, even if they had been soaked in oil for two or three days beforehand, could result in them all clattering with a terrible din for up to 20 mins. As mine didn't I can only assume that putting a drill on the oil pump until oil was visibly pumping out from every tappet and then leaving it for a few days did a better job of priming them than leaving them in a bath of oil off the engine. As for the roughness I rechecked the plug leads on the distributor cap and found I had a pair of them reversed. Restarted the engine and it sounds much smoother, but keeps cutting out then won't start at all. Then I realise that I had removed my fuel pump fuse as a precaution and forgot to replace it. Refit that, wait for the pump to stop chattering, fire it up and run it at 2k or so while the cam beds in, then leave it at a fast idle with no choke while looking over, under and round for oil and water leaks - thankfully none. Some smoke off the manifolds and down-pipes but that is to be expected. Check and adjust the carb airflow balance, both on and off idle, and the timing. Timing needs a little adjustment but the pointer is rock-steady whereas it used to move about a bit before - obviously the effect of the new timing chain and gears. Then switch off and refit the airbox, filters and carb overflow pipes, then run again and adjust the mixtures to the lifting pins. All seems well, engine gets up to temp and the fans cut in, the temp drops, and the fans cut out again. Fix the rad properly, fit the fan grill, reinstall the steering rack and anti-roll bar, refit the road wheels, remove the axle stands, and take it for a short test-drive and all seems well. Exhaust manifolds not blowing for the first time in years. Next day take it for a longer run down the motorway and back through the lanes, when I get back the left-hand manifold is blowing again (see above)! After a couple of days I drain out the plain water and refill with a 33% mixture of antifreeze and check for leaks, seems fine so far.

In conclusion the prospect of doing this job seemed pretty daunting but I thought about it for a long time and read the manual through several times making notes about order of actions, torque figures etc. In the event apart from a couple of snapped bolts it went quite well. The pumping-up seems to have been fixed, the pressure gets up to about 11psi just as the fan cuts in and drops to about 6 when it cuts back out. Running in free air on the motorway reduces it to about 3.5psi. Beforehand it would get up to radiator cap pressure fairly rapidly. The coolant level probe showed a little fluttering on the first day but rock-steady thereafter, I am assuming the initial fluttering was the system purging itself of air. Beforehand it would start fluttering within a few miles and get worse and worse until eventually it indicated that the coolant level had dropped below the bottom of the probe. The disappointing thing is that the tappets still rattle when hot. Opinion now is that the bores in the block are worn and will have to be sleeved. Well, for now they will just have to rattle. One interesting difference to before is that when warming up the needle had always oscillated slightly about 'N' before settling down, not to be confused with the wild oscillations that can occur - I have seen the temp gauge at 60psi on the oil gauge! - when the steam pipe is blocked. Now it rises slowly, possibly slower than before, then slows and comes to a stop on 'N', no oscillation at all. This could be as a result of clearing the bypass pipe inside the inlet manifold. As the (4-cylinder) roadster has also always done the same thing I took no notice of the V8 doing it.

Update October 2003. Something continued to clatter, and very badly during the hot weather. I tried adding a viscosity improver but apart from a slight increase in pressure there was no change in the noise. I took the rocker covers off when hot and rattling and inserted a 20 thou feeler gauge under each rocker, one at a time, to see what happened. All rockers seemed equally difficult to push the feeler gauge in and pull it out, and there was no change in the sound when doing so, so I can't really see how it can be tappets. An engine rebuilder opined "It doesn't look or sound like bearings, it could be pistons, I can't guarantee to clear it, and it sounds 'orrible". I left it that I would take my spare short engine over to him some time in the future for him to give me a price on rebuilding that, and left it at that. Come the V8 Register Tour of Cornwall, and after 70 miles of M5 on the way down the noise seems to be quieter even when stuck in traffic. Seemed to go back to 'normal' after climbing Porlock and Lynton/Lynmouth, but then over the remainder of the tour it just got quieter and quieter, even on the very steep and narrow 'roads' through some of the coastal villages. After the mainly fast A30 and M5 trip back home it seemed to have stopped altogether, and after a couple of weeks back on mainly local and short journeys it is *still* very quiet, only making the slightest tapping at very low revs after a hot start. It is so much nicer to drive without all that noise, time will tell whether it remains quiet.

Update Summer 2005. Just a faint ticking occasionally now, with a louder tick after a hot start even more infrequently. I've been communicating with Nik Henville this year as his V8 had a similar coolant problem to mine. However in his case he found the bottom hose clip loose. It started dripping as soon as he touched it even though it hadn't been leaking before. There could well be negative pressure in the bottom hose from the action of the pump, and this negative pressure could also suck air in past bad pump seals (which was one of the reasons I changed my pump at the same time). Just tightening the bottom hose cured Nik's problem. Mine wasn't loose so I don't think my problem was that, but a good example of why you should think, think, think and go for the easy things first.

For many years getting consistent results on my 4-cylinder roadster had seemed impossible - I would adjust them using the then rotate the crankshaft again to recheck them and some of them would be incorrect. So I would adjust them again, rotate again, and they would be out again! It took me some years to realise that at the strict 'Rule-of-Nine' (RON) point some of the gaps were still changing. And unless I stopped the crank at the same point for each valve each time they would vary considerably between tests. So instead of using the strict RON point I just used it as a starting point, and looked for the point of greatest gap on each valve. On valves 6 and 8 this proved to be significantly after RON and on valve 7 significantly before. What this says about my cam and cam bearings I am not sure, but the tappets are noticeably quieter, and 5000 miles on the gaps are still the same indicating that I am not getting recession or burning.