Midget Bits

Head Gasket    Water Pump   

Head Gasket

A mate has a Midget 1500 with a recently blown head gasket. Now he has the (barn find) MGB GT he's looking to sell the Midget, and asked me if I'd change the head gasket for him.

Looking though his Haynes it looked pretty straightforward, except that I was a bit concerned that the water pump seemed to be attached to both the front of the head and the front of the crankcase. The instructions said just to remove the bolts securing it to the head, but I didn't see how that would let me get the head off, let alone back on with a new water pump gasket, even worse seal across the two surfaces (head and block). In the event the pump is only attached to the head, so it wasn't a problem.

First thing was drain whatever coolant was left by removing the bottom hose where it attaches to the radiator. The clamp on that was well rusted and wasn't going to shift easily, "A good start" I thought to myself. But the other end where it attaches to the water pump came off easily, and was good enough, as I only needed to lower the level in the engine to below the top of the block and not empty it or the radiator. I also started draining the sump, as the oil looked to be badly contaminated with water, all of it looking like mayonnaise. Being cold as well it flowed very slowly, so I left it draining into a bucket whilst I carried on elsewhere.

Next was carbs off. Someone had said to keep the air cleaner box on, which keeps the linkages between the carbs intact. But like that I couldn't see how to get at the bottom carb bolts. So I removed the fuel delivery link pipe between the carbs, the air cleaner, slackened the top bolts a little, removed the bottom bolts, refitted the air cleaner, then removed the top bolts. I'd already removed the accelerator cable bracket from the inlet manifold, so with the choke return spring disconnected from the front carb (left hanging on the exhaust manifold), and with the accelerator and choke cables and linkages still attached, lifted the carbs over to lie across the brake and clutch master cylinders which was neatly out of the way. Next was removing the hoses from the manifold - one at the front and two at the back. I also undid the large nut securing the coolant bypass pipe into the lower part of the water pump, as this pipe has a bracket onto the rear manifold stud and needs to be movable to get it over the stud. Also the cooling fan cover plate so I could get at the water pump to head bolts. Haynes says to remove the heater air pipe, I've no idea why, it isn't remotely in the way - or so I thought.

I didn't fancy trying to remove the three nuts holding the exhaust down-pipe to the manifold, and was hoping to be able to swing the manifold across clear of the head to enable the head to be lifted off. Fortunately the manifold only goes over the outer two studs, which are quite short, fitting between the remainder of the studs which are used with bridging pieces to attach both inlet and exhaust manifolds. It was a bit of a fiddle removing the two lower nuts. I've found in the past that having one spanner of each size (or even one spanner and one socket per size) is no good for many awkward places, and I have about four different types of 1/2" and 9/16" (the most common MGB and Midget sizes, it seems) spanner - flat open ended; flat open/angled ring combination; double-cranked ring; and flat with open ended one end and the same size as a ratchet ring (very convenient for moving nuts and bolts along the threads or where there is only a very small 'swing' space) on the other end; also 1/2" and 3/8" drive socket sets. In this case the only one that would get on the nuts was the 3/8" socket, and then only with a wobble extension. They weren't very tight, so I suppose the previous fitter had only been able to get a flat one on at an angle and hence not much pressure. The top ones are easy, and the inlet manifold was off. One of the two outer exhaust studs came out attached to the nut, which gave me the idea of removing the other one with the double-nut method, which meant the manifold came free without having to be pulled off the studs. In the end not much of an advantage, as we were fitting a different head (in case the old one had warped) and I couldn't get these studs out of that one. Manifold and exhaust moved out of the way quite easily, however I noticed that the face of the manifold was way off being aligned with the head when free, in both vertical and horizontal planes. I wondered how easy it was going to be to get a good seal later on, there being signs of blowing round the bottom of the front port.

Next the rocker shaft came off, making sure the push-rods were unstuck from the rockers, or you can lift the tappets out of their bores, then they drop back across the bores, and you have the job of trying to get them back in again. Now we were down to the bare head, so slackened the nuts in the correct order bit by bit (even though we were going to fit a different head). Before I removed the nuts completely I undid them to be just clear of their washer, then turned the engine over (plugs still in) which should unstick the head from the block. In the event I don't think it was stuck at all, as the old gasket didn't leave any material stuck to head or block. That done I was able to lift the old head off the studs single-handed, and remove the old gasket.

Immediately obvious was the fire-rings blown inwards on the two outer cylinders, in the same (mirror-image) place - the outer corners on the manifold side. These coincided with two curious blind holes in the block, right beside the groove machined for the fire-rings, about 1/2" across, 3/4" deep, with radiused bottoms. I've no idea what they are there for, and they seem a liability, as any pressure seeping in to the holes will come back out again and tend to push the fire-ring reinforcement off the edge of the gasket. I was also concerned about the depth of machining of the groove round the cylinder. This is presumably to accommodate the fire-ring reinforcement in the gasket, but I thought the whole idea of this thicker section was so that greater pressure is applied to the part of the gasket that is going to experience the most extreme conditions. But all that aside, the reason for changing the gasket in the first place was that it had lost its coolant and overheated, with steam coming out of the exhaust and the sump full of water. But I examined the gasket closely and couldn't see any evidence of damage round the waterways. About this time I glanced under the car to see how the oil draining was going on. It had overflowed the bucket, and what a mess that made! Fortunately less than a cup-full, but it doesn't half spread a long way. The worst job was trying to handle the overflowed bucket, tipping it into an empty oil container via a purpose-made 'funnel' - a 1 litre gear oil bottle with the base and the narrow part of the tube cut off, the thick part of the tube being a snug fit into a Halfords or Castrol 5 litre oil 'can'. It was so thick it was like liquid mud, much of it ran down the side of the bucket instead of pouring into the funnel, that which did ran very slowly into the oil can with air bubbles coming up like in a volcanic mud bath. What a mess!

I'd previously cleaned up the (good used) replacement head, removing the valves and grinding them in a bit, and checking they were 'petrol tight'. Because the old gasket hadn't stuck to either head or block the block face needed very little scraping. The new gasket is a Payen, with a matt black surface which seems to have been coated with some kind of clear 'varnish'. It reminded me of my Scimitar GTE (Ford 3 litre V6 'Essex' engine), the gasket for which is 'composition' and needs a 3-stage process for fitting - 1st scrupulously clean/degrease head and block before fitting, torque down and run up to warm with no coolant. 2nd leave to cool right down (i.e. overnight), retorque, fill with water but don't fit the cap and run it up to temperature. 3rd leave it to cool right down again, retorque, top up coolant if required and away you go. Someone on the Midget BBS said much the same thing, so I decided I would use the same process. One thing that bothered me a bit is that a couple of people on the BBS said theirs had blown again only a few hundred miles after previous replacement, which was what had happened to this car, which makes me a bit nervous as to whether this replacement will last any longer. End of day one, and quite pleased with progress.

Day two: With the gasket fitted and torqued down (50 ft lb in stages and in the specified order!) I set about reattaching carbs and exhaust etc. What I should have done was check the oilway in the head before fitting, bit didn't think about it at the time, so tried it with the head fitted but without the rocker gear. Spinning the engine over (no plugs) the first thing I noticed was petrol spurting out of the carb feed pipe - it being a mechanical pump of course. Stuck my thumb over the end, which when I released it after stopping cranking spurted some more out of course. After doing this 2 or 3 times I noticed that I wasn't getting the spurt anymore, but didn't think too much about it. What I was also looking for was oil coming up through the head port, but even with enough cranking to get full oil pressure on the gauge none appeared - a bit worrying, but put it on one side for the moment. Had to pull on the manifold and exhaust quite a bit to get them over the two outer studs (not being able to get these out of the head even when attached to the block), and used the two outer studs to pull that more or less into position before fitting the inlet manifold. Before that I had fitted the new manifold to head gasket over the studs, with a smear of exhaust assembly paste around the exhaust ports, and a smear of Hermetite Red round the inlets. Started fitting the bridging pieces across the exhaust and inlet manifold flanges, before I noticed they were of two different sizes. The four (?) smaller ones go at the top and the two larger ones at the bottom. Snugged all the nuts up bit by bit to pull the exhaust manifold gradually into position.

Next to re-attach the carbs and heat-shield. They had previously been fitted with some kind of silicone sealant, which I hate, as it tends to skin if not dry far too quickly where you have a complicated or long edge to seal. These had been installed with great wodges of the stuff, which had oozed into the inlets and must have reduced them by at least 10%. More Hermetite Red on the face of the inlet manifold and carb flanges, and on both sides of the heat shield. This was now the tricky bit - getting the top two bolts though the carb flanges, the gaskets and heat shield onto those, then getting the bolts started in the inlet manifold, and the gaskets to the correct angle to get the bottom bolts in. Just managed that and removed the air cleaner, when I realised I had got the inner gaskets on (between heat shield and manifold) but forgotten the outers. In the fiddle to get them off again and correct that, the choke linkage fell out, so I thought "What the heck" and removed the throttle linkage as well which actually made things easier to get the carbs and gaskets installed with just the top bolts. Getting the linkages back in was relatively easy - with just the top bolts in you can swing the carbs apart and slot them in, and it is pretty obvious how it all goes together.

Put the air cleaner and fuel delivery pipes back on, refitted the rocker gear (30 ft lb) and set the valve clearances by putting it into 4th and nudging the car forwards. They were all tight by about 1/8th turn, so presumably this head has been skimmed more than the old one. Nothing for it now but to start it up for its first 'dry' run. Cranked, but no start. Took the coil lead out of the distributor, attached it to a spark plug lying on the block and that sparked OK, so check the fuel delivery. With the pipe off the back carb and cranking just a few spots of fuel and that was all, then nothing. Keith thought there was enough in the tank, but we poured in another gallon anyway, which made no difference. I can't believe that the fuel pump chose to pack up during the head gasket change, I had been nowhere near it, but that was how it seemed. Keen to get the engine running, we attached a funnel and pipe to the carbs, poured in some petrol, cranked it, and it fired up straight away, so we kept feeding it with petrol till the block warmed up. It was running very roughly, but the breather hose wasn't attached to anything and so making the mixture weak, so I shoved the threaded end of a spark plug in the end of the hose which solved that. By now it was lunchtime, so a convenient point to let the engine cool while we investigated the pump, hopefully being able to run it again with water in but no cap before the end of the day.

After lunch took the rocker gear off, cranked some more but still no oil coming up through the head. Removed the blanking plug on the back of the head, cranked again, and this time some oil did come out of that hole. So if there was a blockage I should be able to clear it by pocking in through that hole, as well as down from the top. But first I put my thumb over the hole while Keith cranked again to try once more, and this time oil did come up out of the head - phew! Must have been a case of not giving it enough time to work its way through, but the oiling rate seems much less than I am used to on my MGB, little more than tiny splashes.

With that out of the way it was time to investigate the pump. With the top off it was dry, so there seemed nothing else for it but to remove the pump and try and find out what was wrong, and now the heater air pipe did have to come off! As did the solenoid to starter cable, and the coil to distributor HT lead, to give me as much room as possible. 3/8" socket drive and wobble extension comes to the fore again, as the rear bolt is partly obstructed by the firewall for the extensions I have. With the pump off we put a pipe and funnel on what we thought was the inlet, poured in some petrol, and I operated the cam lever. When I released it a jet of petrol shot up into the air, curled over, and straight down the back of my neck! We looked at each other and Keith said "There doesn't seem to be much wrong with that!", and when we got the funnel on the right port indeed fuel was being pumped through. I went back to car and stared at the pipes looking for some inspiration. Idly took hold of the supply pipe and wiggled it, and was surprised how easily it moved, it didn't feel like it was attached to anything! Got the car back up onto axle stands (which in itself is a bit of a fiddle as it has to be 'walked' up in two or three stages to get it high enough), slid underneath, to find a short section of rubber hose that joins the long main pipe from the tank to a separate short pipe that leads up the firewall, was completely detached from the short pipe. There was a short lump of rubber and clip on the end of the short pipe, so the hose had simply perished, split which probably caused air to be sucked in, then when I was fiddling with the pump probably detached completely. Of course it was the upper end that needed to be reattached, getting the stub off first, and that was wedged up between gearbox and transmission tunnel, and only room for one hand. Fortunately the clip was the spring type and not a Jubilee, and both clip and stub came off easily. While Keith held the short pipe steady from above I was able to push the hose on from below and refasten the clip. Time to replace the fuel pump, cut a new gasket out of some card, scraped and cleaned as best I could, and reassembled with Hermetite. Cranked the engine over and sure enough fuel was spurting out the delivery pipe at the carb.

By that time the engine had fully cooled, so gave it the first retorque, by applying 'undo' pressure just enough to get each nut moving, then tightening back up to 50 ft lb. I noticed that each nut tightened more than I had slackened, which proves the point as to why retorquing is necessary. Replaced the rocker gear. Didn't bother resetting the valve clearances as they would have to be done again after the 2nd retorque. Started filling up with water poured in through the plug in the thermostat housing. Quite a slow process as it couldn't take much flow rate. I was using a funnel but it had ribs on the outside of the spout to (normally) let air out while pouring liquid in. But if I poured too fast the level in the thermostat housing rose and the excess came out past these ribs, and without the rocker cover and gasket fitted this would run into the sump through the push-rod holes. I stuffed some paper into the gap both to divert and absorb any overflow, but still erred on the side of caution and a slow fill. I could hear trickling so stopped and looked around underneath but nothing was running down, it was coming from inside the engine/radiator so OK. Kept pouring slowly, then became aware of my leg getting damp, and this time it was running down, but from the overflow of the expansion tank as I had the filler cap removed :o) Replaced that, as what I was pouring in the top was running out the bottom, and that allowed me to fill the thermostat housing, whereupon I replaced the plug.

Ready for its 2nd unpressurised run, so I removed the radiator cap again, cranked her over, and she started up very easily. Watching the rockers one by one oil came up through the little bleed hole on the top of each one and started splashing around, so I just dropped the rocker cover over the studs with no gasket to stop it getting everywhere. Went to wipe some drops off the alternator casing with my hand, and shoved my thumb right into the alternator fan! Hoses and radiator started warming up, and with the heat expansion water started running out of the expansion tank overflow, but as it was completely full that was only to be expected. Checked round the various water, petrol and oil connections we had had apart and everything was fine. Eventually water stopped running out of the overflow, so presumably everything had reached a stable temperature, although the hoses and radiator weren't as warm as I was expecting. End of day two, and a convenient time to stop and leave the engine to cool overnight.

Day three: Get the rocker gear off again, retorque (again I noticed they tightened by more than I had slackened them), replace the rocker gear and reset the gaps. Fitted the rocker cover with gasket, with Hermetite Red just on the cover face. Although the cover is grooved where the gasket goes it is a different shape, and even with Hermetite the gasket didn't stick, which meant care was needed when the cover was on the studs and sitting on the gasket that it was properly located before tightening down the cover, and reattached the breather hose to the rocker cover. Fired her up, while she was warming up we cleared up all the tools, checked all round for any petrol, water or oil leaks, then it was time for a test drive. I wondered whether I ought to drive behind in my car but Keith was confident so we just went for it. Ran like a charm, my first time in a Midget and the 1500 at least is pretty sprightly. After about 20 mins or so we got back home, now the radiator was at the sort of temperature I was expecting. Checked around again and still no leaks, so, job done! Fingers crossed it stays that way.

A couple of days later and on the way to an MOT (annual test) it starts running really badly, and Keith limps it home. When I get there and start it up it runs fine for a few moments, then obviously starts running on 3 cylinders. Clipped the timing light onto the coil lead and each plug lead but everything is flashing as it should be. Quite by chance I pulled No.3 HT lead off the distributor cap and it made no difference - i.e. that was the faulty cylinder. Not believing the chances of that I replaced it and pulled the others off one-by-one, but no it was definitely No.3. Held near the cap No.3 lead sparked just as all the others did, so that and the timing light flashing told me the rotor and cap were OK. Removed the plug and it was fouled. Replaced it (with a used plug) and it fired up OK, but after a few seconds started firing on 3 again. Tried another used plug and that was the same. By now I'm beginning to fear the worst and suspect a sticking valve. There is no popping in the intake so it must be an exhaust, but for that to be affecting running to the extent that removing the HT lead made no difference at all, it has to be sticking for a fair way through the induction stroke, at least. But looking at the valve spring it seems to be doing exactly the same as all the others, and the rocker is showing no signs of excessive looseness. It isn't tight either, for although the gap has closed up somewhat because the engine is hot (they were set to 10 thou cold) there still is a gap, and is much the same as the other valves. Then I have a brain-wave and realise I can use the adjustable timing light to 'freeze' the valve (low ambient light levels), by connecting the light to a suitable plug lead and varying the timing control. Indeed this is exactly what happens, and by varying the control I can freeze the valve anywhere between fully up and fully down. No.3 is regular and consistent, and the timing the same as the other exhaust valves, so I'm beginning to doubt my diagnosis. As a further test I press down on the top of the spring cap with a screwdriver, and lever up under the upper turns of the spring, thinking that if it is a sticking valve one or other of these tests should have some effect, but neither did. Then all of a sudden it started firing on 4 cylinders again. Thinking that possibly this was because the valve cover had been off a while, and the top of the valves might be a bit cooler and it had stopped sticking, we replaced the cover. The engine continued to run perfectly for several minutes even being revved hard, on a short test-drive, and a drive of about 20 miles or so next day. So at the moment it remains a mystery. One thing that did occur to me subsequently was that with the plug being fouled, wouldn't that mean it wasn't firing? I.e. no HT reaching the plug (we'd tried three plugs remember) despite it leaving the cap and going into the lead? Also another test I could have done while it was running on 3 was listen to the misfire in the exhaust while Keith alternately removed and replaced the plug lead for that cylinder. If it was firing, but the exhaust valve was sticking open, the misfire would surely make a very different sound with the plug connected to when it wasn't. Whereas if the problem was simply that the plug wasn't firing, there would be no change in the sound between connected and disconnected.

Water Pump
Shortly after changing the head gasket Keith reported that the water pump seemed to be leaking from one of the bolt holes, tightening it made no difference. My first thought was that it must be one of the bolts the attaches the pump housing to the head, which surprised me as I had used a new gasket with Hermetite Red round the bolt holes as well as well as round the main hole. But when he brought it round, and it was the bottom bolt that secures the pump to the housing (which I hadn't touched), and stopped dripping as soon as he switched off the engine, I realised it was the pump seal that had gone and it was actually dripping out of the hole on the bottom of the pump body. It probably is collateral damage, though, as in the head change I removed the fan-belt and retensioned it afterwards. I don't think I overtightened it, so maybe the pump was on its way out anyway and disturbing it just tipped it over the edge.

Haynes wants you to remove the housing and radiator, which seems an awful lot of bother. If you remove the pump with its blades and housing I can't see why you need to remove anything else but the top fan shroud, as it should lift up vertically. But I didn't want to do that as I have only just refitted it as part of the head gasket change.

I did look at removing the radiator as there is very little space between it and the front of the blades, but when I saw there were two bolts going into the bottom of the radiator I shied away from that fearing I'd turn the bottom out of the rad. I also looked at pulling the shroud up to allow me to pull the radiator a bit further away from the blades and so give me a bit more room, but even when I had removed the two bolts half way down (the captive nut for one of those started turning in its cage!) there seemed to be two more further down that I couldn't see how to get at even with the grill removed, and Haynes didn't mention those. In fact it seemed to me that as the shroud sits under the bonnet slam-panel and over the radiator, and that panel is screwed to the body frame rather than being welded as on the MGB, that in fact the slam-panel has to come off before you can lift the shroud and radiator out!

In the end with the two top bolts and two screws securing the top of the shroud to the bonnet slam panel gave me a bit of wriggle-room. I thought if I could get the pump and blades clear of the housing I might be able to get it across to the left-hand side of the engine bay, and with the top hose and alternator out of the way would have enough room to lift it up from there. The top hose came off easily enough, but I found the long bolt that goes all the way through the bottom mount of the alternator hit the radiator before the alternator and spacers would come off the engine front plate. Should this bolt be inserted from the rear? Does the front spacer have a shoulder that goes into the front plate? It seems to. Eventually I found the alternator swivelled right over and dropped down by the chassis rail, which did get it out of the way without complete removal. However the big question would be, if the alternator does have to be removed at some time, how am I going to do it short of cutting through the long bolt and getting a new one to fit from the other side?

I removed the nuts securing the pump to the housing with the blades in-situ, which with the radiator loose just allowed the pump body to clear the studs, but then I found I needed even more room for the impeller to clear the housing, and that just wasn't available.

Eventually I undid the four bolts securing the fan to the pump, the looseness of the radiator and shroud giving me just enough room to remove them and the fan. With the pump nuts loose and the bottom stud out (it came out with the nut but was captive in the pump body) I could tilt the pump which improved access to the bolts. But I feared problems when replacing as I wanted to bolt the new pump up to the housing with a new gasket and Hermetite Red, so I wouldn't be able to tilt it as I had done for removal to refit the fan.

But marking the pump nose with a pencil to show where the four bolt holes are (they aren't equally spaced) and Keith holding the fan in position I was able to slip each bolt into the blades in turn and get them started quite easily. A point filed onto the end of each bolt (and thread chased clean) would have made it even easier. So false starts excluded it was altogether a much easier job than Haynes indicates.